sandey/blender
1
0
Fork 0
forked from bartvdbraak/blender
Code Pull Requests Activity
226d6fbfc5
blender/intern/cycles/render/session.cpp

909 lines
22 KiB
C++
Raw Normal View History

Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/*
Cycles: relicense GNU GPL source code to Apache version 2.0. More information in this post: http://code.blender.org/ Thanks to all contributes for giving their permission!
2013-08-18 14:16:15 +00:00
* Copyright 2011-2013 Blender Foundation
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
*
Cycles: relicense GNU GPL source code to Apache version 2.0. More information in this post: http://code.blender.org/ Thanks to all contributes for giving their permission!
2013-08-18 14:16:15 +00:00
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
*
Cycles: relicense GNU GPL source code to Apache version 2.0. More information in this post: http://code.blender.org/ Thanks to all contributes for giving their permission!
2013-08-18 14:16:15 +00:00
* http://www.apache.org/licenses/LICENSE-2.0
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
*
Cycles: relicense GNU GPL source code to Apache version 2.0. More information in this post: http://code.blender.org/ Thanks to all contributes for giving their permission!
2013-08-18 14:16:15 +00:00
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
*/
#include <string.h>
#include <limits.h>
#include "buffers.h"
#include "camera.h"
#include "device.h"
Cycles: experimental correlated multi-jittered sampling pattern that can be used instead of sobol. So far one doesn't seem to be consistently better or worse than the other for the same number of samples but more testing is needed. The random number generator itself is slower than sobol for most number of samples, except 16, 64, 256, .. because they can be computed faster. This can probably be optimized, but we can do that when/if this actually turns out to be useful. Paper this implementation is based on: http://graphics.pixar.com/library/MultiJitteredSampling/ Also includes some refactoring of RNG code, fixing a Sobol correlation issue with the first BSDF and < 16 samples, skipping some unneeded RNG calls and using a simpler unit square to unit disk function.
2013-06-07 16:06:22 +00:00
#include "integrator.h"
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
#include "scene.h"
#include "session.h"
#include "util_foreach.h"
#include "util_function.h"
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
#include "util_math.h"
Cycles: border render now works in the viewport, when looking through the camera, same as in render. It draws objects in solid draw mode outside of the border.
2012-06-06 23:27:43 +00:00
#include "util_opengl.h"
Cycles: merging features from tomato branch. === BVH build time optimizations === * BVH building was multithreaded. Not all building is multithreaded, packing and the initial bounding/splitting is still single threaded, but recursive splitting is, which was the main bottleneck. * Object splitting now uses binning rather than sorting of all elements, using code from the Embree raytracer from Intel. http://software.intel.com/en-us/articles/embree-photo-realistic-ray-tracing-kernels/ * Other small changes to avoid allocations, pack memory more tightly, avoid some unnecessary operations, ... These optimizations do not work yet when Spatial Splits are enabled, for that more work is needed. There's also other optimizations still needed, in particular for the case of many low poly objects, the packing step and node memory allocation. BVH raytracing time should remain about the same, but BVH build time should be significantly reduced, test here show speedup of about 5x to 10x on a dual core and 5x to 25x on an 8-core machine, depending on the scene. === Threads === Centralized task scheduler for multithreading, which is basically the CPU device threading code wrapped into something reusable. Basic idea is that there is a single TaskScheduler that keeps a pool of threads, one for each core. Other places in the code can then create a TaskPool that they can drop Tasks in to be executed by the scheduler, and wait for them to complete or cancel them early. === Normal ==== Added a Normal output to the texture coordinate node. This currently gives the object space normal, which is the same under object animation. In the future this might become a "generated" normal so it's also stable for deforming objects, but for now it's already useful for non-deforming objects. === Render Layers === Per render layer Samples control, leaving it to 0 will use the common scene setting. Environment pass will now render environment even if film is set to transparent. Exclude Layers" added. Scene layers (all object that influence the render, directly or indirectly) are shared between all render layers. However sometimes it's useful to leave out some object influence for a particular render layer. That's what this option allows you to do. === Filter Glossy === When using a value higher than 0.0, this will blur glossy reflections after blurry bounces, to reduce noise at the cost of accuracy. 1.0 is a good starting value to tweak. Some light paths have a low probability of being found while contributing much light to the pixel. As a result these light paths will be found in some pixels and not in others, causing fireflies. An example of such a difficult path might be a small light that is causing a small specular highlight on a sharp glossy material, which we are seeing through a rough glossy material. With path tracing it is difficult to find the specular highlight, but if we increase the roughness on the material the highlight gets bigger and softer, and so easier to find. Often this blurring will be hardly noticeable, because we are seeing it through a blurry material anyway, but there are also cases where this will lead to a loss of detail in lighting.
2012-04-28 08:53:59 +00:00
#include "util_task.h"
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
#include "util_time.h"
CCL_NAMESPACE_BEGIN
Fix #32951: Progressive refine crashing Blender when used with multiple cuda Issue was caused by offline rendering could have been allocated the same tile to different devices and in this case buffers would become invalid. Made it more clear in the code, so now it's flag for tile manager to indicate whether tiles should always be allocated for the same device or not. Also cleaned a way how tile index for progressive refine is calculating, which is now avoids tricky computation based on tile coordinate and it's dimensions.
2012-10-23 16:36:53 +00:00
/* Note about preserve_tile_device option for tile manager:
* progressive refine and viewport rendering does requires tiles to
* always be allocated for the same device
*/
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Session::Session(const SessionParams& params_)
: params(params_),
Cycles: change preview "resolution divider" that gave the number of lower resolutions to render, to a "start resolution" which gives the resolution to start at. This avoids unnecessary rendering of small resolutions in small viewports, and avoids long waiting on big viewports.
2012-09-17 10:55:18 +00:00
tile_manager(params.progressive, params.samples, params.tile_size, params.start_resolution,
Cycles / Tile Rendering: * Added new option to chose the tile order. In addition to the "Center" method, 4 new methods are available now, like Top -> Bottom and Right -> Left. Thanks to Sergey for code review and some tweaks!
2013-01-07 19:55:49 +00:00
params.background == false || params.progressive_refine, params.background, params.tile_order,
Cycles: memory usage report This commit adds memory usage information while rendering. It reports memory used by device, meaning: - For CPU it'll report real memory consumption - For GPU rendering it'll report GPU memory consumption, but it'll also mean the same memory is used from host side. This information displays information about memory requested by Cycles, not memory really allocated on a device. Real memory usage might be higher because of memory fragmentation or optimistic memory allocator. There's really nothing we can do against this. Also in contrast with blender internal's render cycles memory usage does not include memory used by scene, only memory needed by cycles itself will be displayed. So don't freak out if memory usage reported by cycles would be much lower than blender internal's. This commit also adds RenderEngine.update_memory_stats callback which is used to tell memory consumption from external engine to blender. This information is used to generate information line after rendering is finished.
2012-11-05 08:04:57 +00:00
max(params.device.multi_devices.size(), 1)),
stats()
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Cycles: device code refactoring, no functional changes.
2012-01-04 18:06:32 +00:00
device_use_gl = ((params.device.type != DEVICE_CPU) && !params.background);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: merging features from tomato branch. === BVH build time optimizations === * BVH building was multithreaded. Not all building is multithreaded, packing and the initial bounding/splitting is still single threaded, but recursive splitting is, which was the main bottleneck. * Object splitting now uses binning rather than sorting of all elements, using code from the Embree raytracer from Intel. http://software.intel.com/en-us/articles/embree-photo-realistic-ray-tracing-kernels/ * Other small changes to avoid allocations, pack memory more tightly, avoid some unnecessary operations, ... These optimizations do not work yet when Spatial Splits are enabled, for that more work is needed. There's also other optimizations still needed, in particular for the case of many low poly objects, the packing step and node memory allocation. BVH raytracing time should remain about the same, but BVH build time should be significantly reduced, test here show speedup of about 5x to 10x on a dual core and 5x to 25x on an 8-core machine, depending on the scene. === Threads === Centralized task scheduler for multithreading, which is basically the CPU device threading code wrapped into something reusable. Basic idea is that there is a single TaskScheduler that keeps a pool of threads, one for each core. Other places in the code can then create a TaskPool that they can drop Tasks in to be executed by the scheduler, and wait for them to complete or cancel them early. === Normal ==== Added a Normal output to the texture coordinate node. This currently gives the object space normal, which is the same under object animation. In the future this might become a "generated" normal so it's also stable for deforming objects, but for now it's already useful for non-deforming objects. === Render Layers === Per render layer Samples control, leaving it to 0 will use the common scene setting. Environment pass will now render environment even if film is set to transparent. Exclude Layers" added. Scene layers (all object that influence the render, directly or indirectly) are shared between all render layers. However sometimes it's useful to leave out some object influence for a particular render layer. That's what this option allows you to do. === Filter Glossy === When using a value higher than 0.0, this will blur glossy reflections after blurry bounces, to reduce noise at the cost of accuracy. 1.0 is a good starting value to tweak. Some light paths have a low probability of being found while contributing much light to the pixel. As a result these light paths will be found in some pixels and not in others, causing fireflies. An example of such a difficult path might be a small light that is causing a small specular highlight on a sharp glossy material, which we are seeing through a rough glossy material. With path tracing it is difficult to find the specular highlight, but if we increase the roughness on the material the highlight gets bigger and softer, and so easier to find. Often this blurring will be hardly noticeable, because we are seeing it through a blurry material anyway, but there are also cases where this will lead to a loss of detail in lighting.
2012-04-28 08:53:59 +00:00
TaskScheduler::init(params.threads);
Fix #33107: cycles fixed threads 1 was still having two cores do work, because main thread works as well.
2012-11-07 21:00:49 +00:00
device = Device::create(params.device, stats, params.background);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(params.background) {
buffers = NULL;
display = NULL;
}
else {
buffers = new RenderBuffers(device);
Cycles: viewport render now takes scene color management settings into account, except for curves, that's still missing from the OpenColorIO GLSL shader. The pixels are stored in a half float texture, converterd from full float with native GPU instructions and SIMD on the CPU, so it should be pretty quick. Using a GLSL shader is useful for GPU render because it avoids a copy through CPU memory.
2013-08-30 23:49:38 +00:00
display = new DisplayBuffer(device, params.display_buffer_linear);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
session_thread = NULL;
scene = NULL;
start_time = 0.0;
reset_time = 0.0;
preview_time = 0.0;
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
paused_time = 0.0;
Fix a few issues reported by coverity scan.
2013-09-03 22:39:21 +00:00
last_update_time = 0.0;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
delayed_reset.do_reset = false;
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
delayed_reset.samples = 0;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
display_outdated = false;
gpu_draw_ready = false;
gpu_need_tonemap = false;
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
pause = false;
Fix #30246: rendering multiple render layers with CUDA still not working, should really be fixed now.
2012-02-23 19:48:18 +00:00
kernels_loaded = false;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Session::~Session()
{
if(session_thread) {
progress.set_cancel("Exiting");
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
gpu_need_tonemap = false;
gpu_need_tonemap_cond.notify_all();
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
{
thread_scoped_lock pause_lock(pause_mutex);
pause = false;
}
pause_cond.notify_all();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
wait();
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(display && params.output_path != "") {
Cycles: some warning fixes, cpu device task tweaks, avoid unnecessary tonemap in non-viewport render, and some utility functions.
2011-09-08 18:58:07 +00:00
tonemap();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
progress.set_status("Writing Image", params.output_path);
display->write(device, params.output_path);
}
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
foreach(RenderBuffers *buffers, tile_buffers)
delete buffers;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
delete buffers;
delete display;
delete scene;
delete device;
Cycles: merging features from tomato branch. === BVH build time optimizations === * BVH building was multithreaded. Not all building is multithreaded, packing and the initial bounding/splitting is still single threaded, but recursive splitting is, which was the main bottleneck. * Object splitting now uses binning rather than sorting of all elements, using code from the Embree raytracer from Intel. http://software.intel.com/en-us/articles/embree-photo-realistic-ray-tracing-kernels/ * Other small changes to avoid allocations, pack memory more tightly, avoid some unnecessary operations, ... These optimizations do not work yet when Spatial Splits are enabled, for that more work is needed. There's also other optimizations still needed, in particular for the case of many low poly objects, the packing step and node memory allocation. BVH raytracing time should remain about the same, but BVH build time should be significantly reduced, test here show speedup of about 5x to 10x on a dual core and 5x to 25x on an 8-core machine, depending on the scene. === Threads === Centralized task scheduler for multithreading, which is basically the CPU device threading code wrapped into something reusable. Basic idea is that there is a single TaskScheduler that keeps a pool of threads, one for each core. Other places in the code can then create a TaskPool that they can drop Tasks in to be executed by the scheduler, and wait for them to complete or cancel them early. === Normal ==== Added a Normal output to the texture coordinate node. This currently gives the object space normal, which is the same under object animation. In the future this might become a "generated" normal so it's also stable for deforming objects, but for now it's already useful for non-deforming objects. === Render Layers === Per render layer Samples control, leaving it to 0 will use the common scene setting. Environment pass will now render environment even if film is set to transparent. Exclude Layers" added. Scene layers (all object that influence the render, directly or indirectly) are shared between all render layers. However sometimes it's useful to leave out some object influence for a particular render layer. That's what this option allows you to do. === Filter Glossy === When using a value higher than 0.0, this will blur glossy reflections after blurry bounces, to reduce noise at the cost of accuracy. 1.0 is a good starting value to tweak. Some light paths have a low probability of being found while contributing much light to the pixel. As a result these light paths will be found in some pixels and not in others, causing fireflies. An example of such a difficult path might be a small light that is causing a small specular highlight on a sharp glossy material, which we are seeing through a rough glossy material. With path tracing it is difficult to find the specular highlight, but if we increase the roughness on the material the highlight gets bigger and softer, and so easier to find. Often this blurring will be hardly noticeable, because we are seeing it through a blurry material anyway, but there are also cases where this will lead to a loss of detail in lighting.
2012-04-28 08:53:59 +00:00
TaskScheduler::exit();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
void Session::start()
{
session_thread = new thread(function_bind(&Session::run, this));
}
bool Session::ready_to_reset()
{
double dt = time_dt() - reset_time;
if(!display_outdated)
return (dt > params.reset_timeout);
else
return (dt > params.cancel_timeout);
}
/* GPU Session */
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
void Session::reset_gpu(BufferParams& buffer_params, int samples)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Fix #34205: Cycles: Zooming in rendered mode during update out of sync with intended zoom Made sure no pause_cond.wait() happens at the same time as session as resetting.
2013-02-12 13:06:52 +00:00
thread_scoped_lock pause_lock(pause_mutex);
Code cleanup: * Some typo fixes.
2013-08-13 08:43:31 +00:00
/* block for buffer access and reset immediately. we can't do this
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* in the thread, because we need to allocate an OpenGL buffer, and
* that only works in the main thread */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock display_lock(display_mutex);
thread_scoped_lock buffers_lock(buffers_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
display_outdated = true;
reset_time = time_dt();
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
reset_(buffer_params, samples);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
gpu_need_tonemap = false;
gpu_need_tonemap_cond.notify_all();
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
pause_cond.notify_all();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
bool Session::draw_gpu(BufferParams& buffer_params)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
/* block for buffer access */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock display_lock(display_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* first check we already rendered something */
if(gpu_draw_ready) {
/* then verify the buffers have the expected size, so we don't
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* draw previous results in a resized window */
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
if(!buffer_params.modified(display->params)) {
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* for CUDA we need to do tonemapping still, since we can
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* only access GL buffers from the main thread */
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
if(gpu_need_tonemap) {
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock buffers_lock(buffers_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
tonemap();
gpu_need_tonemap = false;
gpu_need_tonemap_cond.notify_all();
}
display->draw(device);
if(display_outdated && (time_dt() - reset_time) > params.text_timeout)
return false;
return true;
}
}
return false;
}
void Session::run_gpu()
{
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
bool tiles_written = false;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
start_time = time_dt();
reset_time = time_dt();
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
paused_time = 0.0;
Cycles: update buffers only once in a while when progressive refine is used It'll mimic the same behavior as regular tile rendering update and it should give pretty noticeable boost on simpler scenes.
2012-10-23 17:24:23 +00:00
last_update_time = time_dt();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
if(!params.background)
Fix #30710: cycles wrong render time after pause/unpause in viewport.
2012-04-04 16:11:44 +00:00
progress.set_start_time(start_time + paused_time);
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
while(!progress.get_cancel()) {
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* advance to next tile */
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
bool no_tiles = !tile_manager.next();
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
if(params.background) {
/* if no work left and in background mode, we can stop immediately */
Cycles: progress printing in background mode, print finished at the end and avoid duplicate prints.
2011-10-30 10:12:34 +00:00
if(no_tiles) {
progress.set_status("Finished");
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
break;
Cycles: progress printing in background mode, print finished at the end and avoid duplicate prints.
2011-10-30 10:12:34 +00:00
}
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
}
else {
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
/* if in interactive mode, and we are either paused or done for now,
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* wait for pause condition notify to wake up again */
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
thread_scoped_lock pause_lock(pause_mutex);
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
Correction for fix #34205: Zooming in rendered mode during update out of sync with intended zoom Didn't initially notice one possibility when GPU render would fail here,
2013-02-13 07:50:41 +00:00
if(!pause && !tile_manager.done()) {
/* reset could have happened after no_tiles was set, before this lock.
* in this case we shall not wait for pause condition
*/
}
else if(pause || no_tiles) {
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
update_status_time(pause, no_tiles);
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
while(1) {
double pause_start = time_dt();
pause_cond.wait(pause_lock);
paused_time += time_dt() - pause_start;
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
if(!params.background)
Fix #30710: cycles wrong render time after pause/unpause in viewport.
2012-04-04 16:11:44 +00:00
progress.set_start_time(start_time + paused_time);
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
update_status_time(pause, no_tiles);
progress.set_update();
if(!pause)
break;
}
}
if(progress.get_cancel())
break;
}
if(!no_tiles) {
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* update scene */
update_scene();
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
if(device->error_message() != "")
progress.set_cancel(device->error_message());
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
if(progress.get_cancel())
break;
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
}
if(!no_tiles) {
/* buffers mutex is locked entirely while rendering each
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* sample, and released/reacquired on each iteration to allow
* reset and draw in between */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock buffers_lock(buffers_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* update status and timing */
update_status_time();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* path trace */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
path_trace();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
device->task_wait();
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(device->error_message() != "")
progress.set_cancel(device->error_message());
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* update status and timing */
update_status_time();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
gpu_need_tonemap = true;
gpu_draw_ready = true;
progress.set_update();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* wait for tonemap */
if(!params.background) {
while(gpu_need_tonemap) {
if(progress.get_cancel())
break;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
gpu_need_tonemap_cond.wait(buffers_lock);
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
if(device->error_message() != "")
progress.set_cancel(device->error_message());
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
tiles_written = update_progressive_refine(progress.get_cancel());
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
if(progress.get_cancel())
break;
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(!tiles_written)
update_progressive_refine(true);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
/* CPU Session */
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
void Session::reset_cpu(BufferParams& buffer_params, int samples)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
thread_scoped_lock reset_lock(delayed_reset.mutex);
Fix #34205: Cycles: Zooming in rendered mode during update out of sync with intended zoom Made sure no pause_cond.wait() happens at the same time as session as resetting.
2013-02-12 13:06:52 +00:00
thread_scoped_lock pause_lock(pause_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
display_outdated = true;
reset_time = time_dt();
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
delayed_reset.params = buffer_params;
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
delayed_reset.samples = samples;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
delayed_reset.do_reset = true;
device->task_cancel();
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
pause_cond.notify_all();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
bool Session::draw_cpu(BufferParams& buffer_params)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock display_lock(display_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* first check we already rendered something */
if(display->draw_ready()) {
/* then verify the buffers have the expected size, so we don't
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* draw previous results in a resized window */
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
if(!buffer_params.modified(display->params)) {
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
display->draw(device);
if(display_outdated && (time_dt() - reset_time) > params.text_timeout)
return false;
return true;
}
}
return false;
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
bool Session::acquire_tile(Device *tile_device, RenderTile& rtile)
{
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(progress.get_cancel()) {
if(params.progressive_refine == false) {
/* for progressive refine current sample should be finished for all tiles */
return false;
}
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock tile_lock(tile_mutex);
/* get next tile from manager */
Tile tile;
Fix #32951: Progressive refine crashing Blender when used with multiple cuda Issue was caused by offline rendering could have been allocated the same tile to different devices and in this case buffers would become invalid. Made it more clear in the code, so now it's flag for tile manager to indicate whether tiles should always be allocated for the same device or not. Also cleaned a way how tile index for progressive refine is calculating, which is now avoids tricky computation based on tile coordinate and it's dimensions.
2012-10-23 16:36:53 +00:00
int device_num = device->device_number(tile_device);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(!tile_manager.next_tile(tile, device_num))
return false;
/* fill render tile */
rtile.x = tile_manager.state.buffer.full_x + tile.x;
rtile.y = tile_manager.state.buffer.full_y + tile.y;
rtile.w = tile.w;
rtile.h = tile.h;
rtile.start_sample = tile_manager.state.sample;
rtile.num_samples = tile_manager.state.num_samples;
Cycles: change preview "resolution divider" that gave the number of lower resolutions to render, to a "start resolution" which gives the resolution to start at. This avoids unnecessary rendering of small resolutions in small viewports, and avoids long waiting on big viewports.
2012-09-17 10:55:18 +00:00
rtile.resolution = tile_manager.state.resolution_divider;
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
tile_lock.unlock();
Highlight currently rendering tiles This commit implements highlight of tiles which are being currently rendered for both Blender Internal and Cycles (and should be possible to use it for other external engines as well). Couple of implementation details: - Added one extra boolean flag to render engine which should be set to truth if render engine wants to highlight tiles. If so, property use_highlight_tiles should be set to True. - Render Part's ready boolena was changed by status enum, which could be NONE, IN_PROGRESS and READY. All render part with IN_PROGRESS status will be highlighted in image editor. - For external engines render part's status is filling in automatically. Initially all render parts has got NONE status, then one external engine acquire render result, corresponding part will change status to IN_PROGRESS. As soon as render result is finished, corresponding render part will change status to FINISHED This should make it easy to highlight tiles for other engines as well.
2013-01-01 16:15:13 +00:00
/* in case of a permanent buffer, return it, otherwise we will allocate
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
* a new temporary buffer */
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(!params.background) {
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
tile_manager.state.buffer.get_offset_stride(rtile.offset, rtile.stride);
rtile.buffer = buffers->buffer.device_pointer;
rtile.rng_state = buffers->rng_state.device_pointer;
rtile.buffers = buffers;
device->map_tile(tile_device, rtile);
return true;
}
/* fill buffer parameters */
BufferParams buffer_params = tile_manager.params;
buffer_params.full_x = rtile.x;
buffer_params.full_y = rtile.y;
buffer_params.width = rtile.w;
buffer_params.height = rtile.h;
buffer_params.get_offset_stride(rtile.offset, rtile.stride);
Fix small memory leak in cycles tile render, a few bytes for each tile. Reported by "aquo" on IRC.
2013-02-06 21:32:34 +00:00
RenderBuffers *tilebuffers;
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
/* allocate buffers */
if(params.progressive_refine) {
tile_lock.lock();
if(tile_buffers.size() == 0)
tile_buffers.resize(tile_manager.state.num_tiles, NULL);
Fix #32951: Progressive refine crashing Blender when used with multiple cuda Issue was caused by offline rendering could have been allocated the same tile to different devices and in this case buffers would become invalid. Made it more clear in the code, so now it's flag for tile manager to indicate whether tiles should always be allocated for the same device or not. Also cleaned a way how tile index for progressive refine is calculating, which is now avoids tricky computation based on tile coordinate and it's dimensions.
2012-10-23 16:36:53 +00:00
tilebuffers = tile_buffers[tile.index];
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(tilebuffers == NULL) {
tilebuffers = new RenderBuffers(tile_device);
Fix #32951: Progressive refine crashing Blender when used with multiple cuda Issue was caused by offline rendering could have been allocated the same tile to different devices and in this case buffers would become invalid. Made it more clear in the code, so now it's flag for tile manager to indicate whether tiles should always be allocated for the same device or not. Also cleaned a way how tile index for progressive refine is calculating, which is now avoids tricky computation based on tile coordinate and it's dimensions.
2012-10-23 16:36:53 +00:00
tile_buffers[tile.index] = tilebuffers;
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
tilebuffers->reset(tile_device, buffer_params);
}
tile_lock.unlock();
}
else {
tilebuffers = new RenderBuffers(tile_device);
tilebuffers->reset(tile_device, buffer_params);
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
rtile.buffer = tilebuffers->buffer.device_pointer;
rtile.rng_state = tilebuffers->rng_state.device_pointer;
rtile.buffers = tilebuffers;
Highlight currently rendering tiles This commit implements highlight of tiles which are being currently rendered for both Blender Internal and Cycles (and should be possible to use it for other external engines as well). Couple of implementation details: - Added one extra boolean flag to render engine which should be set to truth if render engine wants to highlight tiles. If so, property use_highlight_tiles should be set to True. - Render Part's ready boolena was changed by status enum, which could be NONE, IN_PROGRESS and READY. All render part with IN_PROGRESS status will be highlighted in image editor. - For external engines render part's status is filling in automatically. Initially all render parts has got NONE status, then one external engine acquire render result, corresponding part will change status to IN_PROGRESS. As soon as render result is finished, corresponding render part will change status to FINISHED This should make it easy to highlight tiles for other engines as well.
2013-01-01 16:15:13 +00:00
/* this will tag tile as IN PROGRESS in blender-side render pipeline,
* which is needed to highlight currently rendering tile before first
* sample was processed for it
*/
update_tile_sample(rtile);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
return true;
}
void Session::update_tile_sample(RenderTile& rtile)
{
thread_scoped_lock tile_lock(tile_mutex);
if(update_render_tile_cb) {
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(params.progressive_refine == false) {
/* todo: optimize this by making it thread safe and removing lock */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
update_render_tile_cb(rtile);
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
}
update_status_time();
}
void Session::release_tile(RenderTile& rtile)
{
thread_scoped_lock tile_lock(tile_mutex);
if(write_render_tile_cb) {
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(params.progressive_refine == false) {
/* todo: optimize this by making it thread safe and removing lock */
write_render_tile_cb(rtile);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
delete rtile.buffers;
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
}
update_status_time();
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
void Session::run_cpu()
{
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
bool tiles_written = false;
Cycles: update buffers only once in a while when progressive refine is used It'll mimic the same behavior as regular tile rendering update and it should give pretty noticeable boost on simpler scenes.
2012-10-23 17:24:23 +00:00
last_update_time = time_dt();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
/* reset once to start */
thread_scoped_lock reset_lock(delayed_reset.mutex);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock buffers_lock(buffers_mutex);
thread_scoped_lock display_lock(display_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
reset_(delayed_reset.params, delayed_reset.samples);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
delayed_reset.do_reset = false;
}
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
while(!progress.get_cancel()) {
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* advance to next tile */
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
bool no_tiles = !tile_manager.next();
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
bool need_tonemap = false;
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
if(params.background) {
/* if no work left and in background mode, we can stop immediately */
Cycles: progress printing in background mode, print finished at the end and avoid duplicate prints.
2011-10-30 10:12:34 +00:00
if(no_tiles) {
progress.set_status("Finished");
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
break;
Cycles: progress printing in background mode, print finished at the end and avoid duplicate prints.
2011-10-30 10:12:34 +00:00
}
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
}
else {
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
/* if in interactive mode, and we are either paused or done for now,
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* wait for pause condition notify to wake up again */
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
thread_scoped_lock pause_lock(pause_mutex);
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
Fix #34205: Cycles: Zooming in rendered mode during update out of sync with intended zoom Made sure no pause_cond.wait() happens at the same time as session as resetting.
2013-02-12 13:06:52 +00:00
if(!pause && delayed_reset.do_reset) {
/* reset once to start */
thread_scoped_lock reset_lock(delayed_reset.mutex);
thread_scoped_lock buffers_lock(buffers_mutex);
thread_scoped_lock display_lock(display_mutex);
reset_(delayed_reset.params, delayed_reset.samples);
delayed_reset.do_reset = false;
}
else if(pause || no_tiles) {
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
update_status_time(pause, no_tiles);
while(1) {
double pause_start = time_dt();
pause_cond.wait(pause_lock);
paused_time += time_dt() - pause_start;
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
if(!params.background)
Fix #30710: cycles wrong render time after pause/unpause in viewport.
2012-04-04 16:11:44 +00:00
progress.set_start_time(start_time + paused_time);
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
update_status_time(pause, no_tiles);
progress.set_update();
if(!pause)
break;
}
}
if(progress.get_cancel())
break;
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
}
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
if(!no_tiles) {
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* buffers mutex is locked entirely while rendering each
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* sample, and released/reacquired on each iteration to allow
* reset and draw in between */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock buffers_lock(buffers_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* update scene */
update_scene();
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
if(device->error_message() != "")
progress.set_cancel(device->error_message());
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
if(progress.get_cancel())
break;
/* update status and timing */
update_status_time();
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* path trace */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
path_trace();
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
/* update status and timing */
update_status_time();
Cycles: some warning fixes, cpu device task tweaks, avoid unnecessary tonemap in non-viewport render, and some utility functions.
2011-09-08 18:58:07 +00:00
if(!params.background)
need_tonemap = true;
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
if(device->error_message() != "")
progress.set_cancel(device->error_message());
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
device->task_wait();
{
thread_scoped_lock reset_lock(delayed_reset.mutex);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
thread_scoped_lock buffers_lock(buffers_mutex);
thread_scoped_lock display_lock(display_mutex);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
if(delayed_reset.do_reset) {
/* reset rendering if request from main thread */
delayed_reset.do_reset = false;
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
reset_(delayed_reset.params, delayed_reset.samples);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
else if(need_tonemap) {
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* tonemap only if we do not reset, we don't we don't
style cleanup: some warnigs & spelling.
2013-02-06 14:02:19 +00:00
* want to show the result of an incomplete sample*/
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
tonemap();
}
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
if(device->error_message() != "")
progress.set_cancel(device->error_message());
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
tiles_written = update_progressive_refine(progress.get_cancel());
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
progress.set_update();
}
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(!tiles_written)
update_progressive_refine(true);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
void Session::run()
{
Cycles: compile opencl kernels in non-blocking thread, and don't crash on build failure but show error message in status text.
2011-09-02 00:10:03 +00:00
/* load kernels */
Fix #30246: rendering multiple render layers with CUDA still not working, should really be fixed now.
2012-02-23 19:48:18 +00:00
if(!kernels_loaded) {
progress.set_status("Loading render kernels (may take a few minutes the first time)");
Cycles: compile opencl kernels in non-blocking thread, and don't crash on build failure but show error message in status text.
2011-09-02 00:10:03 +00:00
Fix #30246: rendering multiple render layers with CUDA still not working, should really be fixed now.
2012-02-23 19:48:18 +00:00
if(!device->load_kernels(params.experimental)) {
string message = device->error_message();
if(message == "")
message = "Failed loading render kernel, see console for errors";
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
Fix #30246: rendering multiple render layers with CUDA still not working, should really be fixed now.
2012-02-23 19:48:18 +00:00
progress.set_status("Error", message);
progress.set_update();
return;
}
kernels_loaded = true;
Cycles: compile opencl kernels in non-blocking thread, and don't crash on build failure but show error message in status text.
2011-09-02 00:10:03 +00:00
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* session thread loop */
progress.set_status("Waiting for render to start");
/* run */
if(!progress.get_cancel()) {
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
/* reset number of rendered samples */
progress.reset_sample();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
if(device_use_gl)
run_gpu();
else
run_cpu();
}
/* progress update */
if(progress.get_cancel())
Cycles: improve error reporting for opencl and cuda, showing error messages in viewport instead of only console.
2011-11-22 20:49:33 +00:00
progress.set_status("Cancel", progress.get_cancel_message());
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
else
progress.set_update();
}
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
bool Session::draw(BufferParams& buffer_params)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
if(device_use_gl)
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
return draw_gpu(buffer_params);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
else
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
return draw_cpu(buffer_params);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
void Session::reset_(BufferParams& buffer_params, int samples)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(buffers) {
if(buffer_params.modified(buffers->params)) {
gpu_draw_ready = false;
buffers->reset(device, buffer_params);
display->reset(device, buffer_params);
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
tile_manager.reset(buffer_params, samples);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
start_time = time_dt();
preview_time = 0.0;
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
paused_time = 0.0;
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
if(!params.background)
Fix #30710: cycles wrong render time after pause/unpause in viewport.
2012-04-04 16:11:44 +00:00
progress.set_start_time(start_time + paused_time);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
void Session::reset(BufferParams& buffer_params, int samples)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
if(device_use_gl)
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
reset_gpu(buffer_params, samples);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
else
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
reset_cpu(buffer_params, samples);
Fix #33337: cycles crash with progressive refine and multiple render layers.
2012-11-28 19:57:29 +00:00
if(params.progressive_refine) {
thread_scoped_lock buffers_lock(buffers_mutex);
foreach(RenderBuffers *buffers, tile_buffers)
delete buffers;
tile_buffers.clear();
}
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
}
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
void Session::set_samples(int samples)
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
{
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
if(samples != params.samples) {
params.samples = samples;
tile_manager.set_samples(samples);
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
{
thread_scoped_lock pause_lock(pause_mutex);
}
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
pause_cond.notify_all();
Cycles: * Add alpha pass output, to use set Transparent option in Film panel. * Add Holdout closure (OSL terminology), this is like the Sky option in the internal renderer, objects with this closure show the background / zero alpha. * Add option to use Gaussian instead of Box pixel filter in the UI. * Remove camera response curves for now, they don't really belong here in the pipeline, should be moved to compositor. * Output full float values for rendering now, previously was only byte precision. * Add a patch from Thomas to get a preview passes option, but still disabled because it isn't quite working right yet. * CUDA: don't compile shader graph evaluation inline. * Convert tabs to spaces in python files.
2011-08-28 13:55:59 +00:00
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
void Session::set_pause(bool pause_)
{
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
bool notify = false;
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
{
thread_scoped_lock pause_lock(pause_mutex);
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
if(pause != pause_) {
pause = pause_;
notify = true;
}
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
}
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
if(notify)
pause_cond.notify_all();
Cycles: preview passes setting to set the max number of passes to render in the viewport working now, set to 0 for unlimited (well, actually 2147483647).
2011-08-29 10:21:10 +00:00
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
void Session::wait()
{
session_thread->join();
delete session_thread;
session_thread = NULL;
}
void Session::update_scene()
{
thread_scoped_lock scene_lock(scene->mutex);
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
/* update camera if dimensions changed for progressive render. the camera
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* knows nothing about progressive or cropped rendering, it just gets the
* image dimensions passed in */
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Camera *cam = scene->camera;
Cycles: some small code refactoring related to buffer parameters.
2011-12-21 13:48:35 +00:00
int width = tile_manager.state.buffer.full_width;
int height = tile_manager.state.buffer.full_height;
Cycles Hair: Strand Minimum Pixel Size Code is added to restrict the pixel size of strands in cycles. It works best with ribbon primitives and a preset for these is included. It uses distance dependent expansion of the strands and then stochastic strand removal to give a fading. To prevent a slowdown for triangle mesh objects in the BVH an extra visibility flag has been added. It is also only applied for camera rays. The strand width settings are also changed, so that the particle size is not included in the width calculation. Instead there is a separate particle system parameter for width scaling.
2013-04-15 21:38:31 +00:00
int resolution = tile_manager.state.resolution_divider;
Cycles: border rendering support, includes some refactoring in how pixels are accessed on devices.
2011-12-20 12:25:37 +00:00
if(width != cam->width || height != cam->height) {
cam->width = width;
cam->height = height;
Cycles Hair: Strand Minimum Pixel Size Code is added to restrict the pixel size of strands in cycles. It works best with ribbon primitives and a preset for these is included. It uses distance dependent expansion of the strands and then stochastic strand removal to give a fading. To prevent a slowdown for triangle mesh objects in the BVH an extra visibility flag has been added. It is also only applied for camera rays. The strand width settings are also changed, so that the particle size is not included in the width calculation. Instead there is a separate particle system parameter for width scaling.
2013-04-15 21:38:31 +00:00
cam->resolution = resolution;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
cam->tag_update();
}
Cycles: experimental correlated multi-jittered sampling pattern that can be used instead of sobol. So far one doesn't seem to be consistently better or worse than the other for the same number of samples but more testing is needed. The random number generator itself is slower than sobol for most number of samples, except 16, 64, 256, .. because they can be computed faster. This can probably be optimized, but we can do that when/if this actually turns out to be useful. Paper this implementation is based on: http://graphics.pixar.com/library/MultiJitteredSampling/ Also includes some refactoring of RNG code, fixing a Sobol correlation issue with the first BSDF and < 16 samples, skipping some unneeded RNG calls and using a simpler unit square to unit disk function.
2013-06-07 16:06:22 +00:00
/* number of samples is needed by multi jittered sampling pattern */
Integrator *integrator = scene->integrator;
if(integrator->sampling_pattern == SAMPLING_PATTERN_CMJ) {
int aa_samples = tile_manager.num_samples;
if(aa_samples != integrator->aa_samples) {
integrator->aa_samples = aa_samples;
integrator->tag_update(scene);
}
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* update scene */
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(scene->need_update()) {
progress.set_status("Updating Scene");
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
scene->device_update(device, progress);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
void Session::update_status_time(bool show_pause, bool show_done)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
int sample = tile_manager.state.sample;
Cycles: change preview "resolution divider" that gave the number of lower resolutions to render, to a "start resolution" which gives the resolution to start at. This avoids unnecessary rendering of small resolutions in small viewports, and avoids long waiting on big viewports.
2012-09-17 10:55:18 +00:00
int resolution = tile_manager.state.resolution_divider;
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
int num_tiles = tile_manager.state.num_tiles;
int tile = tile_manager.state.num_rendered_tiles;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* update status */
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
string status, substatus;
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(!params.progressive) {
Cycles: multi-gpu rendering doesn't know for sure which sample is being sampled, so only report tile number is being processed
2012-11-05 08:04:41 +00:00
bool is_gpu = params.device.type == DEVICE_CUDA || params.device.type == DEVICE_OPENCL;
bool is_multidevice = params.device.multi_devices.size() > 1;
bool is_cpu = params.device.type == DEVICE_CPU;
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
substatus = string_printf("Path Tracing Tile %d/%d", tile, num_tiles);
Cycles: multi-gpu rendering doesn't know for sure which sample is being sampled, so only report tile number is being processed
2012-11-05 08:04:41 +00:00
if((is_gpu && !is_multidevice) || (is_cpu && num_tiles == 1)) {
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
/* when rendering on GPU multithreading happens within single tile, as in
* tiles are handling sequentially and in this case we could display
* currently rendering sample number
* this helps a lot from feedback point of view.
* also display the info on CPU, when using 1 tile only
*/
Fix another part of #34877: cycles progress status text not showing correct with per render layer samples in addition to the progress bar. Also fixed job progress bar not working at all on high DPI / retina, was so small the actual progress was not visible.
2013-04-05 17:57:26 +00:00
int sample = progress.get_sample(), num_samples = tile_manager.num_samples;
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
if(tile > 1) {
/* sample counter is global for all tiles, subtract samples
* from already finished tiles to get sample counter for
* current tile only
*/
sample -= (tile - 1) * num_samples;
}
substatus += string_printf(", Sample %d/%d", sample, num_samples);
}
}
Code cleanup / Cycles: * Use USHRT_MAX rather than manual value, suggested by Campbell.
2013-07-19 22:56:54 +00:00
else if(tile_manager.num_samples == USHRT_MAX)
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
substatus = string_printf("Path Tracing Sample %d", sample+1);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
else
Fix another part of #34877: cycles progress status text not showing correct with per render layer samples in addition to the progress bar. Also fixed job progress bar not working at all on high DPI / retina, was so small the actual progress was not visible.
2013-04-05 17:57:26 +00:00
substatus = string_printf("Path Tracing Sample %d/%d", sample+1, tile_manager.num_samples);
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
Render stats text: show elapsed time for blender internal, hide useless "Single Layer" at the start, more clearly indicate what the render time of the last frame was, some other tweaks for consistency.
2013-06-24 22:41:24 +00:00
if(show_pause) {
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
status = "Paused";
Render stats text: show elapsed time for blender internal, hide useless "Single Layer" at the start, more clearly indicate what the render time of the last frame was, some other tweaks for consistency.
2013-06-24 22:41:24 +00:00
}
else if(show_done) {
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
status = "Done";
Render stats text: show elapsed time for blender internal, hide useless "Single Layer" at the start, more clearly indicate what the render time of the last frame was, some other tweaks for consistency.
2013-06-24 22:41:24 +00:00
}
else {
status = substatus;
substatus = "";
}
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
progress.set_status(status, substatus);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* update timing */
if(preview_time == 0.0 && resolution == 1)
preview_time = time_dt();
Cycles: show elapsed time for F12/background render.
2012-01-09 16:57:46 +00:00
avoid divide by zero in cycles progress calculation
2012-11-13 01:53:07 +00:00
double tile_time = (tile == 0 || sample == 0)? 0.0: (time_dt() - preview_time - paused_time) / sample;
Cycles: pause button to interrupt viewport renders, in the 3d view header.
2011-08-29 16:54:13 +00:00
/* negative can happen when we pause a bit before rendering, can discard that */
if(preview_time < 0.0) preview_time = 0.0;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
progress.set_tile(tile, tile_time);
}
void Session::update_progress_sample()
{
progress.increment_sample();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
void Session::path_trace()
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
/* add path trace task */
DeviceTask task(DeviceTask::PATH_TRACE);
Cycles: merge of changes from tomato branch. Regular rendering now works tiled, and supports save buffers to save memory during render and cache render results. Brick texture node by Thomas. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture Image texture Blended Box Mapping. http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texture http://mango.blender.org/production/blended_box/ Various bug fixes by Sergey and Campbell. * Fix for reading freed memory in some node setups. * Fix incorrect memory read when synchronizing mesh motion. * Fix crash appearing when direct light usage is different on different layers. * Fix for vector pass gives wrong result in some circumstances. * Fix for wrong resolution used for rendering Render Layer node. * Option to cancel rendering when doing initial synchronization. * No more texture limit when using CPU render. * Many fixes for new tiled rendering.
2012-09-04 13:29:07 +00:00
task.acquire_tile = function_bind(&Session::acquire_tile, this, _1, _2);
task.release_tile = function_bind(&Session::release_tile, this, _1);
task.get_cancel = function_bind(&Progress::get_cancel, &this->progress);
task.update_tile_sample = function_bind(&Session::update_tile_sample, this, _1);
task.update_progress_sample = function_bind(&Session::update_progress_sample, this);
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
task.need_finish_queue = params.progressive_refine;
Cycles: more code refactoring to rename things internally as well. Also change property name back so we keep compatibility.
2013-08-23 14:34:34 +00:00
task.integrator_branched = scene->integrator->method == Integrator::BRANCHED_PATH;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
device->task_add(task);
}
void Session::tonemap()
{
/* add tonemap task */
Cycles: viewport render now takes scene color management settings into account, except for curves, that's still missing from the OpenColorIO GLSL shader. The pixels are stored in a half float texture, converterd from full float with native GPU instructions and SIMD on the CPU, so it should be pretty quick. Using a GLSL shader is useful for GPU render because it avoids a copy through CPU memory.
2013-08-30 23:49:38 +00:00
DeviceTask task(DeviceTask::FILM_CONVERT);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: some small code refactoring related to buffer parameters.
2011-12-21 13:48:35 +00:00
task.x = tile_manager.state.buffer.full_x;
task.y = tile_manager.state.buffer.full_y;
task.w = tile_manager.state.buffer.width;
task.h = tile_manager.state.buffer.height;
Cycles: viewport render now takes scene color management settings into account, except for curves, that's still missing from the OpenColorIO GLSL shader. The pixels are stored in a half float texture, converterd from full float with native GPU instructions and SIMD on the CPU, so it should be pretty quick. Using a GLSL shader is useful for GPU render because it avoids a copy through CPU memory.
2013-08-30 23:49:38 +00:00
task.rgba_byte = display->rgba_byte.device_pointer;
task.rgba_half = display->rgba_half.device_pointer;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
task.buffer = buffers->buffer.device_pointer;
Cycles: tweaks to properties and nodes * Passes renamed to samples * Camera lens radius renamed to aperature size/blades/rotation * Glass and fresnel nodes input is now index of refraction * Glossy and velvet fresnel socket removed * Mix/add closure node renamed to mix/add shader node * Blend weight node added for shader mixing weights There is some version patching code for reading existing files, but it's not perfect, so shaders may work a bit different.
2011-09-16 13:14:02 +00:00
task.sample = tile_manager.state.sample;
Cycles: some small code refactoring related to buffer parameters.
2011-12-21 13:48:35 +00:00
tile_manager.state.buffer.get_offset_stride(task.offset, task.stride);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: fix cuda assertion that could be triggerd sometimes, when trying to tonemap an empty image.
2011-08-29 11:47:03 +00:00
if(task.w > 0 && task.h > 0) {
device->task_add(task);
device->task_wait();
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: fix cuda assertion that could be triggerd sometimes, when trying to tonemap an empty image.
2011-08-29 11:47:03 +00:00
/* set display to new size */
display->draw_set(task.w, task.h);
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
display_outdated = false;
}
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
bool Session::update_progressive_refine(bool cancel)
{
int sample = tile_manager.state.sample + 1;
Fix another part of #34877: cycles progress status text not showing correct with per render layer samples in addition to the progress bar. Also fixed job progress bar not working at all on high DPI / retina, was so small the actual progress was not visible.
2013-04-05 17:57:26 +00:00
bool write = sample == tile_manager.num_samples || cancel;
Cycles: update buffers only once in a while when progressive refine is used It'll mimic the same behavior as regular tile rendering update and it should give pretty noticeable boost on simpler scenes.
2012-10-23 17:24:23 +00:00
double current_time = time_dt();
code cleanup: quiet double promotion warnings
2012-11-07 23:52:33 +00:00
if (current_time - last_update_time < 1.0) {
Cycles: update buffers only once in a while when progressive refine is used It'll mimic the same behavior as regular tile rendering update and it should give pretty noticeable boost on simpler scenes.
2012-10-23 17:24:23 +00:00
/* if last sample was processed, we need to write buffers anyway */
if (!write)
return false;
}
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
if(params.progressive_refine) {
foreach(RenderBuffers *buffers, tile_buffers) {
RenderTile rtile;
rtile.buffers = buffers;
rtile.sample = sample;
Cycles: update buffers only once in a while when progressive refine is used It'll mimic the same behavior as regular tile rendering update and it should give pretty noticeable boost on simpler scenes.
2012-10-23 17:24:23 +00:00
if(write)
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
write_render_tile_cb(rtile);
else
update_render_tile_cb(rtile);
}
}
Cycles: update buffers only once in a while when progressive refine is used It'll mimic the same behavior as regular tile rendering update and it should give pretty noticeable boost on simpler scenes.
2012-10-23 17:24:23 +00:00
last_update_time = current_time;
return write;
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: persistent images option This option enables keeping loaded images in the memory in-between of rendering. Implemented by keeping render engine alive for until Render structure is being freed. Cycles will free all data when render finishes, optionally keeping image manager untouched. All shaders, meshes, objects will be re-allocated next time rendering happens. Cycles cession and scene will be re-created from scratch if render/ scene parameters were changed. This will also allow to keep compiled OSL shaders in memory without need to re-compile them again. P.S. Performance panel could be cleaned up a bit, not so much happy with it's vertical alignment currently but not sure how to make it look better. P.P.S. Currently the only way to free images from the device is to disable Persistent Images option and start rendering.
2012-11-09 08:46:53 +00:00
void Session::device_free()
{
scene->device_free();
foreach(RenderBuffers *buffers, tile_buffers)
delete buffers;
tile_buffers.clear();
/* used from background render only, so no need to
* re-create render/display buffers here
*/
}
Cycles: progressive refine option Just makes progressive refine :) This means the whole image would be refined gradually using as much threads as it's set in performance settings. Having enough tiles is required to have this option working as it's expected. Technically it's implemented by repeatedly computing next sample for all the tiles before switching to next sample. This works around 7-12% slower than regular tile-based rendering, so use this option only if you really need it. This commit also fixes progressive update of image when Save Buffers option is enabled. And one more thing this commit fixes is handling display buffer with Save Buffers option enabled. If this option is enabled image buffer wouldn't have neither byte nor float buffer until image is fully rendered which could backfire in missing image while rendering in cases color management cache became full. This issue solved by allocating byte buffer for image buffer from tile update callback. Patch was reviewed by Brecht. He also made some minor edits to original version to patch. Thanks, man!
2012-10-13 12:38:32 +00:00
CCL_NAMESPACE_END
Copy Permalink