sandey/blender
1
0
Fork 0
forked from bartvdbraak/blender
Code Pull Requests Activity
33e12a2983
blender/intern/cycles/kernel/svm/svm_image.h

272 lines
12 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
/*
* Copyright 2011, Blender Foundation.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
CCL_NAMESPACE_BEGIN
Cycles: OpenCL image texture support, fix an attribute node issue and refactor feature enabling #defines a bit.
2012-05-13 12:32:44 +00:00
#ifdef __KERNEL_OPENCL__
/* For OpenCL all images are packed in a single array, and we do manual lookup
* and interpolation. */
__device_inline float4 svm_image_texture_read(KernelGlobals *kg, int offset)
{
uchar4 r = kernel_tex_fetch(__tex_image_packed, offset);
float f = 1.0f/255.0f;
return make_float4(r.x*f, r.y*f, r.z*f, r.w*f);
}
__device_inline int svm_image_texture_wrap_periodic(int x, int width)
{
x %= width;
if(x < 0)
x += width;
return x;
}
__device_inline int svm_image_texture_wrap_clamp(int x, int width)
{
return clamp(x, 0, width-1);
}
__device_inline float svm_image_texture_frac(float x, int *ix)
{
int i = (int)x - ((x < 0.0f)? 1: 0);
*ix = i;
return x - (float)i;
}
__device float4 svm_image_texture(KernelGlobals *kg, int id, float x, float y)
{
uint4 info = kernel_tex_fetch(__tex_image_packed_info, id);
uint width = info.x;
uint height = info.y;
uint offset = info.z;
uint periodic = info.w;
int ix, iy, nix, niy;
float tx = svm_image_texture_frac(x*width, &ix);
float ty = svm_image_texture_frac(y*height, &iy);
if(periodic) {
ix = svm_image_texture_wrap_periodic(ix, width);
iy = svm_image_texture_wrap_periodic(iy, height);
nix = svm_image_texture_wrap_periodic(ix+1, width);
niy = svm_image_texture_wrap_periodic(iy+1, height);
}
else {
ix = svm_image_texture_wrap_clamp(ix, width);
iy = svm_image_texture_wrap_clamp(iy, height);
nix = svm_image_texture_wrap_clamp(ix+1, width);
niy = svm_image_texture_wrap_clamp(iy+1, height);
}
float4 r = (1.0f - ty)*(1.0f - tx)*svm_image_texture_read(kg, offset + ix + iy*width);
r += (1.0f - ty)*tx*svm_image_texture_read(kg, offset + nix + iy*width);
r += ty*(1.0f - tx)*svm_image_texture_read(kg, offset + ix + niy*width);
r += ty*tx*svm_image_texture_read(kg, offset + nix + niy*width);
return r;
}
#else
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 float4 svm_image_texture(KernelGlobals *kg, int id, float x, float y)
{
float4 r;
/* not particularly proud of this massive switch, what are the
style cleanup: block comments
2012-06-09 17:22:52 +00:00
* alternatives?
* - use a single big 1D texture, and do our own lookup/filtering
* - group by size and use a 3d texture, performance impact
* - group into larger texture with some padding for correct lerp
*
* also note that cuda has 128 textures limit, we use 100 now, since
* we still need some for other storage */
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
switch(id) {
case 0: r = kernel_tex_image_interp(__tex_image_000, x, y); break;
case 1: r = kernel_tex_image_interp(__tex_image_001, x, y); break;
case 2: r = kernel_tex_image_interp(__tex_image_002, x, y); break;
case 3: r = kernel_tex_image_interp(__tex_image_003, x, y); break;
case 4: r = kernel_tex_image_interp(__tex_image_004, x, y); break;
case 5: r = kernel_tex_image_interp(__tex_image_005, x, y); break;
case 6: r = kernel_tex_image_interp(__tex_image_006, x, y); break;
case 7: r = kernel_tex_image_interp(__tex_image_007, x, y); break;
case 8: r = kernel_tex_image_interp(__tex_image_008, x, y); break;
case 9: r = kernel_tex_image_interp(__tex_image_009, x, y); break;
case 10: r = kernel_tex_image_interp(__tex_image_010, x, y); break;
case 11: r = kernel_tex_image_interp(__tex_image_011, x, y); break;
case 12: r = kernel_tex_image_interp(__tex_image_012, x, y); break;
case 13: r = kernel_tex_image_interp(__tex_image_013, x, y); break;
case 14: r = kernel_tex_image_interp(__tex_image_014, x, y); break;
case 15: r = kernel_tex_image_interp(__tex_image_015, x, y); break;
case 16: r = kernel_tex_image_interp(__tex_image_016, x, y); break;
case 17: r = kernel_tex_image_interp(__tex_image_017, x, y); break;
case 18: r = kernel_tex_image_interp(__tex_image_018, x, y); break;
case 19: r = kernel_tex_image_interp(__tex_image_019, x, y); break;
case 20: r = kernel_tex_image_interp(__tex_image_020, x, y); break;
case 21: r = kernel_tex_image_interp(__tex_image_021, x, y); break;
case 22: r = kernel_tex_image_interp(__tex_image_022, x, y); break;
case 23: r = kernel_tex_image_interp(__tex_image_023, x, y); break;
case 24: r = kernel_tex_image_interp(__tex_image_024, x, y); break;
case 25: r = kernel_tex_image_interp(__tex_image_025, x, y); break;
case 26: r = kernel_tex_image_interp(__tex_image_026, x, y); break;
case 27: r = kernel_tex_image_interp(__tex_image_027, x, y); break;
case 28: r = kernel_tex_image_interp(__tex_image_028, x, y); break;
case 29: r = kernel_tex_image_interp(__tex_image_029, x, y); break;
case 30: r = kernel_tex_image_interp(__tex_image_030, x, y); break;
case 31: r = kernel_tex_image_interp(__tex_image_031, x, y); break;
case 32: r = kernel_tex_image_interp(__tex_image_032, x, y); break;
case 33: r = kernel_tex_image_interp(__tex_image_033, x, y); break;
case 34: r = kernel_tex_image_interp(__tex_image_034, x, y); break;
case 35: r = kernel_tex_image_interp(__tex_image_035, x, y); break;
case 36: r = kernel_tex_image_interp(__tex_image_036, x, y); break;
case 37: r = kernel_tex_image_interp(__tex_image_037, x, y); break;
case 38: r = kernel_tex_image_interp(__tex_image_038, x, y); break;
case 39: r = kernel_tex_image_interp(__tex_image_039, x, y); break;
case 40: r = kernel_tex_image_interp(__tex_image_040, x, y); break;
case 41: r = kernel_tex_image_interp(__tex_image_041, x, y); break;
case 42: r = kernel_tex_image_interp(__tex_image_042, x, y); break;
case 43: r = kernel_tex_image_interp(__tex_image_043, x, y); break;
case 44: r = kernel_tex_image_interp(__tex_image_044, x, y); break;
case 45: r = kernel_tex_image_interp(__tex_image_045, x, y); break;
case 46: r = kernel_tex_image_interp(__tex_image_046, x, y); break;
case 47: r = kernel_tex_image_interp(__tex_image_047, x, y); break;
case 48: r = kernel_tex_image_interp(__tex_image_048, x, y); break;
case 49: r = kernel_tex_image_interp(__tex_image_049, x, y); break;
case 50: r = kernel_tex_image_interp(__tex_image_050, x, y); break;
case 51: r = kernel_tex_image_interp(__tex_image_051, x, y); break;
case 52: r = kernel_tex_image_interp(__tex_image_052, x, y); break;
case 53: r = kernel_tex_image_interp(__tex_image_053, x, y); break;
case 54: r = kernel_tex_image_interp(__tex_image_054, x, y); break;
case 55: r = kernel_tex_image_interp(__tex_image_055, x, y); break;
case 56: r = kernel_tex_image_interp(__tex_image_056, x, y); break;
case 57: r = kernel_tex_image_interp(__tex_image_057, x, y); break;
case 58: r = kernel_tex_image_interp(__tex_image_058, x, y); break;
case 59: r = kernel_tex_image_interp(__tex_image_059, x, y); break;
case 60: r = kernel_tex_image_interp(__tex_image_060, x, y); break;
case 61: r = kernel_tex_image_interp(__tex_image_061, x, y); break;
case 62: r = kernel_tex_image_interp(__tex_image_062, x, y); break;
case 63: r = kernel_tex_image_interp(__tex_image_063, x, y); break;
case 64: r = kernel_tex_image_interp(__tex_image_064, x, y); break;
case 65: r = kernel_tex_image_interp(__tex_image_065, x, y); break;
case 66: r = kernel_tex_image_interp(__tex_image_066, x, y); break;
case 67: r = kernel_tex_image_interp(__tex_image_067, x, y); break;
case 68: r = kernel_tex_image_interp(__tex_image_068, x, y); break;
case 69: r = kernel_tex_image_interp(__tex_image_069, x, y); break;
case 70: r = kernel_tex_image_interp(__tex_image_070, x, y); break;
case 71: r = kernel_tex_image_interp(__tex_image_071, x, y); break;
case 72: r = kernel_tex_image_interp(__tex_image_072, x, y); break;
case 73: r = kernel_tex_image_interp(__tex_image_073, x, y); break;
case 74: r = kernel_tex_image_interp(__tex_image_074, x, y); break;
case 75: r = kernel_tex_image_interp(__tex_image_075, x, y); break;
case 76: r = kernel_tex_image_interp(__tex_image_076, x, y); break;
case 77: r = kernel_tex_image_interp(__tex_image_077, x, y); break;
case 78: r = kernel_tex_image_interp(__tex_image_078, x, y); break;
case 79: r = kernel_tex_image_interp(__tex_image_079, x, y); break;
case 80: r = kernel_tex_image_interp(__tex_image_080, x, y); break;
case 81: r = kernel_tex_image_interp(__tex_image_081, x, y); break;
case 82: r = kernel_tex_image_interp(__tex_image_082, x, y); break;
case 83: r = kernel_tex_image_interp(__tex_image_083, x, y); break;
case 84: r = kernel_tex_image_interp(__tex_image_084, x, y); break;
case 85: r = kernel_tex_image_interp(__tex_image_085, x, y); break;
case 86: r = kernel_tex_image_interp(__tex_image_086, x, y); break;
case 87: r = kernel_tex_image_interp(__tex_image_087, x, y); break;
case 88: r = kernel_tex_image_interp(__tex_image_088, x, y); break;
case 89: r = kernel_tex_image_interp(__tex_image_089, x, y); break;
case 90: r = kernel_tex_image_interp(__tex_image_090, x, y); break;
case 91: r = kernel_tex_image_interp(__tex_image_091, x, y); break;
case 92: r = kernel_tex_image_interp(__tex_image_092, x, y); break;
case 93: r = kernel_tex_image_interp(__tex_image_093, x, y); break;
case 94: r = kernel_tex_image_interp(__tex_image_094, x, y); break;
Cycles: float texture support. Due to GPU limitations there are now 95 byte, and 5 float image textures. For CPU render this limit will be lifted later on with image cache support. Patch by Mike Farnsworth. Also changed color space option in image/environment texture node, to show options Color and Non-Color Data, instead of sRGB and Linear, this is more descriptive, and it was not really correct to equate Non-Color Data with Linear.
2012-03-07 12:27:18 +00:00
case 95: r = kernel_tex_image_interp(__tex_image_float_095, x, y); break;
case 96: r = kernel_tex_image_interp(__tex_image_float_096, x, y); break;
case 97: r = kernel_tex_image_interp(__tex_image_float_097, x, y); break;
case 98: r = kernel_tex_image_interp(__tex_image_float_098, x, y); break;
case 99: r = kernel_tex_image_interp(__tex_image_float_099, x, y); 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
default:
kernel_assert(0);
return make_float4(0.0f, 0.0f, 0.0f, 0.0f);
}
return r;
}
Cycles: OpenCL image texture support, fix an attribute node issue and refactor feature enabling #defines a bit.
2012-05-13 12:32:44 +00:00
#endif
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +00:00
__device void svm_node_tex_image(KernelGlobals *kg, ShaderData *sd, float *stack, uint4 node)
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: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +00:00
uint id = node.y;
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
uint co_offset, out_offset, alpha_offset, srgb;
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +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
decode_node_uchar4(node.z, &co_offset, &out_offset, &alpha_offset, &srgb);
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +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
float3 co = stack_load_float3(stack, co_offset);
float4 f = svm_image_texture(kg, id, co.x, co.y);
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +00:00
float3 r = make_float3(f.x, f.y, f.z);
Cycles: some steps to getting OpenCL backend to compile.
2011-05-20 12:26:01 +00:00
if(srgb) {
r.x = color_srgb_to_scene_linear(r.x);
r.y = color_srgb_to_scene_linear(r.y);
r.z = color_srgb_to_scene_linear(r.z);
}
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
if(stack_valid(out_offset))
stack_store_float3(stack, out_offset, r);
if(stack_valid(alpha_offset))
stack_store_float(stack, alpha_offset, f.w);
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: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +00:00
__device void svm_node_tex_environment(KernelGlobals *kg, ShaderData *sd, float *stack, uint4 node)
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: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +00:00
uint id = node.y;
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
uint co_offset, out_offset, alpha_offset, srgb;
Cycles: support for environment texture "Mirror Ball" projection mode, next to existing "Equirectangular". This projection is useful to create light probes from a chrome ball placed in a real scene. It expects as input a photograph of the chrome ball, cropped so the ball just fits inside the image boundaries. Example setup with panorama camera and mixing two (poor quality) photographs from different viewpoints to avoid stretching and hide the photographer: http://www.pasteall.org/pic/28036
2012-03-08 19:52:58 +00:00
uint projection = node.w;
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +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
decode_node_uchar4(node.z, &co_offset, &out_offset, &alpha_offset, &srgb);
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +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
float3 co = stack_load_float3(stack, co_offset);
Cycles: support for environment texture "Mirror Ball" projection mode, next to existing "Equirectangular". This projection is useful to create light probes from a chrome ball placed in a real scene. It expects as input a photograph of the chrome ball, cropped so the ball just fits inside the image boundaries. Example setup with panorama camera and mixing two (poor quality) photographs from different viewpoints to avoid stretching and hide the photographer: http://www.pasteall.org/pic/28036
2012-03-08 19:52:58 +00:00
float2 uv;
co = normalize(co);
if(projection == 0)
uv = direction_to_equirectangular(co);
else
uv = direction_to_mirrorball(co);
Sample as Lamp option for world shaders, to enable multiple importance sampling. By default lighting from the world is computed solely with indirect light sampling. However for more complex environment maps this can be too noisy, as sampling the BSDF may not easily find the highlights in the environment map image. By enabling this option, the world background will be sampled as a lamp, with lighter parts automatically given more samples. Map Resolution specifies the size of the importance map (res x res). Before rendering starts, an importance map is generated by "baking" a grayscale image from the world shader. This will then be used to determine which parts of the background are light and so should receive more samples than darker parts. Higher resolutions will result in more accurate sampling but take more setup time and memory. Patch by Mike Farnsworth, thanks!
2012-01-20 17:49:17 +00:00
float4 f = svm_image_texture(kg, id, uv.x, uv.y);
Cycles: color space control for image/environment texture nodes. Ideally would be automated but need to think about how to do this, not so simply in a node system. But guideline for now is, for color textures set to sRGB, for things like bump or roughness map, set to Linear.
2011-05-13 14:32:08 +00:00
float3 r = make_float3(f.x, f.y, f.z);
Cycles: some steps to getting OpenCL backend to compile.
2011-05-20 12:26:01 +00:00
if(srgb) {
r.x = color_srgb_to_scene_linear(r.x);
r.y = color_srgb_to_scene_linear(r.y);
r.z = color_srgb_to_scene_linear(r.z);
}
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
if(stack_valid(out_offset))
stack_store_float3(stack, out_offset, r);
if(stack_valid(alpha_offset))
stack_store_float(stack, alpha_offset, f.w);
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
}
CCL_NAMESPACE_END
Copy Permalink