Commit Graph

9 Commits

Author SHA1 Message Date
Thomas Dinges
9d236ac06c Cycles: Enable half float support (4 channels and 1 channel) on CUDA.
Atm OpenEXR half files benefit from this and will use only 1/2 of the memory now. More space for HDRs!

Part of my GSoC 2016.
2016-08-11 22:47:53 +02:00
Thomas Dinges
6311a9ff23 Cycles: Support half and half4 textures.
This is an initial commit for half texture support in Cycles.
It adds the basic infrastructure inside of the ImageManager and support for these textures on CPU.

Supported:
* Half Float OpenEXR images (can be used for e.g HDRs or Normalmaps) now use 1/2 the memory, when loaded via disk (OIIO).

ToDo:
Various things like support for inbuilt half textures, GPU... will come later, step by step.

Part of my GSoC 2016.
2016-06-19 17:31:16 +02:00
Sv. Lockal
3ec168465d Cycles: fix compilation on 32-bit Windows for half-floats
Reported by IRC user HG1.
2015-05-10 19:06:43 +00:00
Sv. Lockal
2ec221aa28 Cycles: Use native float->half conversion instructions for Haswell CPUs.
This makes OCIO viewport color correction a little bit faster (about -0.5s for 100 samples)
Also set max half float value to 65504.0 to conform with IEEE 754.
2015-05-10 16:35:51 +00:00
Thomas Dinges
ee36e75b85 Cleanup: Fix Cycles Apache header.
This was already mixed a bit, but the dot belongs there.
2014-12-25 02:50:24 +01:00
Thomas Dinges
cd5e1ff74e Cycles Refactor: Add SSE Utility code from Embree for cleaner SSE code.
This makes the code a bit easier to understand, and might come in handy
if we want to reuse more Embree code.

Differential Revision: https://developer.blender.org/D482

Code by Brecht, with fixes by Lockal, Sergey and myself.
2014-06-13 21:59:12 +02:00
Sv. Lockal
5031ca05aa Cycles: compile fix after rB7808360c5f (own mistake) 2014-02-27 17:29:35 +04:00
Sv. Lockal
7808360c5f Cycles: fix crash in SSE hair and half-floats on x86+vc2008
MSVC 2008 ignores alignement attribute when assigning from unaligned
float4 vector, returned from other function. Now Cycles uses unaligned
loads instead of casts for win32 in x86 mode.
2014-02-27 15:01:20 +04:00
d9e52ac98b Code cleanup: move half float functions to separate header file. 2014-01-15 15:29:22 +01:00