- 18 11月, 2020 2 次提交
-
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
No functionality changes. Cleaned up some of the GPU *WorkItem definitions.
-
- 17 11月, 2020 1 次提交
-
-
由 Matt Pharr 提交于
-
- 16 11月, 2020 1 次提交
-
-
由 Matt Pharr 提交于
-
- 15 11月, 2020 2 次提交
-
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
Changes scalar contribution function used in MLT to the luminance of the sample rather than the raw average of the spectral sample values. Otherwise just cosmetic changes.
-
- 14 11月, 2020 6 次提交
-
-
由 Matt Pharr 提交于
Added io_scene_pbrt to blender export directory readme.
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
(And a few small cleanups / cosmetic changes.)
-
由 Matt Pharr 提交于
Fixes issues with interpolating RGB causing spectra to have different bounds than if corresponding SampledSpectrums were interpolated. Issue #82.
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
- 13 11月, 2020 3 次提交
-
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
Issue #82
-
由 Matt Pharr 提交于
-
- 12 11月, 2020 10 次提交
-
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
When rendering on the CPU but with a build that supports the GPU, there's no need to copy the RGB->Spectrum tables into newly allocated memory.
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
- 11 11月, 2020 9 次提交
-
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
-
由 Matt Pharr 提交于
This matches how e.g. the PathIntegrator does it and saves tracing a ray at each intersection point.
-
由 Matt Pharr 提交于
The number of iterations is now determined by the number of pixel samples provided by the sampler. Further, the specified pixel sampler is now used for the camera paths rather than always using a Halton sampler. The "regularize" option was removed just to slightly simplify the implementation.
-
- 10 11月, 2020 5 次提交
-
-
由 Matt Pharr 提交于
Issue #75.
-
由 Matt Pharr 提交于
And issue an error rather than crashing. Also, added a CHECK in the AliasTable constructor to actively catch the case of all-zero weights being provided. Issue #77.
-
由 Matt Pharr 提交于
Issue #78
-
由 Matt Pharr 提交于
The old version was all sorts of buggy if one of the BSDFs terminated secondary wavelengths (including thread races.) Now we start each photon and camera path from scratch with the originally-sampled set of wavelengths. Each is allowed to terminate secondary wavelengths. When photon contributions are added, then we need to fall back to a single wavelength if either the photon path terminated them or the visible point is along a camera path that terminated them.
-
由 Matt Pharr 提交于
-
- 08 11月, 2020 1 次提交
-
-
由 Matt Pharr 提交于
Reworked some details of the bilinear patch intersection code.
-