Why so significant difference in render results??

[Walker]

If you just looking for an old version, check out the archive. The code repository is here: http://src.luxrender.net/

I am looking for source code of the version that is proposed to download at http://www.luxrender.net/en_GB/standalone#windows (v0.8 for windows x64 no open cl). I cant find nesessary source code at http://src.luxrender.net/ to build that version. As i understand, you give me link http://src.luxrender.net/lux/archive/45d3e13eb94c.zip for lux folder that is necessary for building version what i need , right? now i need windows and luxrays folder. but i cant find it in archive... It would be amazing if you can give me this links.

[SATtva]

Have you looked into the archive? The precompiled version of Lux 0.8 for Win x64 (without OpenCL) is located here, just click and download. If you for some reason are going to compile it yourself, i guess you could figure out where to get the relevant pieces on your own.

I'd prefer not to participate in this topic anymore.

[Walker]

SATtva, thanks for your help.

[Meelis]

Here are some weak internal reflections beacause i have only HDR light source.
And with Bidir in 0.9dev ~2 hours --- 850 S/p

[Walker]

Here are some weak internal reflections beacause i have only HDR light source.
And with Bidir in 0.9dev ~2 hours --- 850 S/p

Can you give me your render settings such Sampler, Accelerator, SurfaceIntegrator, VolumeIntegrator, PixelFilter?
Or something else that create internal reflection on your mind.

[Meelis]

Can you give me your render settings such Sampler, Accelerator, SurfaceIntegrator, VolumeIntegrator, PixelFilter?
Or something else that create internal reflection on your mind.

internal reflection will just happen depending on environment lighting.

Like on HDR map how high dynamic range is the map, how bright is sun compared to background.
Bigest impact is the light source size and how close or far it is from glass volume http://www.luxrender.net/wiki/Creating_Beautiful_Caustics

BTW i don't recommend ERPT method because the glow problem:


But compared with random sampler, it kind of magnifies caustics:


1'st image was with Tropical Beach HDR as only light source http://www.hdrlabs.com/sibl/archive/

2'nt and 3'rd was with Ice Lake http://www.hdrlabs.com/sibl/archive/

I think the 1'st image was something like:

Code: Select all

# Main Scene File

Renderer "sampler"

Sampler "random"
   "integer pixelsamples" [16]
   "string pixelsampler" ["vegas"]

Accelerator "qbvh"

SurfaceIntegrator "bidirectional"
   "integer eyedepth" [1024]
   "integer lightdepth" [1024]
   "string strategy" ["all"]
   "float eyerrthreshold" [1]
   "float lightrrthreshold" [1]

VolumeIntegrator "multi"
   "float stepsize" [0.010000000000000]

PixelFilter "gaussian"
   "float xwidth" [3.000000000000000]
   "float ywidth" [3.000000000000000]
   "float alpha" [3.000000000000000]

Here is 9 hour 1500 S/p example i did with following settings:

Code: Select all

Renderer "sampler"

Sampler "random"
   "integer pixelsamples" [64]
   "string pixelsampler" ["vegas"]

Accelerator "qbvh"

SurfaceIntegrator "bidirectional"
   "integer eyedepth" [256]
   "integer lightdepth" [256]
   "string strategy" ["one"]
   "float eyerrthreshold" [1.000000000000000]
   "float lightrrthreshold" [1.000000000000000]

VolumeIntegrator "multi"
   "float stepsize" [0.009999999776483]

PixelFilter "gaussian"
   "float xwidth" [3.000000000000000]
   "float ywidth" [3.000000000000000]
   "float alpha" [3.000000000000000]

Rendered on 2 dual cores

[Walker]

Meelis, thank you for detailed answer.