LuxCoreRender Forums

Sharlybg wrote: Fri Sep 04, 2020 3:40 pm Wich setting control number of sample before transition ?

Code: Select all

            # Each preview step is rendered for n frames.
            definitions["rtpath.resolutionreduction.step"] = 1

It should be this one.

Yes, it is possible and not even that complex but, as usually happen, it is a matter of finding time, establishing priorities, etc.

Rendered twice with latest build of 2.5.

First render set system to OptiX and add on to CUDA. Rendered in 38:21

Second render set system to CUDA and add on to CUDA with OptiX unchecked. Rendered in 38.04


You have rendered both times with CUDA+Optix, the current BlendLuxCore preference ...

Sharlybg wrote: Thu Sep 03, 2020 9:08 pm Is possible to do inside luxcore code 2x2 during input and navigation and switch back to 8x1 automatically right after ?

It is but it is a parameter BlendLuxCore doesn't expose (the number of samples it will take the transition from block to pixel rendering).

I mean Dade even if option B is adding less pressure after first samples compared to A. The fluidity during first sample in option B is simply from another level and this almost no matter how high you set block size in option A.

So for me there is a weird thing we can improve here.


You are ...

AndreasResch wrote: Thu Sep 03, 2020 1:44 pm This might be helpful in some cases, but would be hard to do when you have a lot of objects and you want to randomize them all.

By default, the object ID is random, you can overwrite it to have a fixed value.

This would be great. You might also consider adding this to triplanar mapping as well where randomization makes a lot of sense. In those wooden pieces I actually used triplanar mapping. All the sides are randomly rotated and shifted to get the variation which simulated the randomness of production ...

The settings should be correlated with the view port size. You can not have a single setting responsive for any view port size.

On nvidia‘s graphic transfer rate from RAM to VRAM is 14 GB/s. This is 32 times less than VRAM to GPU cache.


The slide is about CPU/SSD/etc. (any device attached to the PCIe bus) to GPU ram transfer rate. It can have a huge impact on some application.

In our case, it could highly increase the ...

Emmm, in addition, is the wo and wi to determine whether to reflect or transmit in evaluate funtion?


Glass is a specular material (i.e. a delta one) so GlassMaterial::Evaluate() always return "Black": https://github.com/LuxCoreRender/LuxCore/blob/2863718772213dd938913da4fff37b0c0ffaf49f/src/slg ...