Over 60k subdivisions for PPT and SSS

[anchovy]

Vlado-

I know we had chatted about this awhile back but Im wondering if the 60k limit for subdivisions for PPT could be raised or if its a limit imposed by the way the sampler is set up.

Also Is there any way to get SSS to work with PPT since there is no pre-processing the way its currently set up. Perhaps there is a current workaround but I may have missed it.

Thanks!

-Michael

[vlado]

Currently the work-around for sss is to calculate the illumination map, save it to disk, and then reuse it, but this may be not quite what you are looking for.

The 60k limit is not caused by the sampler; just some 32-bit values will overflow. It is not too hard to work around that, I hope that we do this for the next release.

Best regards,
Vlado

[anchovy]

Yep it would be great to see a bit more room on the 60k limit but its not pressing. Yeah the SSS seems like a bear as id like to use it for the bidirectional Path Trace. out of curiosity as I haven't found it in what if read thus far, how are some of the renderers out there calculating sss in an unbiased fashion and doing it without a precompute/cache?

-Michael

[vlado]

how are some of the renderers out there calculating sss in an unbiased fashion and doing it without a precompute/cache?

Slowly

Seriously though, it is certainly possible to calculate the sss effect in an unbiased way and this is what we are doing for the RT engine.

Best regards,
Vlado

[anchovy]

Cool to know, I figured it had possible. Perhaps youll grab what you are doing on the rt side and be able to use it for the path tracer.

Thanks

-Michael