ERPT chain length

[jeanphi]

Hi,

I've pushed a few improvements to ERPT, it should now be about 10 to 20% faster and give slightly less strange results, but it is still far worse than metropolis.
Regarding caustics, you actually need to let it run much longer than metropolis for it to get a chance to access caustic paths and resolve them.

Jeanphi

[Meelis]

Hi,

I've pushed a few improvements to ERPT, it should now be about 10 to 20% faster and give slightly less strange results, but it is still far worse than metropolis.
Regarding caustics, you actually need to let it run much longer than metropolis for it to get a chance to access caustic paths and resolve them.

Jeanphi

Hi

Thank you

[CoolColJ]

After some experimentation, I finally found some good settings for ERPT

ERPT has always puzzled me, but today I got a good feel for it.

Tried these -

Bidrectional
32 depth all round
Random Sampler
Hilbert pixel sampler
2 samples per pixel

I tried chain length 8 and 16 so far for 15 mins each - the lower number means caustics don't develop as quickly, but it seems to render bigger chunks of the image with Hilbert on a semi-tiling basis when using a 3 sec screen update.
So it covers the whole render and develops quite quickly. Reaches higher samples per pixel much quicker.
First pass is sorta darker, and gets brighter every pass in a bucket rendering fashion.

If you use Vegas or Low discrepency pixel samplers it looks like a mess!
Tile and linear is not bad, it goes down in a scanline fashion, but is slower than Hilbert

I'll try some higher chain numbers and report back. When I used to text edit my scene files I have gone as high as 1 million

I'll post some pics of my tests in my render tests thread soon.

I posted my test pics here
viewtopic.php?f=14&t=3885&p=81925#p81925

[CoolColJ]

Hi,

I've pushed a few improvements to ERPT, it should now be about 10 to 20% faster and give slightly less strange results, but it is still far worse than metropolis.
Regarding caustics, you actually need to let it run much longer than metropolis for it to get a chance to access caustic paths and resolve them.

Jeanphi

Thanks

Where can I download this?

[SATtva]

From the repo. If you need a binary, you'll have to wait till the next weeklies or RC.

[Meelis]

Hi

Is it ok when chainlength value = eyedepth/ lightdepth value?
Or must the chainlength value be allways higher?

chainlength 256
eyedepth 256
lightdepth 256

[CoolColJ]

Hi

Is it ok when chainlength value = eyedepth/ lightdepth value?
Or must the chainlength value be allways higher?

chainlength 256
eyedepth 256
lightdepth 256

doesn't matter, but having such high depth values will make it slow

[Meelis]

doesn't matter, but having such high depth values will make it slow

Oo i see, ok thanks.

[CoolColJ]

Use higher chain length values though - better caustics
1024 is good

Eye and depth values of 128 is more than enough, although gems might need higher...