Why IrMap over Brute force for animation rendering?

[sharktacos]

Not necessarily, why do you say that?


Yes, but it's not done daftly, and most of the potential issues are taken care of by Leak Prevention and retrace


Likely because now you're shooting a lot more, smaller LC samples, and the leak prevention goes into overdrive.

It'd be good if you could show what issue you're having which forces you to raise LC subdivs for a still when resolution changes, as it should really be resolution independent by design.

Lele,

When speaking of increasing the light cache subdivs based image size, I was referring to the formula you gave earlier in this thread (post #76):
https://forums.chaosgroup.com/forum/...403#post834403

"1 x 1k LC subdivs (1k) -> Retrace of 2.0^1 (2.0)
2 x 1k LC Subdivs (2k) -> Retrace of 2.0^2 (4.0)
3 x 1k LC Subdivs (3k) -> Retrace of 2.0^3 (8.0)
4 x 1k LC Subdivs (4k) -> Retrace of 2.0^4 (16.0)"

In that formula, the retrace and subdiv values both increase with the image resolution. So that was the starting point for my question. Increasing the subdivs makes sense to me logically, but was not quite sure what the logic was of increasing the Retrace value, and so was hoping you could give some insight into that part of your formula.


FWIW, I'm a VFX dude. So I never render stills, only animations.

[^Lele^]

Ah, no, i never intended image resolution in that post.

It was a simple table to indicate how to increase retrace threshold based on number of LC subdivs, regardless of why one should have increased them.

In fact, I was underlining the fact that for most scenes (particularly the most intricate ones) the LC subdivs will work perfectly fine at their defaults, thanks to Leak Prevention, in charge of oversampling the problematic geometric areas.
This, in contrast to pre-LP, where raising LC subdivs (even for a still, but much more so for a sequence) was the only way to try and have samples reach the hard, difficult places.

I have not done recent tests on what oversampling the LC at the growing of resolution does, surely it won't do the render any harm, besides taking a few seconds longer, but I am not sure it'll make a huge difference in quality on the individual frame (while it's a sure method to increase temporal coherence in animations).

[tricky]

Just read through this thread again as a refresher. Some great stuff, but my brain has melted out of my ear

[^Lele^]

Just read through this thread again as a refresher. Some great stuff, but my brain has melted out of my ear

Hey, if the memory of it helps when in doubts if using "Defaults" or not, then i'm happy.
If not, sorry, here's a bowl for collection.

[coolhand78]

Just a quick question - when using Brute Force (primary) and Light Cache (secondary) is the number of bounces for the light cache controlled by the "bounces" setting in the "Expert" Light Cache roll-out...?

also, is there any more info on how "leak Prevention" works? not having issues just wanting to know...

[^Lele^]

Just a quick question - when using Brute Force (primary) and Light Cache (secondary) is the number of bounces for the light cache controlled by the "bounces" setting in the "Expert" Light Cache roll-out...?"

Yes, which also implicitly controls the number of bounces of the BF retracing (as those have to match the ones from the LC).

also, is there any more info on how "leak Prevention" works? not having issues just wanting to know...

vlado will know.