[Bf-cycles] Optimization project
d4vidfenner at gmail.com
Wed Feb 5 19:12:16 CET 2014
Hi all. My name is David Fenner, I'm 3d director at Loica (www.loica.tv).
Regarding Cycles optimizations, we recently did a very (VERY) heavy scene
in cycles that made us suffer a little more than we expected (about 1 hour
per frame on a GTX TITAN, with minimal bounces and simplified shaders). I
wrote down a few things regarding heavy scene production in cycles that I
belive would help a lot. The scene is a jungle from the inside... it
doesn't really get any more complex than that, I'll share it as soon as my
boss lets me, for now I'd like to share the ideas/comments, that probably
are similar problems than the ones that the gooseberry team will be facing:
1) Perfect frame time estimation: Right now time estimation has an error
margin way too big to the point it is useless. For heavy scene rendering
and tight deadlines a better prediction is mandatory. A great way to do
this would be to make, in f12 rendering, a first pass of "progressive
refine" (let's say 20 samples per tile), then calculate the time based on
samples left, and then continue on each tile until full sampled, with no
more "progressive refine". The problem right now is that time estimation is
done by calculating the tiles being used only, but error comes when some
tiles take 10 secs on some others take 10 min. If we did a time estimation
on all the tiles based on 20 samples then there is no margin for error,
since the next 20 samples will take the same time on each tile, and so on.
Having perfect prediction (thanks to the nature of path tracing) is a
blessing for high end production.
2) The glossy shader could have a button/ticket box to make it "only
direct". Basically this would make the glossy shader react only to direct
light and hdri. For many, many types of shaders, this specular-like usage
of the glossy shader is more than enough, and probably it would save a lot
of bounces. For example, I wanted to do only specular to the leaves (more
than enough, no reflection needed), but I couldn't without lowering all the
glossy samples, therefore killing the reflection in the river (the one
reflection that I did need).
3) Currently, hair particles seem to be the only way to distribute objects
through a surface in a procedural manner (like c4d cloner or 3dsmax
scatter). This is what I used for grass (a few modeled planes distributed,
was faster than hair and better looking), but it seemed that the more I
increased the quantity, the more memory was used. Aren't this supposed to
be instances? As far as I know, when you use and object instead of hair it
is only position, scale and rotation are considered, so I don't see why
they couldn't be instances.
4) Dealing with transparency for custom render passes (object ids, custom
light for compositing, extra character ghost, whatever) is currently very
very hard. Basically you can't get a grey geometry to make a custom light
pass without killing transparency settings (and in the future displacement)
with the material override. Could it be possible that renderlayer material
override respected the last transparency shader of the original material
tree? as well as the displacement? This way you could get custom passes but
keeping the shape/transparency of your render. Currently all sort of tricks
need to be done, like making a giant shader that has many transparency
shaders mixed by custom attributes like UV, vertex color, object ID, stuff
like that. Bad to set up and memory intensive.
5) With the setting above, maybe it could be easier to do an extra render
pass for Normal and vector, like a separate, 30 sample render? This way
some complexity could be taken a way for final (heavy) scene render, by
taking out AO pass, normal, vector, mist, object id, etc. And make an
override for another, less complex and less sampled render that respects
transparency and displacement, that gives antialiased normal and vector,
mist, AO, anti aliased object id, etc. I know two pass isn't ideal, but is
a very descent workaround and could be part of the same render (just with a
"AOV" stage). On the other hand, GPU's really went down to their knees on
this jungle render... to the point that adding an AO pass was simple
impossible. Having it separate could ease a little the burden for GPU's
that clearly don't do as well as in simple scenes. (In fact, TITAN is
usually about 3 to 5 times faster than our 12 core xeon cpus, but on this
jungle scene it was about 1.6 times faster only).
6) The mist pass has artifacts when transparency limit is hit. If you have
many leaves and a top of, for example, 7 transparency levels, if the limit
is hit in one leave this leave will be seen white on the mist pass.
7) I think this is quite obvious, but I'll point it out anyway: Normal and
vector pass are a necessity for compositing but are currently useless (no
anti-aliasing, doesn't take transparency into account).
That's about it... I hope you guys find some points useful and good luck
with the project. I'll upload the jungle scene as benchmark too as soon as
I am able too.
2014-02-05 Brecht Van Lommel <brechtvanlommel at pandora.be>:
> Hi all,
> With Gooseberry coming up, it would be good to start focusing on (CPU)
> optimization for the next few release cycles. For new features I still
> plan to finish volume rendering and add deformation motion blur soon,
> but besides that a more organized optimization project is something
> I'd like to spend time on.
> See this wiki page for more details:
> There's already great work being done by Sv. Lockal and others in this
> area, on the low level optimization front, but there's many more
> opportunities. I'll try to add more detail and ideas over time, this
> is just what I could think of off the top of my head.
> You can help out in various ways:
> * Contribute code to implement optimizations
> * Help gathering of test scenes representative of Gooseberry shots
> * Suggest practical optimizations ideas
> It would also be cool if someone could set up some webpage with a
> graph that tracks Cycles performance on test scenes over time, maybe
> doing a render with the latest Git version once a week or so (like
> Bf-cycles mailing list
> Bf-cycles at blender.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Bf-cycles