E-Cycles - up to 2.5x faster rendering for Blender 2.79 to 2.82

If your are interest in learning and understanding what E-Cycles is doing too improve performance in Cycles and build your own Blender version of E-Cycles.

Here is my review of bliblubli course see attach link Create your own Blender.

And if you want to both learn the code and have builds support, you take the course and get E-Cycles for half the price, just PM me :slight_smile:

so i gave it a try
2080ti in sli comparison before and after e-cycles

2.8 Windows 10
2000 samples 2000x1500px

ORIGINAL CYCLES 64x64 ------> 15min 43sec
E-CYCLES +AUTOTILE ----------->11min 03sec

= 28.6% faster = 1,28x faster

also the render is not exactly the same ? just a different noise seed ?

it feels quite low compare to the massive number that the 1080ti got with e-cycles tho
i still need to do some tests to see if sli is maybe not as fast as it should be, maybe the scaling is bad idk

+30% is amazing dont get me wrong but i was hoping for a good +50% with this card like the other did :cry:

i will try the evermotion scene tomorrow, same samples ? same resolutions ? If someone is interested in comparing perf i can share my scène in PM

thanks for the comparison. SLI should be disabled for good performance in cycles. First try with a single card to compare E-Cycles and Cycles to ensure SLI or anything else is not triggering a slowdown.
The noise is indeed most of the time lower (at least at the same level), it’s another benefit of E-Cycles and it will be even lower at same spp count in the next version.
If you send me the file per PM, I can have a look to it :slight_smile:

If you have more than one graphic card, dont join them in Sli because it is slower than rendering with them apart from each other. it is known for long time now

1 Like

GREEN RESTAURANT BENCHES if you want it i can share the file in PM (win10 on 2.8 only)

ORIGINAL RENDER BELOW


E CYCLES RENDER BELOW





Cycles 1x 2080ti (64x64) ------------------------------> 30.48 min
E-Cycl 1x 2080ti (Auto-tiles)--------------------------> 19.30 min = 1.57x Faster than master


Cycles 1x 1080ti (64x64) ------------------------------> 40.03 min
Cycles 1x 1080ti vs 1x 2080ti > = 2080ti is 1,3x Faster
E-Cycl 1x 1080ti (Auto-tiles)--------------------------> 25.34 min = 1.57x Faster than master
E-Cycl 1080ti vs 2080ti > = 1080ti in E-cycle is 1,2x Faster than a 2080ti in master



Cycles 2x 2080ti (64x64) SLI ENABLED ------------> 15.43 min
Cycles Scaling compared to a Solo card > = 1.97x Faster

E-Cycl 2x 2080ti (64x64) SLI ENABLED ------------> 11.03 min = 1.39x Faster
E-Cyc Scaling compared to a Solo card > = 1.74x Faster

this mean that i could got to 9.82 if the scaling was the same with two cards
:angel: please ?



SLI DISABLED OR NOT, IS THERE AN IMPACT ON THE RTX CARDS ?

Cycles 2x 2080ti (64x64) SLI DISABLED ---------> 15.30 = 1.5% faster
E-Cycl 2x 2080ti (64x64) SLI DISABLED ----------> 11.00 = 0.3% faster

as you can see, with the rtx series sli disabled or not the changes are not representative
Not with the RTX series, they use a new NV-link and this rules don’t apply with thoses cards visibly
https://www.chaosgroup.com/blog/profiling-the-nvidia-rtx-cards https://www.pugetsystems.com/labs/articles/NVLink-on-NVIDIA-GeForce-RTX-2080-2080-Ti-in-Windows-10-1253/
this new Nvlink connector can also share vram, so if the devs optimize the set up i could have up to 22Gb of vram…



HAVING A WINDOWS MINIMIZED COULD AFFECT RENDER ?

E-cycl 2x 2080ti + WINDOWS MINIMIZED = 10.15 = 1.07x Faster than an active windows
so yes



IMPACT OF A COMMANDLINE RENDER ?

i got 10.37 with a command line render, still faster
also the VRAM went from 4500Mo to 3200Mo so the UI take 1.3Go of VRAM

4 Likes

Thanks for the scene and the tests. I’ll have a look asap. Can you also try to render from command line. Your config is very fast, so a lot of thing like the UI can become an overhead that reduce the visible improvement. To render from the command line:
c:\path\to\blender.exe -b c:\path\file.blend -f 1
Or simply minimize the window when rendering.

1 Like

So on a 1080Ti, the scene renders in 25min34s here, just pressing F12. So it seems your single card setup works properly. But there is something with the dual GPU setup. I’ll make a new build tomorrow which should make rendering faster in your case. In the mean time, would be awesome if you could make a command line render using the both GPU, with the command line also minimized.

600e gpu faster than a 1200e gpu, how amazing is this ! :slight_smile: (well i hope this rtx technology will get my perfs and money back… )
what about the render witouth e-cycles ? it will took an hour ?

i got 10.15 min with blender minimized, e-cycles, sli disabled
and 10.37 with the batch render with everything minimized, noted 1Gb less of ram during the render, 3232Mo instead of 4500Mo

no other apps running of course

ok, thanks so it’s 930/615=1,51x faster with 2 GPU and E-Cycles, that’s what I thought, the UI takes a lot of memory and also makes rendering slower in 2.8. It’s more noticeable for you as you have a configuration that does the path tracing extremely fast, so the UI is updated more often.

Interesting info about the UI, i’d totally forgotten that a couple years ago i’d read that maximising/Ctrl+Space the render window could help speed up renders too. i’ll give that a try when i’ve got some time. it’s a shame there’s not some built-in code to disable/minimise/whatever the UI when you hit F12 if it’s a known issue.

Well, it’s fast to minimize the window manually if you don’t need to see the image updates.
@BD3D with master, it takes 40min05s, so E-Cycles is “only” 1,6x faster in the scene indeed. With 2x 2080Ti, it’s still a boost worth 2400*0,6 = 1440€, so pretty good already :wink: I’m working on making it even more. In the meantime, I would be happy to see the render time for the BMW scene at 100% resolution (it’s 50% in the official version) compared to vanilla Cycles, with both window minimized to be fair.

1 Like

Sneak peek of march update:


21min18s on a single 1080Ti, so 1,88x faster than master.


2min14s for BMW at 100%, so a 600€ card is nearly as fast as a 5K configuration with 2x 2080Ti
34sec at 50% = 2,87 faster than master :slight_smile:

5 Likes

BWM BENCHMARK 2080ti vs 1080ti at 100% RECAP

Cycles 1x 2080ti ------> 3.20 min
Cycles 2x 2080ti ------> 1.44 min
Scaling of 1x vs 2x Cards = 1.97x


E-Cycl 1x 2080ti -----> 2.00 min = 1.666x Faster
E-Cycl 2x 2080ti -----> 1.04 min = 1.625x Faster
Scaling of 1x vs 2x Cards = 1.87x


Cycles 1x 1080ti -----> Need someone to give me the stat
E-Cycl 1x 1080ti -----> Need to compare with the old E-cycles result before the update of march


But i think its stupid to compare on such a short benchmark, anything that have a little influence could have a big impact on the result, its not a good way to compare two builds/cards

There is something wrong going on it seems.
Another user with 2x 2080Ti there render it in 1:54, you render it in 1:44, which means 13% faster using master. Did you overclock your cards? It could lead to overheating -> throttling thus the smaller speedup with E-Cycles. The RTX 2070 render also 2,02x faster on BMW scene using E-Cycles, so I don’t think it can come from the RTX architecture. Maybe you should check with GPU-Z if the card is working properly (100% usage and 100% frequency).

So the resolution indeed has an impact on the performance. At 100%, master renders in 5min07 while E-Cycles renders in 2min45 (single 1080Ti as usual, using the 20190125 build). It means it’s only 1,86x faster at 100% (compared to 2,42x at 50%). So their is room for even more speedup their too.
By the way, I find your summaries very helpful :slight_smile:

How does it look when animating? Are there any artifacts?

1 Like

i have a small Factory OC
i scout constantly the gpu/cpu activity and its always on 99% drop to 97% sometimes but its minimal
in therms of thermals it never go above 81* degrees.so i doublt that its throttling. i can do some tests if you want but i dont know how to remove my factory OC?

With MSI afterburner, you can, but indeed, 81° doesn’t look like it should be throttling. My 1080Ti only start throttling at 84° by default. However, I do all my test with a down clock of 5% and with -0,08v. It keeps the card at about 60°. It’s very quiet and ensures it never throttle to make timings reproducible.
@BigBlend at least in walk-through it works artifact-free, I’m rendering a video right now.

Edit: here 2 videos of Classroom:
with default settings, 50%, 14sec per frame, no artifact

Just for fun, with Denoising, (current official denoising, not animation version which is in work, so with artifacts but not due to E-Cycles and actually more subtle than I thought), 17sec per frame.

So the short answer: no artifacts in E-Cycles for walk-through.

2 Likes

just a comparison with another rtx benchmark just to be sure
RTX 2070 bwm 50% Master -----> 91 sec
RTX 2070 bmw 50% E-Cycl -----> 45 sec = 2,02x Faster than master

RTX 2080ti bwm 50% Master ----> 58 sec
= RTX 2080ti is 1,56x Faster than a 2070
RTX 2080ti bmw 50% E-Cycl -----> 34 sec = 1,7x Faster than master

so e-cycle build is 20% more efficient with a 2070 compare to my card :cry:
will future e-cycles build be more optimized for 2080ti’s and multi cards set up ?