E-Cycles Pro 2021

Thanks to code optimizations, rendering with NVidia GPUs and E-Cycles is equivalent to doubling the number of GPUs in your machine, on average.

E-Cycles AI-Denoiser requires between 4 and 10 times less samples than other AI-based denoisers to get a similar level of quality.

Using less samples for the same quality means a proportional speed increase for your renders – that’s on top of the ‘regular’ speed increase E-Cycles offers on NVidia cards. For comparable image quality, with both the regular speed increase and the new AI-Denoiser, you can get up to 25x faster rendering. And it's a one click setup!

Switching between real time render engines and path tracers is time consuming and prone to errors. For producing high quality fly-through animations, E-Cycles can match the render speed of the most common real time render engines, while offering better quality and a faster scene setup.

When you tweak lights in the viewport, it will restart at 0 samples by default each time you change a parameter. But when using E-Cycles and light groups, you can tweak your lighting setup while the viewport continues to render, offering you very fast, high quality previews.

When you tweak lights in the viewport, it will restart at 0 samples by default each time you change a parameter. But when using E-Cycles and light groups, you can tweak your lighting setup while the viewport continues to render, offering you very fast, high quality previews.

For even faster rendering and fast previews of your geometry and lighting, E-Cycles offers 2 new types of rendering kernels: AO and Clay. Both presets will also automatically setup material overrides for you and then switch to those kernels, for up to 4x faster rendering – on top of the 2x base speed increase – for up to 8x faster rendering.

Viewport denoising is awesome, but there are times when it washes out the fine details in textures, hair and dark areas. With SSAA, E-Cycles preserves more details and renders faster at a comparable sample count.

Comparison of viewport outputs in E-Cycles with SSAA off and on after 10 seconds of rendering.

E-Cycles is regularly updated:

On top of the regular updates, here is the changelog from January 2019 until now:

February 2019 update

• New Dithered sobol for cleaner renders at low samples.
• up to 5% faster rendering in some scenes.

March 2019 Update:

• Cleaner Viewport renders in complex scenes.

April 2019 Update:

• Intel Open Image Denoise Patch from Stefan Werner is included.
• Latest Filmic log encoding for top-notch post-pro as recommended by Bartek Skorupa https://www.youtube.com/watch?v=kVKnhJN-BrQ.
• New option based on the scrambling distance patch from Lukas Stockner, allows to render up to 40% faster (on top of the existing speedup).
• Uses half the memory when denoising with all passes (transmission and SSS included).
• New AI denoiser addon which can now generate denoising node trees with one click, even in complex scenes, and at very high quality.

May 2019 Update:

• E-Cycles 2.8x got 2 improvements ported from the 2.79x branch. Memory usage is reduced in most scenes, allowing bigger scenes to fit in the GPU memory. Rendering is also around 10% faster compared to the April update.
• Both the 2.8x and 2.79x versions can render up to 2x faster when using the old denoiser (the one from Blender) on multi-GPU setups.
• Updated to the latest Open Image Denoise, using 38% less memory to denoise.

June 2019 Update:

• E-Cycles will automatically choose the most memory efficient code path depending on the scene, using up to 40% less memory than master/buildbots (BMW scene on a 2080Ti goes from 2GB on buildbots to 1.2GB memory usage in E-Cycles).
• An option to force a memory efficient option has also been added, making transparent shadows optional. Render times can be greatly reduced in some scenes.

July 2019 Update:

• The AI denoiser tree generation has been improved and now has a better auto layout.
• Adaptive sampling is now available for CUDA and can bring good speed increases in some scenes. My own test with smoke+fire shows up to 2.5x faster rendering compared to the June update. (Although only CUDA is officially supported, OpenCL users also report a noticeable speed-up)

August 2019 Update:

• Now based on the official 2.80 version.
• AI denoiser updated to 1.0, allowing you to denoise renders of any size with low memory usage.
• New buttons for a quick background command line render. You can then close the UI, which gives better stability, makes more memory available for your scene, and also speeds up rendering in many cases.
• Baking render speed was optimised in some scenarios.

September 2019 update:

Completely reworked AI denoiser:

• faster by default, particularly on multi-GPU setups
• much better memory management thanks to a patch by Stefan Werner
• streamlined denoiser UI
• automatic optimizations of the denoising tree. Improves denoising and rendering speed.

Faster baking:

• Baking is now up to 20x faster by taking smart decisions for optimal performance.
• Baking will also require a little less memory.

October 2019 update:

• Added a remove double modifier.
• Added an experimental bridge modifier.
• Added bevel and crease weight to the Boolean modifier for non-destructive and easy to use bevels after Boolean.
• Added light groups patch from Lukas Stockner (experimental).

November 2019 update:

• out of core Rendering for RTX (speed limited by system RAM, may require Quadros or Teslas for good performance)
• much faster rendering for RTX
• further optimizations for CUDA

December 2019 update:

• Pre-processing optimizations
• Update stable build to 2.81a
• Adds 2.82 builds

January 2020 update:

• Viewport denoising for the RTX version (original patch from P.Mours @Nvidia)
• Added experimental support for GTX cards in the RTX builds, including viewport denoising.

February and March 2020 update:

• Very fast volume rendering
• OpenVDB import preview builds
• Update stable build to 2.82a

April 2020 update:

• Faster viewport rendering
• AO and Clay kernel for very fast previews
• NVLink support to stack memory on multi-GPU configurations. Allows you to almost double the size of your scenes, while still being much faster than with system memory.
• New presets with improved performance and quality
• Improved UI
• easy switch between CUDA and OptiX

May 2020 update:

• Slightly improved performance
• Better Clay Preset

June 2020 update:

• Faster Denoising
• Updated OIDN for higher quality denoising
• New splash screens and logo to easily identify the version you’re working with.
• early access to the new sky model.

July 2020 update:

• New tutorials for up to 41x faster rendering, higher quality rendering, general overview of workflows and functionality, tips and tricks, etc.
• Improved denoising quality when using the slow pass mode.

August 2020 update:

• New Light Groups add-on (included for free)

September 2020 update:

• Improved Light Groups add-on. You can now easily change the tint of your lights and apply both the strength and tint to the lights themselves for final rendering.
• Improved integration of the Light Groups Add-on with the AI-denoiser.

October 2020 update:

• Light Groups can now be tweaked directly in the viewport. Previously, tweaking a light would reset the sample count to 1 and viewport resolution to 64x64. E-Cycles 2.91 now offers the possibility to create the perfect light setup really quickly thanks to a high sample count, full resolution and real time updates!
• Persistent Data option: this option will export the scene only once per animation. Especially useful for fly-through and turntable animations of complex scenes.

November update:

• High quality denoising mode for the E-Cycles AI-Denoiser
• CUDA support on E-Cycles for Mac 2.91
• even faster viewport rendering
• enhanced clay and AO rendering with reflections and refractions supported

December update

• SSAA in the viewport for high quality details preservation using denoising

What the artists say

"E-Cycles lets me render & proof huge print ready assets locally on 3 gpus – no renderfarm required!" – Paul, CG Artist

Made with E-cycles

Render by Petra Mikulicic https://petramikulicic.artstation.com/

Render by Marco Iozzi https://www.marcoiozzi.com/

Render by Marco Iozzi https://www.marcoiozzi.com/

Render by Torsten Dudai https://dudai.artstation.com/

Render by Hyesung

Nebulae by Gleb Alexandrov

"Cookies & Milk", original by Pixar and ported to E-Cycles by Matthieu Barbié

Render by Rashed Abdullah https://www.artstation.com/artwork/VdxDxZ

Bedroom by Luan3D

Dark Interior by Entity Designer https://blendermarket.com/products/dark-interior-scene

Render by Johannes Wilde

Render by Planungsdetail

Render by Albert Shamseev

Render by Entity Designer

Render by Piero Desopo

Render by Luan3D

Render by Luan3D

Render by Marco Iozzi https://www.marcoiozzi.com/

Render by Marco Iozzi https://www.marcoiozzi.com/

Render by Petra Mikulicic https://petramikulicic.artstation.com/

Render by Petra Mikulicic https://petramikulicic.artstation.com/

Render by Petra Mikulicic https://petramikulicic.artstation.com/

Render by Torsten Dudai https://dudai.artstation.com/

Render by Helmy Ardiansyah - Studio HV

Animations by Ryan Plunkett at Premier Lift Group. More renders are available on their website, all made with E-Cycles.

What about OpenCL and CPU rendering?

Only NVidia GPUs are officially supported.

For AMD GPU users, OpenCL is currently only around 20% faster, and is still a work in progress. CPU rendering is the same as in the official Blender release. However, you can still benefit from the new AI denoiser, render profiles, dithered sobol and more for up to 10x faster rendering on all platforms and with similar quality. Features like Light Groups, clay and AO rendering etc. also work on CPU and OpenCL and can massively speed up your workflow.

If you want to use E-Cycles with CPU or OpenCL, I’ve collected some user tips in the documentation to get the best experience possible.

Pre- and post-processing:

E-Cycles is on average like adding another GPU to your configuration. And just like adding another GPU, E-Cycles will make the rendering phase faster. Pre-and post-processing are done on the CPU in Blender, so these will be just as fast as before in E-Cycles.

Known limitations present in Blender and E-Cycles:

• If you use a gamer GPU, scenes have to fit into the GPU memory for good performance. The NVidia drivers only allows fast system memory access on Tesla and Quadro cards.
• CPU+GPU works, but generally won't give any benefit. Today's RTX GPUs are too fast to be significantly helped by even the best CPUs.
• Multi-GPU configurations and NLM denoising: there is a bug in Blender 2.8x and currently 2.9x when using multiple GPUs with CUDA and the old denoiser (or its passes). This can slow down rendering. Solution in E-Cycles: using E-Cycles AI-Denoiser
• For optimal performance and integration quality, E-Cycles like Cycles is included in a full version of Blender. This will be the case until there is a suitable render Engine API to allow the integration of an external renderer with optimal performance.

E-Cycles-specific limitations:

• Support is only for the E-Cycles parts. Bugs also found in Blender builds are to be reported to the Blender Foundation on developer.blender.org
• The overall speed-up on low sample renders (under 64 samples per pixel) are usually low. Scene export, compositing and file saving are most of the time the bottleneck here. Using E-Cycles AI-Denoiser and lowering the sample count even further can help to speed-up renders in this case.
• The out-of-the-box speed-up is mainly for Path Tracing. Branched Path Tracing may also benefit from a speed-up, but this may vary depending on the scene.

System requirements:

Same as Blender plus:

• Any NVidia GPU supported by Blender with driver v456+. (Although only NVidia GPUs are officially supported, many users now use E-Cycles on CPU or AMD GPUs. I added some user tips in the documentation about this.)
• A 64bit version of Windows, MacOS** or Linux supported by Blender.

*

**