RENDER ZONE

http://www.ontmoeting.nl/renderzone/

homepage

 

Comparisons render Engines


Comparison of render programs and their endproduct still pictures, animation or film is very (very) hard to do it right, as it should. Herewith some contributions that are worth mentioning, if you are planning a comparison test yourself.

Always read comparisions with greatest respect for the work of the makers, but beware of the lack of the even the smallest piece of information, that can make the balance to wobble.

What is Furryball RT

A very fast GPU renderer with photorealistic results and good features like CPU render Engines. For animation, film, product and other render subjects.

 

 

FurryBall RT

With GPU development of NVIDIA, many companies are now leveraging this compute power to greatly accelerate their solutions and define new interactive possibilities.

FurryBall is a unique, real-time, GPU production-quality, final-frame renderer with advanced rendering techniques. It can be implemented directly into Maya and 3ds Max with network rendering and multi-GPU support. FurryBall 4 features NVIDIA OptiX‚Ñ¢ physically based, full global illumination, and supports biased techniques. It combines the speed of GPU rendering with the quality and features of CPU rendering to deliver proven results in real movie and game production by multiple studios.

 

Furryball RT

GPU Renderer

The old comparison battle between CPU versus GPU

http://furryball.aaa-studio.eu/aboutFurryBall/compare.html

GPU vs CPU - GPU rendering vs Software CPU rendering

FurryBall vs CPU render - FurryBall vs Vray, FurryBall vs Arnold, FurryBall vs Mental Ray, FurryBall vs IRay

Scene 1 - up to 6x faster

Unbiased rendering and Physically based full global illumination, brute force, with 4x light bounces and Path tracing if available

The most similar render for compare the speed and quality is CPU render Arnold, because it use also Path tracing.

The scene is very complicated for ray tracing and all scene is lit by indirect (4x bounced) light.

In FurryBall setup there are 400 rays for 14 min render and 200 rays for 6:30 render with our Noice reduction filter.

Arnold is doing 41 minutes with comparable results

 

Comparable other renderders

 

Render Engine

minutes

FurryBall (GPU)

14

Mental Ray 3ds Max

44

V-Ray 2.0

47

Maxwell Render

46

Arnold

41

With no realtime Viewport in Mental Ray, Maxwell and Arnold.

Realtime Viewport in FurryBall (GPU), VRay 2.0.

As you can see in the pictures Maxwell Render does incorporate to much noise at 46 min, to get a better result it has to render more then one and a half hour or more. Vray shows also noise in the rendering, but very mild and not disturbing, the middel ball dit not render very well, something like to much of motion blur. Even at a renderingtime of 6:30 min for FurryBall there is some blur, but the noise has cancelled out quite respectable with the use of noise reduction filter.

Workstation specs:

Intel Core i7-930 2,80 GHz, 16 GB RAM, Nvidia Geforce GTX Titan - 6 GB Memory

Because every renderer has a totally different setting, there is no possible to have the same settings. We tried to set the optically similar quality for each renderer. We set brute force in all renderer (without any tweaking).


Scene 2 - 14x faster

Biased rendering

Simple cartoon like scene without reflections. (Two lights, raytraced shadows)

 

Render Engine

minutes

FurryBall (GPU)

: 20

Mental Ray 3ds Max

4:41

V-Ray 2.0

47

Maxwell Render

-

Arnold

-

 


 

CYCLES

Versus the giants of rendering

in the Blender GURU

August 5, 2015 by Dario Baldi

http://www.blenderguru.com/articles/
render-engine-comparison-cycles-vs-giants/

 

This is a very good comparison test (lasting 3 months), knowing that every comparison of render programs is hard to do it right and very easy to do it completely wrong. With the results changing from fair, good to not applicable in any way.

Part of the text is repeated here, for the full text and pictures and statements see the link above.

Orignal text with some compression, revision and rewriting.

 

Render Engine Comparison: Cycles vs The Rest

Cycles has come a long way since we first saw that super cool train demo from Brecht back in 2011, but how does it compare to other path tracing renderers available today?

Well in this article, I (Bladi) put cycles to the test against 5 other render engines that can be used with Blender:

* LuxRender

* Maxwell

* Mitsuba

* Octane

* V-Ray

 

Naturally, it's not easy to compare all the renderers fairly, so here's how I did it:

Testing Process

All the rendering was done using CPU rendering, on an Intel i7 3770 in a desktop with 12 GB of RAM and Windows 7. For Octane, a purely GPU renderer, an NVIDIA GTX 650 was used (*it performs very similarly to the Intel i7 3770).

I started by doing simple tests with cubes and monkeys, comparing materials and lighting between each renderer and finding out where the terminology differed.

Then I gathered all the information that I could from the official documentation, forums, YouTube videos, etc, and worked out which rendering algorithms from each engine would work the best in different situations.

Once I was confident enough, I started to adapt three scenes for the comparison:

1. a car render

2. an interior

3. a scene from a short film.

I jumped randomly between each renderer and slowly narrowed down which options were better suited for each scene. All in all, it was a bit more than 3 months of testing and believe me, it was fun.

Cycles and LuxRender
are both integrated using add-ons created by the developers of that renderer, which means they're generally reliable and well-documented.
 
Maxwell and Mitsuba
are also integrated using add-ons, but those add-ons are developed by people who did not work on the renderers themselves. This isn't necessarily a bad thing, but it might mean the documentation isn't up to scratch and getting support could be difficult.
 
Then Octane and V-Ray
actually have their own separate version of Blender, which can be a bit annoying to manage. It also means that there is a bit of a delay after official Blender releases, while the renderer developers tweak their code to be compatible with the new Blender before releasing an update.
Node-based materials and live Viewport rendering are all the rage these days, but they’re not completely necessary so it’s just something to keep in mind.


Round 1: Car Render

* It's light using an HDR image, which can be difficult to sample correctly and thus produce fireflies.

* Glossy shaders everywhere! That’s also a big cause of fireflies!

* There's a lot of intricate model detail, which could slow down the render a bit.

 

The rendertime results

 

Render Engine

minutes

Cyckes (GPU)

10:02

LuxRender

8:21

Maxwell Render

10:07

Misuba

7:27

Octane Render

7:35

V-Ray for Blender

7:45

Cycles
didn't do too well, but there are surprisingly few fireflies to be found.
Maxwell
was definitely the slowest and noisiest of them all, but the noise is slightly appealing in a way, like a natural film-grain effect.
LuxRender
got a fairly decent time, but it’s not quite as clean as Cycles if you look closely.
Mitsuba, Octane and V-Ray
all came very close in the end, and are all quite equally noise-free.

Recommended:

Mitsuba for being really fast and having great results directly out of the box.

 


Round 2: Interior Scene

Interior scenes are quite problematic to render since an important part of the lighting information comes from indirect light (bouncing off objects), which is a huge source of noise.

So to keep in mind for this scene:

* Light with a single sun lamp and 2 big area lights (one on the ceiling and one in the window)

* All of the fill-light comes from indirect bounces.

* Many image textures.

* Some fur on the carpet.


Round 2

 

Render Engine

minutes

Cyckes (GPU)

10:02

LuxRender

8:51

Maxwell Render

10:07

Misuba

7:27

Octane Render

7:35

V-Ray for Blender

7:45

Cycles
Although Cycles has technically the second slowest time, it's not far off from the average. And again, even for an interior scene, there are surprisingly few fireflies.
 

Maxwell

Again, Maxwell came last with the slowest time by far, but there’s something about that film-grain effect which makes it stand out.
 

Mitsuba, Octane and V-Ray

came close again, although it it does seem like they struggled with noise.
To address the elephant in the room, holy crap LuxRender! That new API is fast! It does look a little out of focus, but still!

Recommended:

Despite LuxRender's crazy speed, I'll have to recommend V-Ray for this one. It's much sharper than LuxRender and pretty quick too. It also gives some very appealing indirect illumination.


Round 3: Caminandes

This scene has quite a few things to keep in mind:

* Tons of hair!

* Real depth of field (not post-processed).

* Asset linking and instancing through particle systems (grass and rocks).


Round 3 Results

 

Render Engine

minutes

Cyckes (GPU)

26:01

LuxRender

32:21

Maxwell Render

21:21

Misuba

-

Octane Render

-

V-Ray for Blender

20:24

Cycles
was a little below-average on this one, but it was actually a lot cleaner than the other renderers.
 
LuxRender's
new API (which is what made it super fast in the last scene) doesn't actually support hair yet, so I had to use the old API and the result speaks for itself.
 

Maxwell Render

For a change, Maxwell did very well in this one, and in my opinion produced the nicest looking hair. Octane and V-Ray came close, but their hair looks pretty odd to me.
 
Mitsuba
But what a disappointment to find out that Mitsuba doesn't support hair in Blender! In cases like Caminades that's a deal breaker.

Recommended:

Definitely Maxwell. That hair is just plain sexy.

But if you'd like to have fast, clean results and tons of settings to optimize production rendering, V-Ray might be a better bet.

Conclusions

So at the end of all that, here are the average render times for each engine:

Average Render Times, from fast to slow

V-Ray
Octane
Mitsuba
Lux Render
Cycles
Maxwell Render

If you base it purely of speed, V-Ray seems like the obvious choice, but it's never really that simple.

Cycles

Peace of Mind - by Steve Lund

Cycles is definitely the youngest of all these renderers, but it's able to stand proudly among them.

Is there still a long way to go? Sure. But that's why projects like Cosmos Laundromat are so important.

I'd also dare to say that it' developing the fastest just take a look at the last few release notes and you'll see all the cool new features that get added for every version.

So lets take an objective look

Pros

* Fairly simple to learn (it's very interactive).

* Material nodes and viewport rendering are the bees knees.

* Good documentation

Cons

* Still young, not completely optimized.

* Only 2 rendering algorithms, neither suitable for caustics.


LuxRender

Mr. Snippy by Simon Wendsche

My impression of LuxRender has always been that it's very accurate, but very slow. However with the new API and transition to version 2.0, it seems it'll soon be a real contender with the top-dollar renderers.

Pros

* CPU and GPU rendering.

* Many rendering algorithms and samplers to choose from.

* Lots of post-processing filters like lens effects and noise reduction.

* Great documentation with many examples.

Cons

* New version doesn't support all features yet (e.g. hair)

* It's pretty complicated and hard to learn.


Maxwell

by 49_corp

On average it seems Maxwell was the slowest, but it has a unique look to it that I find appealing. After all, real cameras have curves noise.

Pros

* Uses real-life units of measurement, making it very easy to learn.

* Simple, but powerful material system.

* The Multilight feature lets you change the intensity of any light after the render is done.

Cons

* No material nodes, which makes it a bit tricky to work with textures.

* The slowest and noisiest on average.


Mitsuba

Test render - scene by "zeealpal", render by "bashi"

Mitsuba is actually a renderer focused on research, and is not actually intended to be used for 3D production. That means you may encounter some serious show-stoppers like no hair support, but it's great for testing new rendering algorithms.

 

Pros

* Many rendering algorithms and samplers to play with.

* Excellent documentation.

* Open-source and popular among researchers, making it the perfect playground for features that haven't made it into mainstream renderers just yet.

Cons

* Generally not suited to serious 3D production (lacks important features).

* Very complex and hard to learn.


Octane

River Side by Enrico Cerica

As someone coming from Cycles, Octane felt quite familiar. It was the first commercial engine to officially support Blender, which I hope is a trend other engines will follow (I'm looking at you Arnold).

 

Pros

* Like Cycles, it's very interactive and easy to learn.

* Nice documentation.

* Has the ability to optimize meshes for animation rendering.

Cons

* Only a few rendering algorithms available.

* You need a license for both Octane, and the Blender plugin (which is actually a separate modified version of Blender)


V-Ray

Realistic CG portrait of Robert Downey Jr by Frank Tzeng and Yibing Jiang

Among all of these renderers, V-Ray is by far the most popular.

Why? Because it's the fastest, and most adjustable renderer of them all.

Pros

* Extremely tweakable many algorithms, samplers and settings for each to optimize rendering for both animations and still renders.

* You can even use more than one rendering algorithm at the same time.

* Very popular, meaning there's lots of tutorials and examples floating around the interwebs.

Cons

* Because of it's complexity, it has a pretty big learning curve.

* The documentation is actually pretty bad.


Final thoughts

If 3D is just a hobby for you, forking out $300 - $700 probably doesn't sound like much fun. But luckily, we're spoiled with a choice of excellent free renderers (RenderMan recently became free too). Or if you want to just try the paid ones, most of them offer free demo versions, or even full versions for non-commercial use.

If you use Blender professionally, it's a good idea to stick your fingers in all the pies. Try all the renderers, and figure out for yourself which you prefer the most. Not only will that give you a better understanding of rendering in general, but it will also increase your chances of getting a job.

Obligatory Disclaimer

Now I'm sure you’re going to jump straight to the comments and tell me how I'm biased and this is not a completely accurate comparison. Of course it’s not!

Rendering is a crazy complex field, and each renderer could be tweaked for days to optimize the result in a passionate effort to prove its worth.

But a part of this comparison involved learning each one, and finding out how difficult it is to use and how long it takes to get good enough to achieve decent results, from the perspective of somebody that is already fairly experienced with Cycles (as you are).

Of course, I didn't even mention stuff like GPU rendering, Subsurface scattering, volumetrics and motion blur, nevermind animation and distributed rendering!

That's it!

I hope this article has given you that itchy feeling that we call curiosity and that you feel like going out there and do your own experiments with renderers. Once you get over the first frustrations it's a fun ride!

Happy blending everybody!

 

Acknowledgements

* Thanks to Maxwell, Octane and Chaos Group (V-Ray) for giving us a free license for testing their renderers,
* Mitsuba, LuxRender and Blender for making their renderers available to the public,
* Andrew Price and Greg Zaal for their feedback
* The artists that shared their scenes on Blendswap or the Blender Foundation, allowing us to run the tests.


Reactions about this comparison test of important Render Engines

Here we see the real problem of comparing several Render Engines with one and the same 3D model. Above test system and conclusions were very good and 3 months spending to fullfill. Quite remarkable.

But then again the reactions where people are interpreting more or less noise, not yet ready, something lighter or darker, not sharp enough. Less noise in depth of field. Different hairs and Shaders and light functions.

Or as Brecht Van Lommel (Blender developer of the first hour, now Arnold developer) tells:

But overall the comparison seems impossible, different hair shapes an shaders and the different light setups have a big effect on render time. I expect all renderers can achieve a look similar to Maxwell, though it may be more difficult to set up.

See also the special about Brecht van Lommel on the interview page


Artlantis 5 (former render version with a quite different render Engine then version 6).

comparison on quality of renders, not with specs or rendertime info.

http://artlantis.com/_forums/viewtopic.php?f=10&t=12196

date Apr 05, 2013

I suggest you download a real 3d studio to perform multiple views with your favorite software. This model is a great Mexican style home ideal for expressing your talents.

Best regards Franck Gérard.

Link to 3D model not found any more . . .


Comparison with render times and specs of hardware

http://artlantis.com/_forums/viewtopic.php?f=10&t=11227

 


New free 3D model of mine

May 6, 2013

http://artlantis.com/_forums/viewtopic.php?f=10&t=12269


Another look into comparison of a standard 3D model

http://artlantis.com/_forums/viewtopic.php?f=10&t=11218


Render Engine comparisons

http://forums.cgarchitect.com/18036-render-engine-comparisons.html

I'm sure this question has been posted before, so perhaps someone can point me to a thread where Vray and Maxwell are compared.

Maxwell and Vray compete, but in very different leagues as they're not based on the same type of calculations and algorithms. Vray is based around GI, Maxwell is not. Which makes comparisons irrelevant.

well, that's not exactly the point. maxwell is based "around gi", if it didn't calculate any global illumination or indirect lighting it wouldn't be much of a light simulator. so, both deal with gi, but in very different ways.

if speed is your main concern, then vray is what you should go for. it has a whole bunch of tools that can make your life much more easier if you had to deal with tight deadlines (fast gi methods, caching, baking, g buffer channels...), it's a very versatile and highly controllable renderer. plus, if you don't need a standalone renderer, it's perfectly integrated into max.


Benchwell Test

Maxwell Render documenation

http://support.nextlimit.com/display/maxwelldocs/Benchwell+Test

Benchwell is the official Maxwell Render benchmarking test which enables you to check your computer's performance under controlled conditions, so that you can compare your rendering power with other hardware configurations.


Speedcomparisons Octane versus CPU

https://render.otoy.com/forum/viewtopic.php?f=25&t=19928

Having in mind that my curent GPUs do the work faster than CPUs like 20times..if I would ivest into upgrade it's clear where to go, at least for me.

So in this case, when You Do comparison, compare apples to apples, not apples to oranges =)

(but always keep in mind the trade-offs, like vRam limitations and other things, like power, heat, etc =)

 


Indigo versus Maxwell Render

http://www.indigorenderer.com/forum/viewtopic.php?f=7&t=9271

Indigo have better tonemaping imo. Maxwell renders are kinda "dull".

Both renderers are powerful and depends on artist, which prefer what he can do with engine. Every renderer have some + and - and depends on artist how this could be spreaded.

Both renderers are powerful and depends on artist, which prefer what he can do with engine.

Thats true for every render engine. Artists produce nice renderings, not engines (also with povray).

I can't see any difference in quality between maxwell renders and Indigo renders. Indigo however is simpler and it's free. I would also say that Maxwell has a more artist friendly UI.


Maxwell versus SU Podium?

sept 2007

http://supodiumforum.websitetoolbox.com/
post/maxwell-render-or-podium-2180875

Maxwell is an awesome engine, as is V-ray for Max. The learning curve is by far greater than Podium and the render times by far longer. Podium at this moment cannot compete with these at full settings YET, however version 2 will be in the fight. With that said, I have taken a Max V-ray model and come bloody close with Podium, and if anyone has a Maxwell model and final render for me to try I would like to see how far I can push Podium in the Maxwell direction.

Maxwell is a good tool for exteriors.

But I find it very expensive just to use for that purpose.

As for interiors, Maxwell just is very limited because the render times are counted in days (for scenes with a lot of indirect light and dark parts).

Vray is a better solution concerning render times, but the price is just too high and you have to struggle through a very high learning curve. Results can be pretty though.

Podium still is a very good and a well 'price/quality balanced' piece of software.

The low price and the ease of use just makes this a top tool to use. In good hands, Podium can produce great images.

As solo said, the next Podium version (v2.0) will have an improved render engine so watch out for that.

When it concerns workflow, Podium is your friend.

The new internal beta('carotene') is even better in that perspective (very handy presets) than the current official 1.4.1.


V-Ray versus Maxwell?

http://forums.augi.com/showthread.php?18583-VRay-vs-Maxwell

 


Thea or Maxwell?

http://sketchucation.com/forums/viewtopic.php?f=80&t=29411

speed is important as is the possibility to do animation.

So far i've tried SU Podium and Kerkythea.

Podium is easy to use and you quickly make very nice renders and you run it inside SU.

Kerkythea has much more to offer but isn't easy to learn... i usually find myself spending hours just testing settings i don't yet understand :?

 

Have you seen Octane render?

Man... it blew me away, now i'm going for a new computer so i can fully use Octane.

You can export from SU to Octane... work in Octane with materials and lights and do test renders.. then go back to Su and continue modeling and again export back to Octane and all your previous lights n materials are still there.


Link

Apr 27, 2015

I am interested in best render for product stills (no animation) that ... on youtube

about a guy comparing several renderer, Maxwell is noted to ...


Comparing Global Illuniation (GI) renderers

using Blender, Maxwell, Indigo, V-Ray, Mental Ray, Octane Render, iRay, Mantra, Povray, Yafaray etc.

Vray Vs Arion Vs Maxwell Vs Octane Render Comparison Cinema 4d

http://mp3.topsong.co/download/lagu_vray_vs_ari
on_vs_maxwell_vs_octane_render_comparison_cinema_4d/


Comparison between 4 SketchUp renderers

Nov. 9 2011

http://ankitsurti.blogspot.nl/2011/11/comparison-between-four-sketchup.html

Indigo render
SU Podium
Maxwell
V-Ray

 


Spectral versus Non-spectral

http://blenderartists.org/forum/archive/index.ph
p/t-251118.html?s=9fb14ad93ff607ad6c7cd821b88258b3

I think maxwell is several levels ahead of everything else. Luxrender looks "comic" compared to luxrender. It is the way the wavelengths are being used, probably maxwell does these calculations much better (perhaps it uses more scientific data of how the light goes).

What I'm saying is that if Cycles was changed to a spectral-renderer, than it might actually be more difficult for new developers who haven't yet done a lot of deep study on raytracing to get up to speed with the Cycles code because it would be more complex and there would be more areas where something could go wrong.

So the question is if you use a spectral renderer but don't use actual "measured" data on your materials would you still get the benefit of spectral rendering ?

Concerning Maxwell I must say one of the best features it has is Multilight but I think that Vray has something like that too.


12 core versus 8 core versus 2 cores

Cinema 4D render speed test

http://greyscalegorilla.com/blog/2010/11/12-core-
vs-8-core-vs-2-core-cinema-4-render-speed-test/
 

Vray-vs-arion-vs-maxwell-vs-octane-render-comparison-tutorial

http://play.tojsiab.com/SF9WZERYUGp1Qm8z

 


https://www.behance.net/gallery/21615831/Testing-render-engines

by Dusan Vukcevic

testing render engines with Cinema 4D

V-Ray
Corona
Octane
C4D Physical
Element 3D V2
Keyshot
Thea


http://www.skyscrapercity.com/showthread.php?t=1605826

Best Sketchup Render Program

I'm using SketchUp for a while now. I'm make mainly skyscrapers, but also houses, and for some other things.

And I'm wondering what the best Render Program is for SketchUp.

I search for a rendering program, which shows windows transparent and also reflect them good.

 


http://sketchup-interior-design.blogspot.nl/2012/03/five-rendering-plugins-comparison-part.html

5 Rendering Plug-ins comparison

March 16, 2012

 


Render engines comparison

http://blenderartists.org/forum/showthread.p
hp?331127-Render-engines-comparison

 


p451-lee.pdf (470 KB)

Debunking the 100 x GPU versus CPU Myth

An Evaluation of Throughput Computing on CPU and GPU

 

http://www.hwsw.hu/kepek/hirek/2010/06/p451-lee.pdf

In the past few years there have been many studies claiming GPUs deliver substantial speedups (between 10X and 1000X) over multi-core CPUs on these kernels. To understand where such large performance difference comes from, we perform a rigorous performance analysis and find that after applying optimizations appropriate for both CPUs and GPUs the performance gap between an Nvidia GTX280 processor and the Intel Core i7 960 processor narrows to only 2.5x on average. In this paper, we discuss optimization techniques for both CPU and GPU, analyze what architecture features contributed to performance differences between the two architectures, and recommend a set of architectural features which provide significant improvement in architectural efficiency for throughput kernels.

 

CONCLUSION

In this paper, we analyzed the performance of an important set of throughput computing kernels on Intel Core i7-960 and Nvidia GTX280. We show that CPUs and GPUs are much closer in performance (2.5X) than the previously reported orders of magnitude difference.

We believe many factors contributed to the reported large gap in performance, such as which CPU and GPU are used and what optimizations are applied to the code. Optimizations for CPU that contributed to performance improvements are: multithreading, cache blocking, and reorganization of memory accesses for SIMDification.

Optimizations for GPU that contributed to performance improvements are: minimizing global synchronization and using local shared buffers are the two key techniques to improve performance.

Our analysis of the optimized code on the current CPU and GPU platforms led us to identify the key hardware architecture features for future throughput computing machines high compute

and bandwidth, large caches, gather/scatter support, efficient synchronization, and fixed functional units. We plan to perform power efficiency study on CPUs and GPUs in the future.

homepage

omhoog