Autodesk develops its own rendering engines


They most likely got Lagoa more for the cloud side of things than for the actual rendering contained. There are a handful of rumours floating around about why and who else was involved (allegedly a bid war got slightly out of hand, but underline and bolden allegedly there), but it’s one of those things we’ll probably never know precisely.


Inconceivable to me that they killed Softimage in favor of pumping money into garbage like this and Stingray.


Eeeehhhh - I was bit off from news, didn’t knew Autodesk had Lagoa under its belt as well now. For me what first comes into mind hearing Lagoa, is quite a fantastic multi-physics solver developed by Thiago for Softimage 2011. Hadn’t been Softimage user myself, but was gravitating towards it and tech demo of Lagoa was truly captivating when it was released.

I know, I know :blush: , eeer - former XSI/Soft users may scream and yell at me for what Autodesk did to their favorite baby. But still it was pretty cool from autodesk side to make possible such addition to softimage tools.

But reading now that they own Lagoa cloud-rendering company, founded as far as i know by the same Thiago guy, and recalling recent Softimage eol event quite frankly gives some irony and creeps as to into what all those autodesk efforts might turn into at the end of the day :curious:


They got Lagoa for the Cloud CAD market. After the Lagoa acquisition…all of a sudden…there’s free cloud rendering in Fusion 360.


First, I wouldn’t say Mental Ray is dead, or hardly used. I believe we have like with everything, many lemmings, what is popular gets more attention, that is not to say Mental Ray is dead, that is like saying no one buys milk alternatives, just drink milk.

For years we were told that reading a book on a tablet would take off, those numbers are falling apart; like when the government tells you that thousands of Jobs were created, when you break it down, it’s a spit in the bucket.

The whole purpose behind cloud rendering is to make it a service, I’d be pretty angry if they stopped packaging mental ray with Maya, it may not be fast but Mental Ray has brought us many films and TV shows.

How many of these current render engines will be around years from now; besides RenderMan ? And I’m a fan of pure GPU renders, I think they are fantastic, even I am skeptical on they’re life span.

It’s when companies force us to do something against what we want that annoys alot of people, and it should, because a small number want, for example, cloud rendering for some assumption, we all want cloud rendering, reminds me of the clip from the breakfast club, in a similar ideology.


As far as feature VFX and animation work is concerned, it is very dead.


You might wanna download the fully functional watermarked demo of RedShift 3D. The future is now dude.


I’m ahead of you. RedShift 3D is hybrid, I’m trying Octane it’s pure GPU but it’s very fast. That is not any reason to abandon Mental Ray.


The CPU in Redshift is used only for exporting data to the GPU (geometry, textures, preprocessing data required by the GPU) as with all the other GPU renders. The whole render is GPU only there is no CPU involved in the rendering just the GPU.
Why do you think Redshift is hybrid? Could you clarify please?


Well its gonna take a little more effort to even use MentalRay going forward as it is no longer apart of the release package, but a separate install all together.

Make of that what you will…


Maybe he’s confusing biased vs unbiased?

Because like you said, all GPU renderers use the CPU to load scene data.


I read an article a while ago that mentioned that RedShift uses a hybrid approach; sorry I don’t have that link anymore, although I could try to hunt it down if necessary ?

You are saying that RedShift and Octane both use the CPU first to load the data, then to send it to the GPU ?

RedShift is more forgiving on your GPU; hence it’s popularity, unlike Octane which strictly relies on your GPU, obviously a GPU or a sequence of GPU with wads or a combined wads of Vram would have no problems with Octane, that comes down to budget.

I personally like Octane, what I don’t like about oToy is there price system; it’s only in Euros and if you are not aware of the major currencies around the world, the Canadian dollar is to go below $0.70 the price for the Maya + Standalone would be wonderful if there was some consideration for our dollar, for example a sub-$600 would be more attractive, atleast RedShift deals in US dollars.


RedShift is hybrid because it provides both biased and unbiased rendering.

Octane is only unbiased.

“Unbiased” means, when rendering, light behaves as it does in nature using the light equation. Solving that equation takes alot of processing to arrive at the solution (image).

“Biased” means, when rendering, light doesnt behave exactly as in nature, and assumptions are made, to arrive at the solution (image) faster.


Octane also has a biased mode, it’s called the Direct Light kernel. The Path Tracing and PMC kernels are unbiased.


Can RedShift do both ? I believe Arnold can but I know nothing of Arnold.


Usually a hybrid render is one that uses both CPU and GPU, you are referring to a combination of unbiased and biased solvers.
As far as I know Arnold and Maxwell only do brute force solving, Renderman, VRay, Redshift, Mental Ray can combine brute force with other approximation methods for speedups and flexibility.
In the end the result is the only thing that matters, if it looks good it is good no matter the road that takes you there.


I think you are mixing up unbiased/biased with physically based.
Unbiased/biased only states if statistically the image is correct on average independent on number of samples. A biased renderer may still converge towards a correct answer (if the bias goes toward zero as the renderer progresses). A biased renderer may also be more physically accurate than another unbiased renderer (think caustics for instance, not all unbiased renderers will account for it, whereas SPPM does). Also compare a biased renderer that uses measured material BRDFs with an unbiased path tracer that uses Phong.


Right, thanks for clarifying that. I was “thinking” in that direction but it didnt translate into words as clearly as I’d imagined.


You’re welcome! :thumbsup:


Hi :slight_smile: some information about Lagoa Render;

The ray tracer is uni-driectional but the company does not like working inside labels such as ‘biased or unbiased’. “Generally, I think the biased vs. unbiased battle is over - consistency is the key,” says Zinke. “We are currently exploring the use of other methods (including fully bidirectional path tracing and a novel progressive consistent method) but the current implementation is uni-directional. On top there is a path space analysis to reduce ‘fireflies’. The method goes beyond standard approaches, like clamping or BRDF smoothing and is less aggressive (more selective) when dealing with hard-to-sample paths.”

There is extensive use of importance sampling, on materials and light sources and IBL. “We use (multiple) importance (re)sampling for lights (also in spectral domain, when having for example spectrally varying scattering coefficients in case of SSS), image reconstruction filter, phase functions and all other materials,” adds Zinke.

The team at Lagao have explored MLT(Metropolis Light Transport). “We considered using MLT. I agree that, in its pure form, the unavoidable impact on rendering speed and the bad stratification of samples etc. are serious shortcomings. However, for certain light paths Metropolis sampling (or similar) is the only way to go. As said we are actively working on several methods for improving speed and quality. For us the perfect method has to be practical: interactive/progressive, must scale, has to allow for dynamic scene updates and has to deliver consistent results,” explained co-founder of Lagoa Arno Zinke.

it could be a serious contender for Arnord and Renderman RIS and of course Nvidia Iray and mental ray…:keenly:
Ps: i love Mental ray :slight_smile: