Well that was the intention, to essentially make it easier for ADSK and MR to hopefully address any issues, etc.

I can understand the argument that you can´t just implement any feature right away because its always a risk.
But like Kzin pointed out there are features which exist since (i think) 3.7 and we still don't have access to them in 3.10... lets take progressive refinement as an example (I think its called like that).
And this feature alone would be a huge time saver.

And then you hear we only want to expose features when we know they are stable and working, which I always liked and think its the right way.
But as the years are going by it seems nobody does care about that anymore and it will never happen. Or does it mean that this feature is still not working correctly after 3 new versions of mr?

I don´t always need the newest thing right away and love it when maya/max users are then betatesters for things which don´t work like they should but like I said after 2 -3 new versions you start to wonder, what happened to this feature?

I agree, unfortunately I don't have an answer. Not because I'm not allowed, tbh I simply don't know.

While this sounds like a very good idea, I think in practice it might not be as good as we think. I assume AD pays a VERY reduced price for MR because of the huge volume. MR is comparable cost to the other renderers when purchased separately. AD probably pays far less than half of that. So just a wild guess, maybe $300 goes to MR from each SI seat (remember, just a guess). Take it out and you get $300 off SI. Is that really worth it? Then you'd have to pay the $1200 to $1600 or so for a new render engine. They don't sell the same volume so they won't give the same deal. Is it that important to save just a few bucks and not have MR?

I've done a fair bit of testing (and watching other people test) other rendering systems. I won't name specific software but they all have plenty of problems too! Some of those problems, if they were in MR people would hit the roof! People like to complain about MR but it really does provide a pretty complete solution and also people have used it for years and have scenes and assets developed for it. If it suddenly wasn't in SI that might be a major hassle for many.

As much as I'd like to see this sort of sales model, I'm not sure it would be as great as it sounds.

I always thought Autodesk should do what Autodesk does best: BUY a company that produces a renderer and then integrate it in other apps. In the long run it probably even wouldn't be such a good idea, but in the here and now that might be quite helpful...

You are right it would probably be wrong move, especially for people that use MR.
But there are two important things, first, less and less people are using it, and to be precise less studios are using it. So it means less feedback, and that means less features implemented and bugfixes.
Second, also you are right that there are tons of problems with alternatives, but difference is that with other render engines bugfixes are fast. For example, I found bug with Vray, reported it, and 5 days later I can download build with fixed bug. With MR you need to wait Nvidai to fix it, which is usually fast, but when that will be implemented by AD you can not know, a year at least.

its the " the gras is always more green on the other side" effect. i also had this with all render engine i used sofar.
but for mr come more things together, first, the old school presets and also users which are not willing to learn and read some things, david aka bitter wrotes some words about this on cgtalk.
i personally dont like people which are crying without to know that much about mr (yes i know, i am doing the same from time to time. ).
but when you explain and they start to understand its getting better and better, but this needs time and people who are willing to learn and ad to implement the current and widely used features.

Yes, every time I try a new render engine I run into things that make me say what? You have to be kidding me! Often it makes MR look not so bad

It's not that they're not good render engines, it's just that they're not as mature and are often missing VERY key features that I use. They may have faster bug fixes but they still have huge gaping holes in their feature set.

For example, has anyone tried rendering volume particles in V-ray, 3Delight etc.? Good luck with that. Just an example.

Also, the render delay from when you hit go that MR had at the beginning of 2012 made big news and people were furious. Including me. But have you checked the rendering delay on some others? Far more in most cases. Just another example.

In a way, having MR gives me the freedom to test other systems because I know that if it doesn't work out, I can always just go back to MR and I know I can do it. It would be pretty risky otherwise.

Ironically, I think if MR wasn't included (and never was) people would want it more.

Well ok. This is the first version of V-ray for XSI, and you can excuse it for not having support to all features. But that does not make MR a good thing
Again nobody is saying strip MR completaly out of the package. Just have it as an optional bundle...


True, but... Autodesk is responsible for the actual implementation in the software. Quoting Vlado (founder of V-Ray) implementation is about 80% of the work, vs 20% for actually building the render engine. So while NVIDIA is getting little money per license, most of the work is up to the Softimage team to deal with. Even more so because historically MentalImages was known for giving sh*t about software implementation and leaving tasks such as Shaders up to the ones doing implementation. (read this: https://groups.google.com/d/msg/xsi_list/BoSKNhwLkJ8/VQ5JzLG-_hcJ)
So I can only imagine treating MR as a plugin like any other would only benefit everyone since (1) the Dev Team would be able to focus on the SDK (2) NVIDIA would have to (FINALLY) work for theire meal and not just tell us "hey if you want the juicy stuff you got to buy the Standalone". A transition period would, of course, be needed.

Wise words Gustavo, but really, do you think there is any AD personnel (the men taking decision) listen your ideas about?