Vray was worse. It was 100% Max based in the beginning :pBecause they've written Vray around Maya's architecture, Mental Ray was an existing standalone engine and they've cut and chopped it in places to make it fit within the architecture sometimes cutting away features or diminishing them in some way.
you're right that is worse LOL, lucky for us Autodesk is making Maya more like 3ds Max with each release Hahaha.Vray was worse. It was 100% Max based in the beginning :p
Yeah, I'm aware of that, I think there was a mix up there. I should have been clearer, the question was directed towards SirCharles.It's not dying out. It's just getting old. Just like Renderman. With all the years of development, it's bound to have some legacy code in it. Newer renderers may benefit from having smaller, fresher code bases, which allows them to put in some pretty neat new features, and have faster releases; but the older ones have been tried and tested in production.
Let me just say this, and try to avoid more people jumping onto this; use whatever tool gets the job done, and you feel comfortable with. Don't switch just because something's becoming popular. Switch because it does something that your current solution doesn't. In production, whole pipelines are built around a single solution, bet it Renderman, Mental Ray, Vray, etc; and it's therefore extremely hard, if not impossible, to change solutions without grinding production down to a halt.
Use what you want to.
Just to be clear, the animation was done in Lightwave and it was rendered in Modo. Probably now they would do it all in Modo since the upgrade.@DJ - Given that they used LW for the rendering I'd guess it was done in Modo 501 or before since there's no way that kind of CA could have been done without much heartache prior to the new release.
Very impressive short though.
Very nice thread and points of view from everyone.I agree that mental ray is too damn complex. You seriously have to have a PhD in light theory to half understand what a quarter the damn switches and knobs do. That is why in the end people just turn a shit ton of unnecessary stuff on (since most have no clue what each really does and how each affects the other) and end up with painfully slow renders and mediocre results.
I honestly spent a day reading through the docs and they read like PhD research papers in quantum hyper pan dimensional light theory originally written in German and then translated to English by a French person that only speaks Chinese.
To do it right you have to spend a metric butt ton of time twiddling mysterious knobs and dials to get a decent result. So seriously, what is the point of getting a render down to some small time if you have to spend 5 days twiddling knobs and dials to get a result that only applies to a single frame!
If you have not seen the modo render engine, I recommend you look at it (I believe it is the same that was just added to lightwave 10 and 11). I have used it several times and it's as close to the magic "make it look photo real" button as I have seen. It's very intuitive and relatively easy to use. I am not sure what is under the hood but it's also fast, and seems to have a maxwell constantly iterating feel to it.