Spin: My graduation short, fully rendered in Viewport 2.0
polycounter lvl 7
Hey everyone!
This is a short film I've been working on at school with 2 other blokes.
It's fully rendered in real time using Maya 2014's Viewport 2.0, DX11 shaders.
Due to lack of time we really couldn't bring the assets to their full potential, but the whole thing connects better than expected.
Hope you like it and would love to hear some opinions!
https://vimeo.com/77894325
This is a short film I've been working on at school with 2 other blokes.
It's fully rendered in real time using Maya 2014's Viewport 2.0, DX11 shaders.
Due to lack of time we really couldn't bring the assets to their full potential, but the whole thing connects better than expected.
Hope you like it and would love to hear some opinions!
https://vimeo.com/77894325
Replies
So working out all the early stages took us most of the college year, around 5 months or maybe more. During that time we were busy designing characters and doing render tests on Maya, designing shots and directing the film (storyboarding).
Then with the remaining 3-4 months we modeled everything, rigged it all (thank you advanced skeleton), set up cameras, lighting, render settings, did the animation and exported it out to After Effects for some final post work.
The beauty is that because rendering was so quick, the majority of post work was done within Maya, and we only added some lens flares, smoke and did minor color fixes in After Effects.
I do believe this is going to become a sort of a standard for film makers as time goes by. I chose VP2.0 mainly out of belief that it's inevitable for us to be using real time engines for everything eventually.
If you're a student working on a short film, my suggestion is this: Work with a real time engine. you probably don't have time to bring out the full potential of a standard rendering engine, so might as well use the extra time saved by real time rendering to push the limits of a game engine instead. The end result can be awesome, this film is just the tip of the tip of the iceberg, I mean it. The potential is awesome. And if you use a game engine, that's even more awesome.
This film was a hell of a ride, I really had no idea if it's gonna be any good, but response has been positive, it makes us all thrilled.
I think some professional studios are at least experimenting with real-time engines... I've seen stuff on real-time live mocap and that recent lucasfilm thing where they used 1313's engine for real-time animation.
Yeah, the Lucas Films Star Wars thing came out about a month after we finished the work on this film. Obviously I was blown away by the level of realism they achieved. 1313's engine is based on unreal 3, but still the quality there is staggering. I bet it's some monster PC with 200 tesla cards, and the absolute beyond maximum of 1313's engine.
Either way, my plan for this film was to render it out of Cryengine 3, but that was a huge mess to handle since I was the only one in the team with any knowledge in game art, or game engines, and my knowledge in CE3's cinematic tools is very limited. So we stuck with Maya's tools, which posed a bunch of problems (mainly shadows and managing big scenes with lots of textures).
I was kinda hoping to get my hands on Cinebox but that didn't happen. Still hoping. I'd be happy.
So I'm curious as to how you achieved this through Viewport 2.0. And what do you mean by render? To the best of my knowledge you can only playblast Viewport 2.0. Would you be willing to share your Viewport settings, and which render you utilized (i.e. maya software, maya hardware, hardware 2.0) I can give you my email if you'd prefer not to make it public.
Best regards
It's Viewport 2.0.
The project was rendered using Maya 2014, the first version where VP2.0 first had DirectX11 support.
Since you can't batch render using VP2.0, I remember using the standard rendering options.
I rendered the project scene-by-scene. So you render through the normal render screen and you set it up so it renders a frame range, instead of just a single picture.
It was a bit of a hacky approach, I admit. We wanted to use a game engine to do it, but I was the only one up to the task and I didn't have enough experience with any engine back then, to confidently attempt it with the time we had.
If you want more info, feel free to ask more specific questions. I don't mind sharing anything here, but images from the original setup would be a problem to get. Not to mention, 4 years later I realize how much I'd do differently.