Saturday, March 31, 2012

VFX Student F.A.Q.

Hello all,
I've had a number of students send me mails asking questions about real time VFX for games. A number of questions have been the same so I figured I'd post them in case a student came by and had a question here or there. Feel free to ask if your question is not below.

Ok here goes:

1) Are there any books I can read for creating VFX? 
I think Elemental Magic Vol 1 and 2 are great starts to get a handle on forces and an overview of elements and visual fx in general. Its a great base to build from, and I can't suggest the two books enough. There aren't a lot of books out on the subject that aren't a part of a tome sized instruction manual. I've used some of the maya innovator books, and I've looked into a dynamics workbook, but generally its more beneficial to look on line for solutions for direct problems.  I've done a couple book reviews here take a look:

2) What programs do you use?
Being a VFX artist is a learners profession, meaning that you'll be continually picking up and learning new tools and techniques as new problems and technical benchmarks arise. So any tool that you have a chance to try out or use then use it and try out experiments with them! I've learned the most when I've failed, since those experimental setbacks have caused me to look at my VFX issues in a new light and I come to a solution that is better or new... eventually... haha. Before working with VFX, my background is in traditional and digital illustration, textures, skinning and modeling, so I have a working knowledge of mesh generation, normal maps, specular maps, general texturing, and UVing.

I use Maya for modeling and dynamics to generate fluid simulations, Photoshop for general texture creation and editing, AfterFX for looping and FX work for the renders, Premier for editing. I mess around with Zbrush, and general modeling stuff. Lately I've been looking into Houdini, Nuke and physics based destruction sims to see what they can do. I use the UDK for building shader materials and I use it as a work table for some effect tech I want to implement. The beauty of that is its free to use, and is an invaluable tool for testing ideas. They also have a searchable database on their engine and they've put out a couple pages that are invaluable for VFX for the UDK.

To build in game effects, I'm usually using an in game tool or a proprietary toolset to build particles and emitters. I know some studios use a limited data set from 3d packages directly in their games. All in game VFX systems have their quirks, learn them and learn how to troubleshoot your VFX. There are a lot of smoke and mirrors in the job, the best way to do that is to learn your system.

3) Do I need to know HLSL/python/MEL to do VFX?
Its always good to gain a working knowledge of any scripting if you have the time, since it will help you to communicate with a tech artist or a programmer. I don't know any of these scripting languages directly, but I've looked into them enough to talk with others about what I'm trying to do. Its also helped with my experiments in Unreal materials. The realtime rendering books are also good to look at for getting a general idea on graphics pipelines and rendring... a bit math heavy, but there are some good nuggets of info there.

I would say if you learn anything like this learn how to make shaders/materials. The easiest way is to download the UDK and start with the Unreal Material system. The more options you have with your Visual effects the better.

4) How do you get a flipbook/subUV from Maya to a game state?
I have a blog entry that goes over the procedure to get frames from a render to a game state (Unreal 3). Its here:

5) How do I break into video game VFX?
This is a matter of debate. For me I was already in a game company when I was figuring out how to make VFX. First I'd cut a reel of your Visual FX work and see what that yields. If you're not working at a game company and want to get into it, start looking at the UDK and learning how to make VFX textures, shaders and using Cascade. Then I'd either work on that and cut a reel or find a mod group that is looking for someone to make VFX for them.

6) What should I put on my demo reel?
This question is another matter of debate. Before starting, ask yourself a couple questions. What games and VFX are you looking at for inspiration and how do you compare to them? Can I do anything else to my work to give it a little extra polish? What kind of game company do I want to work at?

So what should you put on your demo reel? The easy answer is your best stuff.

I've looked at a lot of VFX reels and talked to a couple ADs on this and generally try to keep your reel between 2 and a half to 3 minutes. Its a good way to limit yourself from going overboard and to edit your own work to a manageable state. As far as what to put on there, if this is your first reel, then I'd cover as many elements as you're able to do. Fire, water, water splashes, full screen FX, environmental effects, weapon and hit effects, explosions... and if you can, cater some of the reel to the company you are applying to. If you don't have all of that then that's ok, but show your best work. Before sending it out, have some other people (instructors, forum peoples) to look at it, getting more sets of eyes on it will help you make informed decisions. Its easy to fall in love with a shot or over work a reel, so getting an outside opinion always helps.

This is a personal preference, but I like to show all of my VFX in context. In game if possible, although I've seen reels with tests from a third party product in them that are good too (like fume or fluids). Usually they transition from the test to in game to show its use. If you can't do all of that, then make sure all of the stuff you do show is worth showing. I've seen long kitchen sink reels that aren't all impressive and short reels that blow me away, so don't automatically think that if you have a lot of footage that its good. Remember to have a call out sheet if multiple hands have worked on the scene or VFX.

This is a little harder but try to treat the reel like a trailer. It sounds hokey, but if you look at it as a trailer with beats, and rhythms then you'll have a more compelling cut and will be more impressive. If anything, start and end strong.

7) Will you look at my reel and let me know what you think?
Sure I'd be happy to look at your reel if I have time. It all depends on how tight my deadlines are and if I have some free time to give good feedback.

Ok that's it for now, if I get any more questions I'll do an FAQ part 2! Thanks for reading.