Game Animation Reel: Examples and Advice?

[DrAdamDinosaur]

When we review reels at work, the first thing we go for is the shot/breakdown list where you tell us what elemnts of the reel are yours and what aren't. If employers aren't sure what your input on a peice is then they can't judge it.

As for motions, show peices that show you can do weight and timing. A guy jumping onto a ledge and pulling themself up. Diving away from something. Runs and walks are good but showing good clear actions help a whole lot. Two characters interacting is also a good idea. So would a facial animation test (a good one). I wouldn't worry about repeating motions, if they want to see it again they'll rewind. If your motions are to short think about ways to extend them. Don't just show one loop of your run, show ehough that people can see it and that they can see you cna make it loop well.

Also, make sure that your characters are easy to read on screen. Don't do a bright red background with a bright blue character(I've seen it, it makes my eyes hurt!). You said you're not going to go over the top on lighting, good, but you have to make sure that your character reads well.

Frame count, thumbnails etc are all good but I don't think they're necessary for a reel.

Hope that helped.

Good luck

[Headless]

Thanks. I'll bear all that in mind. :)

Anyone else have opinions?

[switchblade327]

I'd say for a reel where you're shooting to do just real time, in-game animations, it'd be good to show 3 different characters, as different as possible, doing some of the same cycle animations. Walks, runs, jumps, ducks, climbs, lifts, basic and distinct attacks are all good but dont feel you need to show every animation for a character. Maybe pick the 10-15 best? Ive done characters that had 100+ moves and not all of them are terrible exciting.

That variety shows you can handle different styles and character personalities, as opposed to just being good at doing things one way. Even three different characters from the same game would be good if it was a knight, a princess and a monster or something along those lines.

If you have any good work showing animal/nonbipedal animation, show it! I expect we'll see more MoCap with next gen games but they cant easily mocap a hampster or an elephant!

No backgrounds except maybe a matte/shadow plane (if in Max). I just went back and noticed you're already doing this. Good job :) In line with what Dr Dinosaur said, low saturation colors are probably your best bet.

For looping on my own reel, I play combat actions once and cycles (walks, idles) 3-4 times, depending on length. Between big actions Ill use throw in an idle loop to let the viewer get their bearings before the next big thing. This also depends on the media you submit your reel in. If I can loop a run in media player, great. Not so great on a VCR.
I wouldnt loop anything for more then about 3-5 seconds though.

I dont know if showing animation from multiple angles is really necessary, but some people might disagree with me. In game type cycles arent usually the kind of actions that can be fudged by camera angle. I always go for a 3/4 camera, a foot or two above the character's head, angled down. I would try to keep motion in place for the most part, though you might find some animations would look better in motion (ie run in place, flying jump kick crosses the screen).

I agree that demonstrating weight and timing, plus a little bit of acting is going to help any reel. Face animation is going to be a bigger deal in future games so I agree that you should show that too.

Shot breakdowns are more important for completed scenes and I think it's not as necessary when you have one untextured character under a key light and no background, but its a good and easy way to cover your butt just the same. Id just include a note saying all animation is mine (since in my case, it is); Models, textures, rigging etc is not.

Good luck!

[Headless]

Cheers switchblade, that's all really useful info and I appreciate you taking the time to write so much. :)

[Lockan]

I'm also trying to do an animation reel with a game focus. I finished my animation schooling 6 months ago and got some really great stuff out of the program. I learned a lot, but my demo reel is rather lacking in quality content. It's not terrible, but it could certainly be a lot better.The information on here has been really helpful, and I'll be sure to incorporate a few of these ideas the next time I plan a new version of my reel. But I've got similar questions to the starter of this thread.

I can't seem to get many straight answers as to what industry actually wants to see in this sort of reel. The conclusion I've drawn is that everybody wants something a little different, so it's been very hard to focus my reel and do some better quality work.
Maybe somebody on this forum can help me out. The main issue I've had is this:

My feeling is that if the focus of the reel is animation, then the quality of the models/textures shouldn't matter so much. The motion is the most important. However, A few of the comments I've had about my current reel are things like "well the models aren't very good". I'm starting to think that I need to do one of two things:
a) Do some really really good models, and then animate those really well, or
b) Animate simple primitive characters or stick figures with really good motion and not even worry about the models or textures.

Is there anybody out there who can suggest one route over the other? I'm sure each way has it's pros and cons. What I'm more concerned about is which method is more likely to get me an animation job?

[eek]

I was reviewing work at my last company, and generallY if the first 30 secs doesnt grab me, i close the file. I look at just animation, no textures/lighting rendering just animations: playblasts etc. I basically have a checklist:

Walks
Runs
idles
Weight & timing
performance based: fights, twists,leaps etc
acting pieces
quads: trots, gallops, etc
winged/tail creatures

with walks DO NOT do standard generic walks, give the character an attitude whatever it is!

Additional flares/points: facial stuff, a short film etc etc

latest work first

rules: keep it simple, no strong music. Add you contact details, and credits if you need to accredit the models/rigs.

length 30 sec - 3mins no longer (its basically a trailer)

on the cd provide additional contact information in a text file, if they cant open the cd up. Include your cv/resume on the disk. Upload two versions: quicktime , msvideo. Dont use codec like divx xvid. Dont make it a dvd. Cd's are fine. If possible make a cover for the cd, titled to the company your applying for so they can keep it. Keep the aspect ration standard. 1.333 360X240, 640X480. If you have to go widescreen use 1:185 400X300 etc.

eek

[switchblade327]

However, A few of the comments I've had about my current reel are things like "well the models aren't very good". I'm starting to think that I need to do one of two things:
a) Do some really really good models, and then animate those really well, or
b) Animate simple primitive characters or stick figures with really good motion and not even worry about the models or textures.



Or

c)Get ahold of some good, simple public domain, skinned models (I think there are some at http://10secondclub.net/) and use those. Especially in this case though, it'd be important to note very clearly that the models aren't yours. You might find one that gives you a chance to do some facial animation too. I've never done this myself, but I've seen the same characters in a lot of different animation reels.

Otherwise B. Definitely not A. It's a massive diversion of your attention and waste of time to try and do a great model for the sole purpose of showing off animation.

Bad models/textures/lighting/etc *can* be distracting on a reel, so if these aren't your strengths, you should leave them out entirely.

[velarde]

Dont make it a dvd. Cd's are fine.
eek


Charles.. sorry to but in in the conversation but I was wondering the same thing...

The best format for a reel... Why no DVD...?

I thought most people will have a DVD player on their machines by now... and a autostarting DVD seems like a good idea...

You prefer VCDs...?

Thanks

fjv

[switchblade327]

Im pretty sure any dvd player can also play VCDs. If I'm wrong, I'd very much like to be corrected since I've never found out for certain. But if that's the case, why waste a DVD for 2-3 minutes of footage? Personally, I've found burning DVDs to be a more delicate process then burning VCDs, but maybe my software and/or burner just sucks.

Format is something I think you need to address on a per-application basis. I'd be ready to do versions on cd-rom, vcd, dvd and vhs, and send whatever is specifically asked for on a job posting. It seems to vary quite a bit, even now.

[eek]

region code.


Say i send my work to ILM, with region 2 - how will they play it? The studios ive looked into hate dvds, they want to open up folders and pick and chose.

eek

[Lockan]

The other reason I've been told to use CDs instead of DVD is this control. Especially for an animation reel, it's helpful for those reviewing it to be able to scrub the video back and forth and review specific portions of the reel. DVD tends to take away some of this control and turn it over to the player/software, whereas a quicktime movie on CD allows them more control that is needed to scrub properly.

[zenfull]

on the cd provide additional contact information in a text file, if they cant open the cd up. Include your cv/resume on the disk. Upload two versions: quicktime , msvideo. Dont use codec like divx xvid. Dont make it a dvd. Cd's are fine. If possible make a cover for the cd, titled to the company your applying for so they can keep it. Keep the aspect ration standard. 1.333 360X240, 640X480. If you have to go widescreen use 1:185 400X300 etc.

eek

I am currently working on my gaming demo reel for Blizzard Entertainment (The company I am trying to get hired unto) and what I have seen is that they do not want CD's at all. They want VHS or DVD so one of the best things I suggest about finding out what type of media to have your demo reel on is go to the website of the company you are making your demo reel for, check their job section they are usually pretty good at putting that information up.

Also beyond this, one of the things I have noticed and heard while I have been talking with some of their lead artists is this, currently, they do NOT want to see facial animation, at least I know this goes for Blizzard Entertainment, and if you are focusing on animation go ahead and focus on that, but it does not hurt to show you can model, or texture and light as well, at least that is what it says on the Blizzard Tips for Animators. I personally am still focusing on animation, yes I can do the rest, but I prefer not to, and when I get the time I will make a demo reel for those other catagories.

Hope some of this helps.

Kimberly

[edit]

Allright so I was thinking while reading some of the other posts in here, and I know that there are idle movements and a few other things so I would like to pose a few questions myself.

When showing run to idle to movement and such, should all of these movements be captured as their own entity so to speak and placed with fades or cuts of sort to move directly towards the next or should there be some sort of flow like for example character a will move in a walk cycle, then flow into a run, then come to a stop and move into an idle pose, looking around or something along the lines and then lift back up pulling a weapon and moving into an attack only to move back into a pose or something?

Also when working on a death pose which is one of the things that I was told to add in, I was planning on showing two characters fighting, one victorious, the other dead, how would I go about doing this? Should it be something simple? Dramatic? Should I make note of the deaths that happen in games or should I be watching movies that have a lot of battle sequences to get some inspiration from that?

Thanks in advance,

Kimberly

[switchblade327]

I am currently working on my gaming demo reel for Blizzard Entertainment (The company I am trying to get hired unto) and what I have seen is that they do not want CD's at all. They want VHS or DVD so one of the best things I suggest about finding out what type of media to have your demo reel on is go to the website of the company you are making your demo reel for, check their job section they are usually pretty good at putting that information up.

Also beyond this, one of the things I have noticed and heard while I have been talking with some of their lead artists is this, currently, they do NOT want to see facial animation, at least I know this goes for Blizzard Entertainment, and if you are focusing on animation go ahead and focus on that, but it does not hurt to show you can model, or texture and light as well, at least that is what it says on the Blizzard Tips for Animators. I personally am still focusing on animation, yes I can do the rest, but I prefer not to, and when I get the time I will make a demo reel for those other catagories.

Personally, I'd read that as "it doesnt hurt to show you can model, texture or light if you're good at it." Mediocre models can distract the viewer from the quality of your animation, whereas no serious lead animator/art director is going to frown on a reel that shows awesome animation on a very basic, simple character or even a biped. I wouldnt want to show any discipline on my reel that wasn't professional quality, whether that's the actual job I'm pursuing or not.

The short version of my rebuttal is "read what your desired employer asks for and try to show as much of that as possible." I wouldn't hold up a reel trying to meet their every last request or suggestion as a:)it's a hold up to your application and b:)a good character animation reel is a good reel.

I'd also advise that while it's good to have a dream company, you shouldn't put all your eggs in one basket.




Allright so I was thinking while reading some of the other posts in here, and I know that there are idle movements and a few other things so I would like to pose a few questions myself.

When showing run to idle to movement and such, should all of these movements be captured as their own entity so to speak and placed with fades or cuts of sort to move directly towards the next or should there be some sort of flow like for example character a will move in a walk cycle, then flow into a run, then come to a stop and move into an idle pose, looking around or something along the lines and then lift back up pulling a weapon and moving into an attack only to move back into a pose or something?

Also when working on a death pose which is one of the things that I was told to add in, I was planning on showing two characters fighting, one victorious, the other dead, how would I go about doing this? Should it be something simple? Dramatic? Should I make note of the deaths that happen in games or should I be watching movies that have a lot of battle sequences to get some inspiration from that?

Thanks in advance,

Kimberly

I think I sort of get what your asking. I'd show transitions where ever possible and logical, since game animations do need to transition smoothly to one another. If you have to jump between poses, it's not the end of the world. I'd do a hard cut rather then waste time with a fade.

For death animations, the rule we used on the last game I did real-time animation for was "chumps get simple deaths; bosses get dramatic ones." The more you see an animation repeat in game, the more generic it needs to be, hence the 'chump' rule. But for your demo reel? Go for the drama.

As for reference, I've never used it for a death animation beyond acting it out myself but if you see a cool death in a game or movie then by all means, add it to your library.

Also, two characters fighting together isn't quite as relevant to game animation as two characters' cycles, just FYI. Only us cutscene guys usually have the luxury of synchronizing our characters :P (Except in the case of grapples, fancy combos and other unique situations). Not that what you're proposing is a bad idea. Just thought Id mention it.

[zenfull]

Thank you for your advice switchblade, good advice indeed. Though I am curious when you mention about the dream company and not just going for that, do you mean I should go ahead and add stuff that the Blizzard guy told me I should stay away from such as facial animation? I ask this because in the last few weeks here at school Ive been working on my demo reel stuff, and I had this great scene planned out, got all the background work done for it, have all the GeneriRigs laid out in their starting position, created a simple set, picked out the voice clip that I was going to use from DailyWav and was just about to start in on the body movements before heading into facial animation. So I guess my question is, should I continue with this anyways and throw it into the reel just because I can and because I know I can do a good job with it so that I can use it for other companies or should I follow the advice of the guy from Blizz?

Also for the description of the differences on deaths, definately a good thing to know. So now my question for that if you don't mind, should it be a simple fall to the ground with of course follow through of arms legs hands and feet as well? Or should it be a little more then that? Should I just move around in the games I got and die a lot to see what other games are doing or something?

Thanks again,
Kimberly

[switchblade327]

Did the Blizzard guy say he doesn't need to see facial animation or to stay away from it? They may have code automated facial animation (very likely for WoW), facial moCap or they may have dedicated animators for facial stuff. I can't imagine why they'd expressedly not want to see any at all.

But yeah, what I meant by not putting your eggs in one basket is that you want to make sure your reel is marketable to as many companies as possible. There's always a very real chance that the first job you score might not be your dream company so be prepared for that. You'll notice on my handle that I don't work at Pixar :P

For deaths? I think you're thinking about it too much :) Do what looks cool.

They could...

-stare at the wound, look to the attacker in disbelief then collapse
-freeze up for a moment and drop like a sack of potatoes(if you want a sort of comic death)
-stagger around for a few moments, trying to keep fighting
-stagger towards the attacker, then fall forward and crawl a few more inches before going limp
-go punch drunk and sway limply back and forth for a moment before falling down
-get knocked off their feet by the final blow and land akwardly, dead
-If just a simple biped, you could dismember it's bones

There's hundreds of fun ways a character could die. Use your imagination and have fun with it.

A few classic deaths that pop to mind from movies:

-Fezzig (the cicilian) from "The Princess Bride"
-The six-fingered man from "The Princess Bride"
-Sgt Elias from "Platoon"
-Pee Wee Herman's vampire in the Buffy the Vampire Slayer movie (the original, before the series one)
-Boromir from "Fellowship of the Rings"
-Darth Maul from that really crappy movie with the kid and the dumb alien

Good luck.

[zenfull]

Thank you so much for all of your help switchblade, I guess I just get confused because my teacher says I should do one thing, but then again hes a Cinematic Animator for CG Films and such, and yet I want to work for games, and I hear other things so it becomes a very confusing line up. Though I will take your advice and make my reel a little more available, and if I have to just cut out the facial animation, but yes he said to stay away from the facial animation because they don't care for it.

As for the references you've given me, I guess I gotta go home and pop in my Princess Bride DVD. Thanks! I know that right away I probably won't be able to get a job working as an animator for Blizz, but I also know that I have a few extreamly strong feet in the door so to speak. I know a few people that work there that are willing to help me in many ways in getting into the company, though sadly no one from the animation department itself. Though I have made contacts from that department, so that is good.

Thank you agian, you are very helpful and informitive.

Kimberly

[switchblade327]

No problem. Glad to help.

Actually, I agree with your teacher though, if he means to focus on one discipline (animation, modeling, texturing, etc). As teams get bigger in games, they get more compartmentalized and people focus more on just one thing. Hence, why I think a well-rounded demo reel is not as valuable as a tight, focused one: I believe there are more jobs that want to see the latter since they're hiring an animator and not a generalist.

By making your reel 'available' I just meant you should make sure it's good for any game animation job you might find open; not just Blizzard's specific (and apparently rather weird) requirements. And I think facial animation is skill that's going to be more and more required of character animators for games in the future.

[gilley]

One thing that hasn't been touched on that's a real pet-peeve of mine after watching many demo reels...... Don't put motion blur on your animations. It blurs your poses and in my experience can be used to hide "pops" in your animation(sloppy animation). Most game engines don't put motion blur on their characters while fighting/running/etc, so don't do it either. You want your animation to look as close as possible to how it would look in-game.

And please don't try and do any kind of fancy lighting. I need to be able to see the entire character in a strong pose. I loooked at one reel the other day that had a single spot light on the character with a black background. When the character's limb moved out of the spot light's "cone" I could no longer see what that limb was doing......next reel please :)

No fancy models either, I don't want to see dynamic hair and cloth moving around. It looks really nice and all, but it can also distract your viewer so that they don't see the main character.

One example of some kind of throw attack with two characters interacting with each other in some way would be nice to see but not required. It's good to know if you can do these types of animations because they're ALOT more complicated than your standard navigation animations, etc. The animation doesn't have to be a throw like in wrestling(it can be if you want), but could be a quick "canned" sequence that plays when a button is pressed......alot of games have a throw button.

[Wabit]

With the next generation of consoles motion blur is now being implemented into the engines.



Ive never been to sure about the idea of making a reel for games companies full of boring cycles. If you can animate well to an audio clip, or show personality through your animation (which is by far more interesting) – then im sure its not to difficult to do a run cycle.

[eks]

Ive never been to sure about the idea of making a reel for games companies full of boring cycles. If you can animate well to an audio clip, or show personality through your animation (which is by far more interesting) – then im sure its not to difficult to do a run cycle.
Yeah, from the experience I have from both cycles and acting, it seems acting is a step beyond cycles. I have the same impression, if you can do a very expressive and interesting acting (as long as you have all the body on the screen and not only a sitting char with torso/head animated), cycles is a piece of cake.

But in doubt, I'll end doing both...

Btw, it seems much more difficult to have "top notch parameters" for an animation demo reel than for a modeling demo reel. I mean, with modeling you have a much broader grasp of reality to base your work (and to judge the work the HR will be seeing). You can see if that cloth volume was made correctly or not, or if the hair is painted cartoonishly or realistically. With animation it comes much more to taste, it seems way more subjective to personal view. Of course there are still some basic principles of weight and timing that you need to get correctly, but that is, well, basic. You can have nearly inifinte ways of walk and, if the weight is right, you can't really tell that one way is "more right than another". Any more thoughts on this?



eks
PS: And because of that last paragraph, I'm thinking about doing a modeling reel, since working with games is more important for me than working with animation...

[switchblade327]

Whether a game engine can do motion blur or not is beside the point that it can be used to obscure the quality of an animation. If it looks tight without motion blur, it'll only look better with it. I think it can be fitting for acting/short pieces that have nice models/textures/lighting (the kinds of projects that so many schools require as a final project) but it just doesn't belong in demonstartions of game cycles.

Besides, remember that the original xbox was supposed to be able render real time fur and the ps2 would have Toy Story resolution characters? True real-time motion blur is a campaign promise until it's in the box, as far as Im concerned.

As for the acting vs cycles, I mostly agree but game animation does have some unique requirements that performance animation doesn't quite cover, or even goes contrary to. For example, an attack animation in a movie must look cool and dramatic above all else. An attack animation in a game (for a player controlled character) must be responsive above all else, sometimes at the expense of a comfortable amount of anticipation. It must also return complete control to the player as soon as possible, sometimes at the expense of a comfortable amount of follow through. Likewise, the timing of an enemys' combat animation is going to be partially dictated by the intended difficulty level of the character by design, as far as how much time the player has to dodge/counter/otherwise respond. This will trump making it look 'perfect' at times.

This can't be demonstrated well on a reel, animation is often at the mercy of design, who sometimes don't have the best eye for it. Your art will likely get trumped by design needs at times.

In addition, game animators don't have the luxury of staging an animation for maximum effect. Game animation by nature breaks that principle and must be tight and clear from all angles.

Granted, these aren't massive hurdles for a skilled animator to overcome but I've heard stories of talented people struggling with the transition into this much more restrictive type of work. They're not as similar as you'd first think.

While performance animation isn't going to hurt a demo reel, a lack of showing targeting game animation I think could and if it is truly a cakewalk, why take chances? If you were applying for a texture art position, showing your beautiful oil paintings will help demonstrate your skill, but a potential employer will also likely want to make sure you can make things pretty at 512x512 too. Even next gen games will be restricted by technology and the nature of the medium so it wont hurt to show you can play by those rules.

[eks]

It must also return complete control to the player as soon as possible, sometimes at the expense of a comfortable amount of follow through. (...)
This can't be demonstrated well on a reel, animation is often at the mercy of design, who sometimes don't have the best eye for it.
I completely agree with your points about limitations on game animations. And that's exactly what I was mentioning, and is one of the issues of this thread, how do you show that on a reel?

For example, there's a crounch cycle on my reel that, previously, the pelvis was stiff. That was because of restrictions of the engine, we were still prototyping and I could animate the pelvis only on the Z axis. One of the critiques (http://forums.cgsociety.org/showthread.php?t=283178) I've received was about the stiffness of the pelvis, which I later corrected. Should I have done that or not?

Will a recruiter look at the stiff and see "ok, this guy did this because of the engine" or see "it's stiff, it's bad"? Or with the swing he'll see "this is lame, the pelvis can't move on all the axis on any engine, this guy can't animate for games" or "ok, its ok"?



eks

[switchblade327]

I completely agree with your points about limitations on game animations. And that's exactly what I was mentioning, and is one of the issues of this thread, how do you show that on a reel?

For example, there's a crounch cycle on my reel that, previously, the pelvis was stiff. That was because of restrictions of the engine, we were still prototyping and I could animate the pelvis only on the Z axis. One of the critiques (http://forums.cgsociety.org/showthread.php?t=283178) I've received was about the stiffness of the pelvis, which I later corrected. Should I have done that or not?

Will a recruiter look at the stiff and see "ok, this guy did this because of the engine" or see "it's stiff, it's bad"? Or with the swing he'll see "this is lame, the pelvis can't move on all the axis on any engine, this guy can't animate for games" or "ok, its ok"?

eks

All of what I said, said, you shouldn't hesitate to fix an animation for your reel. Game engines can be restrictive and different ones have different restrictions. No part of your reel should require an explanation as to why it's stiff or otherwise bad. In this case, I'd say fix it or toss it.

Some cases might be excusable, like partial body animations (Think Unreal Tournament or Counter-Strike; the upper and lower body pretty much operate independantly of each other to allow the player to run and aim in any, independant direction). This is something that if shown right, will pretty clearly be a restriction of the game engine/gameplay and not bad animation.

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.