next »

[gimymblert]

Game animation also suffer from past craft illusion, an animation like aladin would always seem like an improvement (even if it may does some regression) because it break expectation with shiny new stuff. But may look pale and crude regarding the whole art principle.

It bug me because as a child i was looking at disney movie frame by frame, and if i could, i would do the same with video games (sonic let you frame advence with cheat code) generally by recording them. Through observation i get the principle of both. I could see some game were good at mixing both world, and that some game fails. I have a lot of idea to make truly alive animated game but it mean adapting gameplay, sometimes with paradigm shift.

For exemple something like the high profile mode in assassin's creed could create a true breakthrough by making anticipation interesting from a gameplay standpoint. I'm working on a concept platformer game like that in 3D, but yet have to find a core interesting gameplay that goes with the imposed universe (something with "7 matching power" as the gimmick is "dominoes")

[ChevyRay]

would the still jacket texture work in a game? I'd imagine that would maybe need to be done procedurally, or move backwards at the same speed the player moves forward in order for it to keep its rad stillness.

Yeah, it wouldn't be too difficult with some simple masking method I imagine.

[TooMuchSpareTime]

That running spacebear looks pretty sweet. The "moving up and down" issue is only because there's no background - it'd look fine in a game if the ground was a sort of slightly-above view rather than just a flat horizontal line, I guess.
2D animation in games is generally okay, even if it looks a bit stiff or uninspired at lot of the time. 3D's where the problems seem to be. 3D animation in movies has improved in leaps and bounds in the last 10 years, and nobody can get away with making a big movie with animation that looks like "Beast Wars" ("Beasties", in Canada) anymore. But that level of animation quality seems to have remained generally acceptable in videogames. Not to say that it's not improving at all, though.

Interesting walk cycles are great for games. They give the character so much personality, and remind the player of that personality the whole way through, like Cool Spot's over-the-top swaggering strut, or Laverne from Day of the Tentacle's airheaded flouncing.

[Greg Game Man]

its about how the thing feels to play; not how technically brilliant the animation is

[gimymblert]

It's not even how technically brilliant an animation can be, but how much it fit a "scene".

Take the basic walking animation, no matter how good it could be, the fact it does not change to fit emotion of the context made it "off" and only functional >> "display a state". Worst, the more "technically" brilliant it may be, the more it would draw attention into itself and jar with the gameplay. A subtle balance is to be found.

Now even if the character react "emotionally" to situation with things like blended and procedurals animations (gaze, fatigue, facial animation, feet adjustment, etc... think shadow of colossus), the problem would remain of anticipation and purpose. With the random walking promote by current way of navigating in game, animation would always stay a reward gimmick. Also because good animation is about deeper psychology revelation of a character, the way control are about reaction does not promote that (see anticipation and purpose). Last problem is gameplay, the more blended and adaptable are animation the less easy it's to read game state, animation is one of the main channel to read player's state, "good animation" (fluid and blended with lot of details) just blur that.

For animation to be "really good" and "less gamey", we need to rethought fundation of how we interact within game, most notably navigation, purpose of action, emotional and psychologic interface (intent), "social aware" gameplay. And less "dead time" where nothing happen.

Heavy rain tried just that!

[Greg Game Man]

It's not even how technically brilliant an animation can be, but how much it fit a "scene".

Take the basic walking animation, no matter how good it could be, the fact it does not change to fit emotion of the context made it "off" and only functional >> "display a state". Worst, the more "technically" brilliant it may be, the more it would draw attention into itself and jar with the gameplay. A subtle balance is to be found.

Now even if the character react "emotionally" to situation with things like blended and procedurals animations (gaze, fatigue, facial animation, feet adjustment, etc... think shadow of colossus), the problem would remain of anticipation and purpose. With the random walking promote by current way of navigating in game, animation would always stay a reward gimmick. Also because good animation is about deeper psychology revelation of a character, the way control are about reaction does not promote that (see anticipation and purpose). Last problem is gameplay, the more blended and adaptable are animation the less easy it's to read game state, animation is one of the main channel to read player's state, "good animation" (fluid and blended with lot of details) just blur that.

For animation to be "really good" and "less gamey", we need to rethought fundation of how we interact within game, most notably navigation, purpose of action, emotional and psychologic interface (intent), "social aware" gameplay. And less "dead time" where nothing happen.

Heavy rain tried just that!

wat

[Xion]

makes sense to me.

edit; though I disagree.

edit; though I agree, to an extent.

edit; but I mostly disagree.

[gimymblert]

I also disagree with the implicit statement that "good animation" is "theatrical" animation only

But that was implicit in the post i was responding to (post after i had define good animation for game with megaman and SMB1 one as exemple). Actually it can be read as a statement against theatrical quality animation (unless certain condition) as it is at odd with game principle (especially when not taken into account).

There is still a lot of improvement we can build with theatrical animation principle in mind and also because all gameplay are not the same, and new one need to be create. There is a lot of innovation to make in term of transition for example, there is lot breakthrough to do.

And new principle of animation along theatrical quality could help define new way of thinking gameplay (by adding relevant emotional feedback and system) even for a simple plateformer (don't ask our experimental game so far, they are stuck in gamey gamish usual stuff).

But if i continue i will become WAY to subtle and messy, nobody could follow the trail of mind (since i also have some idea in mind in term of controls and animation).

[HöllenKobold]

Fighting game animation is probably more theatrical than animation in other genres, but they usually have a greater emphasis on controls and "thinking gameplay."

[gimymblert]

yep but only when "ultra" which is non interactive.

Game like sf3 or marvel vs capcom 2 have pretty good balance but notice how the "scene" is fixed in tone and emotion. And still does not compare to any low quality action flick movie fighting sequence. But still very high quality video game like animation.

The best moment in fighting gaming justin vs daigo Evo2K4 expose the shortcoming, the fight look nowhere like real fight.

But we are SOOOO off topic now, which is not shortcoming of game animation compare to theater one, but how to improve indie animation!

[baconman]

I'm still saying it's all a matter of appropriate pacing/framerating. "Functional sprites" can be replaced at any time with "theatrical" ones, but only if the framerates and dimensions line up correctly. Fighting games are considerably more flexible, as most of those (even SF II) are coded to contain and overlap limbs, heads, and bodies as independent sprites; whereas most other genres of gaming (platform, RPG, etc.) are fully-comprised body-sprites.

MegaMan: The Wily Wars for Genesis illustrates the concept pretty nicely - the graphical updates are really good-looking, and the games retain functionality for the most part. But then there's a few gameplay elements that were altered to fit the extra animation, most notably the Magnet Missile; a modification which proved to be gamebreaking.

Moving SMB's simplistic mechanics and animation to the SNES in Super Mario All Stars proved to be very smooth, however. So long as animation improvements don't destroy the gameplay functionality and pacing, it's all good.

[The Monster King]

speaking of animation from fighting games, i think Guilty Gear series really has a lot of things to teach about animation! I was really surprised to find out there were "so few" sprites in most moves since the animation looks so fluid!

by contrast KOF series tends to have slightly less good animation but it doesn't show because it has 12000000000 frames of animation

anyways what is important to remember from this is that you can do really good with less frames as long as there is an illusion of movement!

[Richard Kain]

Part of the problem is that animation is one of the most time-intensive tasks that an indie developer can work on. It also requires a decent amount of training, practice, and talent. Animation tends to be a very artistic discipline. Coders who are used to pure logic don't tend to be as good at animation. (and art in general) There are a few exceptions, but the overall trend holds true.

For most indie developers the solution would be to focus on a more technical art style, and implement animations through code. This would play to the strengths of the coders developing those indie titles. The title "Monaco" that won the Seamus McNally award this year seemed to take this approach. It could have focused on much more traditional hand-drawn animation. Instead it focused on what the developer was good at, and used its art more for player feedback than artistic expression. A focus on the core strengths of the developer payed off with a solid title.

By the same token, developers who are more artistic than technically focused should work on games with fairly simple mechancics but great art and animation.

[Chris Whitman]

Animation tends to be a very artistic discipline. Coders who are used to pure logic don't tend to be as good at animation. (and art in general) There are a few exceptions, but the overall trend holds true.

Honestly, I think people cripple themselves with this kind of reasoning.

It's a bit like saying doctors can't be good at cooking, or golfers can't be good at knitting. The idea that human thinking is split into creative or rational pursuits is just fallout from Romanticism, and has little neurological basis.

[Seth]

I agree with Chris, I think usually the whole "coders aren't good at art" and "artists aren't good at coding" or whatever most often comes from the fact that coders usually care about and practice coding a lot more than art and vice versa for artists

[gimymblert]

And also there is a lot of reasoning in making animation (or art in general). Art is heavy under rules and procedure.

[Bree]

Okay, reviving this thread because I'd like to bring up an alternative: Limited Animation.



Seriously, if animating characters smoothly is a problem, why not go in the complete opposite direction? This doesn't mean you can just cop out and do stunted work. Rather, your efforts are focused on making the individual keyframes look as good as possible, and the hardest part is learning how to use them efficiently. Flash animation naturally lends itself to this method- entire walk cycles can be saved for later use, pieces can be rotated and moved with minimal effort.

the UPA cartoons are known for pioneering this style as well. So long as the creator has an eye for design, the results can look just as aesthetically pleasing as any big-name production.

[Shaun]

The problem with limited animation is that in cartoons there are a number of techniques used to make up for a lack of animation, such as an abundance of dialogue or fast editing, that might not transfer well to video games. That being said 2d video games don't require the level of visual immersion that cartoons do in order to keep the audience interested, so I'm sure there are ways to go about it. It would actually be pretty interesting to see it done well.

[jonschubbe]

Part of the problem is that animation is one of the most time-intensive tasks that an indie developer can work on. It also requires a decent amount of training, practice, and talent. Animation tends to be a very artistic discipline. Coders who are used to pure logic don't tend to be as good at animation. (and art in general) There are a few exceptions, but the overall trend holds true.

I'm pretty sure Andy has a degree in Fine Arts as well as computer science stuff. I could be mistaken though

Also, i think the main deal is that it's time consuming to do the animation. If there are multiple characters, all of them run at high frame rates, and want to look equally smooth compared to the rest of the moving objects in the game, it's just really time consuming to redraw and polish the drawings for every frame.

There IS a point where you lose the flow of interactivity-to-animation ratio, but i think the main problem is that getting an expressive and in some cases not-cartoony animation correct takes multiple attempts sometimes. It's just really time consuming for me. It takes a ton of experience getting everything right quickly.

[Zaratustra]

It's also very hard to coordinate an artist and a programmer together to do decent 3d art, let alone 2d art. At least in 3d the artist won't give you two animations, one of which has the character 25% larger because he set every frame to fit on a 256x256 square just like you asked.