May Contain Nuts

From Joystiq:

Who says mo-cap is the technical capturing of data to create realistic, fluid motion in gaming that’s devoid of soul?

Anyone who knows anything about how to properly animate a skeleton? Seriously, compare Polar Express to any film made by Pixar (even the original Toy Story. I’m that confident about this). The difference is mind-boggling. Pixar even went so far as to put a disclaimer at the end of Ratatoullie stating that absolutely no motion capture or other production shortcuts were used in the making of the movie. Animation is inherently art, and while you can use motion capture as a basis for your intentions if you’re under some serious time constraints, there is nothing motion capture can do that a trained animator can’t do better. Nothing.

Now, as I said, if you’re under pressure to get all the character animation in a game done in an extremely short time frame (especially when you’re aiming for realistic and unique movements, like in sports games, where each player has their own way of moving), you’re likely going to need to fall back on motion capture just because studying that many people’s movements is a lot more time consuming than just having them come in and strap into a mo-cap suit. However, if Polar Express has taught us anything, it’s that un-altered motion capture is creepy. Just watching the in-game footage that’s spliced rather spastic-ly into the end of the video embedded in the linked article, the players look pretty good, and move rather well, but they just strike me as slightly undead. It’s the Uncanny Valley all over again, but while most people tend to focus on the visual appearance of these creepy specimens themselves, the way they move is just as important. Motion capture seems to be a really good way of quickly moving from the null end of the Uncanny Valley graph straight into the deepest part of the trough; it’s what the artists and animators do with that raw footage that ultimately determines whether their characters manage to crawl back out of that pit, but putting your full faith in the motion capture to solve all of your problems is a really bad idea.

So to answer Joystiq’s query in a less sarcastic manner, the people who have only seen un-edited mo-cap are likely to assert wholeheartedly that it’s quite devoid of soul. Well-utilized raw mo-cap footage, on the other hand, may be indistinguishable enough from hand-animated motion that for all intents and purposes, you could either suck the rest of the soul out of it and hit the wrong end of your target, or you can punch it up a notch and hopefully strike the sweet spot of realism/stylization and interest/creepiness.

So I was bored at work yesterday and decided, for the sake of learning MAX’s Particle Flow system again and also to see if a certain idea I had would be feasible, to create a rudimentary cannon using nothing but a cylinder and some particles.

The results are here.

It’s in the air…

There will be major changes to both the Realm and Revision1.com in the next couple of days. What started out as a time-killing project at work one night has evolved into a full-blown dual-site overhaul. My portfolio site has been in serious need of some serious loving for quite some time now, and it’s finally going to get the attention it deserves. The Realm, on the other hand, is something I’ve been wanting to re-skin for the hell of it for a while, because I dislike using other people’s skins on my sites when I’m advertising my services as a web designer. It just looks bad.

These changes are part of my effort to complete as much of my long-standing-projects-to-be-done-sometime collection as possible before we throw caution to the bitter cold wind and move to Washington at the end of the month.

For those curious, this means that unless something miraculous happens, DPWR probably won’t be upgraded until after New Year’s… Invision Power Board 2.2 isn’t out yet, and there’s an additional delay after the release during which I have to re-code the DPWR site back-end and integrate the new Archive system before anything can go live. I was hoping to launch the new site before Live was released, but that is increasingly looking like it won’t be the case.

I’ve got new artwork done, but because I’m lazy and need to go to work, it’s not on the Realm’s gallery yet (I’m getting kind of tired of maintaining 3 galleries, but oh wells, the portfolio gallery isn’t for just anything, and HBI is). I do have it in my portfolio gallery, however, so you can check it out here. It’s the two iPod Mini renders. I’m really incredibly proud of them.

Also, in two weeks I will have been at AMC Theatres for 4 years. This depresses me. I got my 4-year pin in the mail today, and while I appreciate the small humorous irony in the pin being in the shape of a projector (1 year is popcorn, 2 year I forget, 3 year is a film reel (which I received with a sympathy card, so that at least made me chuckle)), I really didn’t want to be here this long. I have a potential job offer, but I’m third in line for it and am arguably the least qualified of the three people applying. I’m supposed to hear back by the end of the week, but I’ll give it to the end of next before I start giving up on the idea. If it falls through, something drastic is going to happen, because I’m sick of being stuck here. I’ve actually been encouraged to pick up and take off if this new job doesn’t pan out. I have something I can do where I’m going (and I’m being intentionally vague because I’ve been asked to be), but it will require a second job to pull it off, and AMC has so far been slow beyond the point of heel-dragging in figuring out if I can transfer to a theatre there, so I’ve all but given up on that happening. I’ll get a job at a restaurant if I have to, and the fact that I’ve publicly stated that I will never do that again should indicate how desperate I am for a change of pace and scenery at this point in time. It also will mean probably leaving Osc here, something I hate doing because I know she wants to get out of this house as much as I do, and because it’ll mean being apart for at least a couple of months. Still, I need to do something, because I’m not going to stick around here much longer. I don’t think I can do so and stay sane (or at least, motivated).

So, after being told by Josh that, as I suspected, I needed to work on my texturing and most especially on my lighting, I re-opened an old scene I originally built at ITT, and then re-textured after graduation, and began revamping the lighting. This time around I had the advantage of having a reference photo of booth to work from. So, here’s the new new Booth scene… some textures are missing, and I’ll be plussing the modeling sometime later, but here ya go… lighting nearly complete, but I wanted to get a shot up: New New Booth

The gallery here at the Realm is currently experiencing technical difficulties, so I’ll just point you all to my portfolio site, where I’ve uploaded a few WIP shots :).

Hooray for low-poly work.

Okay, I know I said HBI would be ugly and blue for a long time to come, but I decided to upgrade the thing to the Gallery 2.0.2 script, because it was running on a version that was over a year old and a real pain in the butt to manage. Not saying the new version is very much easier (still don’t know how to edit the language files without fighting with Terminal to re-compile the damn things), but the templates make a slightly larger bit of sense and I was able to get the whole thing wrapped in a Blog-like interface, so there’s at least a marginal amount of consistency now. Whee. I even managed to keep all of the old album data, so huzzah.

And now, I need to go and die for a while.

I’ve been thinking about this a lot lately, and I think it’s finally time I started putting these thoughts down in a more permanent form. I’ve been trying to figure out what makes certain games and CG movies visually appealing, while others come off as either flat (Unreal), full of super-creepy zombie people (Polar Express), or just lacking in effort (Madagascar).

I think the ultimate problem is that people are pushing too hard for photo-realism, and because the technology isn’t there (and it probably won’t ever be), the people who shoot for it in its entirety end up making something that looks uninspired and boring at best, and incredibly creepy (in the unintentional “it’s not supposed to be creepy!” kind of way) at worst. The reason for this being that we live in a photorealistic world, and we know what it’s supposed to look like. So, when a game (or even a CG film) comes along and tries to pass itself off as being completely photorealistic, we pounce on any little thing that could possibly be wrong with it, especially when it comes to human characters. It’s not that we do it intentionally (often times we don’t even realize we’re doing it), it’s just an innate behavior for whatever reason. Another part of the problem is that if you *do* manage to attain a close enough semblance of photo-reality that you overcome the gut instinct that something’s not quite right, your scenes often come across as visually quite dull. A bland cityscape with a boring blue sky and unremarkable clouds and flowerboxes isn’t going to get you very far in the “interesting” category. While such things have their application - especially as effects shots in live-action films - I don’t think games are the best place to apply them.

So, the question now becomes, if we can’t do photorealism (or at least, if we want to be high-detail without falling into the uncanny valley trap), what can we do? Well, the first solution would be to take it to the other end of the spectrum and create highly-stylized scenes and characters, which is what I think Dreamworks tried to do with Madagascar. However, lack of effort, decent story, acting, and a complete inability to rig a character decently (seriously, I could probably rig the characters just as well as they did, if not better… I just suck at thumbs) really damaged the film’s appearance. I’ve got more to bring up on why Madagascar in particular is a disaster of a style choice, but I’ll get to that later.

If you don’t want to stylize things and you still want to retain the same level of high detail and texture that you find in the real world, the last option, I think, is to create a sort of “optimized reality”. It’s a sort of expansion on the concept of surrealism introduced in Myst, where things often looked, sounded, and felt more realistic than they actually were. It’s an incredibly fine line to ride, but interestingly enough, there’s a considerable amount of play in what you can get away with under that style. What it boils down to is this: take reality, and cut away everything but its essence. You’re not translating an object into a game or film, you’re translating the Platonic ideal of that object into a game or film. You don’t just build a “house”, you build a “House”… the ideal of a house that may not even be quite as detailed, ornate, or intricate as an actual house, but which carries a vibrance of shape and color you’d never want to see on a house in the real world, but which surprisingly enough happens to work better than a plain old house when used in a game or all-CG film.

It’s a long-standing principle of animation that you don’t draw the totality of a person or object, you simply draw the necessary elements of that person or object needed to get across to the viewer that what they’re looking at is the person or object in question. I think this concept still very much applies to CG films and computer games, as they exist outside of the mold of live-action composites, which by definition have to fit the very realistic appearance of the actors and other real set pieces in the shot. Interestingly enough, Myst is one of the few series I’ve encountered that actually pulls off live-action compositing with “optimized reality” scenery, because real actors tend to stand out against scenery that doesn’t look as real as they do. Riven, obviously, is in a class of its own in terms of photo-realism, but even Riven had an optimization to the reality of the scenes that gave it a sort of elevated sense of design and order. The real world lacks such design and order, but without it in a game or film, the scene fails to convey any emotion or generate any connection with the viewer. It’s odd. Anyway, back to my point. “Optimized reality” very rarely lends itself to photo-realism when it comes to interactive environments. In that particular instance, there’s a lot less you can still get away with, because interacting with an optimized reality betrays the lack of complexity that actual reality has, at which point the illusion falls down like a house of cards. It’s for this reason that I think Edanna (and even haven to some extent) failed for me as a concept. it’s a great concept, but it relies on utilizing systems that are far beyond the ability of computers to replicate, and interacting with their optimized forms creates a very mechanical-feeling experience, which you definitely don’t want from a nature-based puzzle.

I think Cyan and Pixar are leaders in their fields when it comes to creating believable “optimized realities”. Their products have always portrayed reality not as it actually is, but in a very visceral form. What you get from Myst, Uru, End of Ages, Finding Nemo, and The Incredibles is a lightly stylized view of nature, with colors, forms, shapes, and scenery that couldn’t exist in reality, but which balances stylization, realistic surfaces and sounds, and a dash of artistic know-how in just the right quantities that you’re left with a vision of a world in its most pure form; a form which, when you think back on it, your mind fills in the blanks for, but which, when actually analyzing it, fails to stand up to the rigors of actual reality. The trick, then, is to look beyond the modeling simplicities, the simplified forms, the over-vibrant colors, and let the world soak into you, so that when you revisit it in your mind, you don’t remember the polygon count on a sphere, but rather what was on the sphere and the room it was sitting in. Obviously, this artistic style doesn’t work for everyone, but it definitely works for me, and it’s the sort of thing I strive to do in my own work.
And having gotten to the end of my point, I realize I never did go back into how Madagascar is a fiasco of a film from numerous aspects, so I guess I’ll save my rant against the Dreamworks Feature Animation department for another time.

I am continually bemused by the people who comment on someone’s art (mostly on deviantArt… forums tend to get less of this behavior) with what basically amounts to “you should draw more stuff for free because I said so.” Usually these take the form of an otherwise innocuous request (like “could you give me a size-comparison between these two characters?”), but are phrased in such a way as to make it abundantly clear that the commenter demands that you do as they wish, or you’ll be late (as in, the late Dentarthurdent… *cough*). They typically begin with the words “you should…”, as opposed to “could you…” and involve no question marks, simply statements of something that you, as an artist, must do because someone commenting on your art said you should. I mean, it’s not like you had anything else to do anyway, right?

Yes, 19 days after Cyan put out their call for new artists, I’ve finally cobbled together what I can definitively call the best of my 5 years of stumbling around in MAX trying to teach myself how to use it, plus a couple of new things I put together because I didn’t think what I had was good enough as it was (I’m paranoid like that…).

So, now all that’s left is to get my resume put together as a .pdf for download in addition to the version already on Revision 1, and make a couple of edits on the website itself to make the 3D stuff easier to find, and I’ll be ready to send an email off to Chris saying “hire me pleeeeeeeeeeeeeeeeze!”, only in much more formal words than that (really though, that’s about what this boils down to).

For those who want to see my demo reel without having to go hunting for it, you can get it here. Please save it to your computer, as it’s 7 megs and I don’t want to blow up my site before Cyan gets here <_ <.

This is officially the shortest entry I’ve made in probably the last 3 months…