yeah I read this too -- it's a critical argument I think, and one lost on the majorty of the gaming community. at gdc, it was interesting to watch warren spector back out of his love story presentation, citing that humans could never emphasize with virtual characters until they were at the appropriate level of reality. mori presents the alternative, and one that I think holds a lot more water.

Scott McCloud has a great theory about abstraction, realism and iconography in 'Understanding Comics', and why iconographic faces attract empathy from the reader.

That's why Tintin is such a simple character, in richly detailed environments: allows us to become the protagonist in the story world.

I totally agree. The more realistic the game graphic looks like, the more attention will be dropped on the comparing between reality and game world. However, this process won't last long if a game succeeded in limiting the rules and letting the player realiizng its abstract and unrealistic rules are the game-play. To keep people confusing between real and game is the key here. Additionally, racing and simulation game do require realistic graphics, I don't think a nearly realistic car is something people don't like.

OK, an article like that just gets me started on a roll...

The article states something that I've been thinking about over the years as well. There's a gap when elements (especially humans) are almost realistic. Comparing these entites with corpses seems particularly apt...

As Jenova points out, objects like cars are different in this respect...I suppose a car is less complicated (in terms of how we identify it and, more importantly, identify with it). Enviornments are similar as well...we'll probably be able to render photo- believable terrain and structures a lot sooner than humans.

Another key part here is _motion_. The Final Fantasy movie humans were almost perfectly photoreal...if one looked at a still image. They still didn't move or talk correctly. I think if you can get animation working well, you can get away with a lot less "static" detail...an eye watching something moving (as long as it moves believably) is very forgiving of details.

It also seems a lot of the "Ken and Barbie corpse" models forget the animated laws of squashing and stretching...things real objects don't do, but nevertheless are almost required if you want to animate objects. Classic animation (Warner Bros. Disney, Fleischer, etc.) did this all the time...it doesn't normally look right if you don't. An ironic case of exaggeration required for "realism".

If people are waiting for facial animation before trying to get an "emotional" game, I think they're barking up the wrong tree. People are perfectly capable of identifying with an abstracted character...think of how many people guffaw at bugs bunny, or how many children were scared witless watching any number of classic Disney cartoons (my mother tells stories of hiding from Monstro the Whale from Pinocchio).

Nevertheless, people seem to get used to just about anything like this, and the "almost human" style is now established enough to make fun of. A lot of Machinima animation is humorous, and a large element of the humor is making fun of the grotesque movements and mannerisms of the "replicant" style people described in the Slate article. Red VS. Blue wouldn't really be the same if it were animated in a way that was, well, more animated. The goofy run cycles and semi-canned reactions add to the irreverent humor.

What's tempting, though, is that graphical break throughs are so close to getting photoreal imagery, a lot of people think the goal is in sight. Breakthroughs in lighting, shading, not to mention geometrical trick all give more graphical power that would have been unthinkable five years ago. If we're almost photoreal...why not go the extra 1%, even if that last bit is the hardest? While that's cool, it's not something I, personally, want to develop. I want to look down some of the side roads that digital/computer imagery could provide.

So, in a partially unrelated conclusion, I hope someone is figuring out different ways to use the 300 million polygons per second we'll be getting on future graphics cards...there's so many other things you can do besides rendering plastic, puppety humanoids.

By the way, is it a rant if you totally agree with the article?

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

URL:
Remember me? YesNo

Comments: