Wednesday, August 3, 2011 7:38:50 PM
i could see this being an issue in a game environment when the game engine designer have to figure out collision. It's easy when you just compare polygons, but compare millions of atoms to millions of other atoms. This won`t be easy to integrate into games, but it eventually will be done, and I for one can`t wait cuz it`ll be awesome.
Wednesday, August 3, 2011 5:03:24 PM
[quote">So from what I see, this works great for fixed scenery, and poorly for animation. Perhaps game designers should go to a hybrid technique where backgrounds are rendered this way, and all moving objects are added in with conventional polygon rendering.[/quote">
This isn`t a new idea and it isn`t unlimited no matter how many times they say it is. The hybrid approach also has problems, which is why it isn`t used more often despite the fact that it`s been around for 20 years.
If this bunch have some secret new techniques that nobody else has managed to think of that solves problems that appear to be impossible with current technology, they`ll need to produce some evidence of it to convince anyone other than co
Wednesday, August 3, 2011 4:54:02 PM
uunxx, they just haven't got the resources on the design side so repeating is easy to show lots of a large landscape but they are individual objects being handled as such, if they were unique it would be no different.
Yes it would. The most obvious difference is the amount of data required. You can`t use procedural generation for a whole gameworld, it just doesn`t work at all realistically. So you must have some data and the more unique objects you have in the gameworld the more data you must have for rendering them. Data that you need to store, data that you need to transfer between storage and processing, data that you need to process. It rapidly exceeds the abilities of current technology by multiple orders of magnitude.