<!--QuoteBegin-Mantrid+Jun 11 2004, 02:51 AM--></div><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Mantrid @ Jun 11 2004, 02:51 AM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin--> Because it looks better. <!--QuoteEnd--> </td></tr></table><div class='postcolor'> <!--QuoteEEnd--> Er... was that replying to me? If so, what the result looks like and how the programmer specifies that result are pretty much unrelated. OpenGL already exists to allow programmers to draw graphics; why not just extend OpenGL to provide this kind of stuff, rather than invent an entirely new library standard?
And i want this in a game, why? <!--emo&???--><img src='http://www.unknownworlds.com/forums/html//emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif' /><!--endemo-->
"Too much of a good thing is bad." comes to mind...
<!--QuoteBegin-SoulSkorpion+Jun 11 2004, 02:27 AM--></div><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (SoulSkorpion @ Jun 11 2004, 02:27 AM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin--> OpenGL already exists to allow programmers to draw graphics; why not just extend OpenGL to provide this kind of stuff, rather than invent an entirely new library standard? <!--QuoteEnd--> </td></tr></table><div class='postcolor'> <!--QuoteEEnd--> Presumably because OpenGL isn't really suited to this sort of thing, so they're designing an API (imaginatively called OpenRT) which is.
One impressive aspect is that it can use multiple computers to get things done in a real-time manner - rendering, say, a 20th of a frame isn't necessarily too difficult, but when that 20th of a frame might take significantly more or less time than the rest, things become more complicated. Perhaps the system automatically balances stuff out, so one computer might render a couple of difficult pixels while another renders a load of easier ones...
Then you have antialiasing, with multiple samples for certain output pixels. Fun!
Comments
Er... was that replying to me? If so, what the result looks like and how the programmer specifies that result are pretty much unrelated. OpenGL already exists to allow programmers to draw graphics; why not just extend OpenGL to provide this kind of stuff, rather than invent an entirely new library standard?
And i want this in a game, why? <!--emo&???--><img src='http://www.unknownworlds.com/forums/html//emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif' /><!--endemo-->
"Too much of a good thing is bad." comes to mind...
Presumably because OpenGL isn't really suited to this sort of thing, so they're designing an API (imaginatively called OpenRT) which is.
One impressive aspect is that it can use multiple computers to get things done in a real-time manner - rendering, say, a 20th of a frame isn't necessarily too difficult, but when that 20th of a frame might take significantly more or less time than the rest, things become more complicated. Perhaps the system automatically balances stuff out, so one computer might render a couple of difficult pixels while another renders a load of easier ones...
Then you have antialiasing, with multiple samples for certain output pixels. Fun!