CompuServe Messages

>I do not disagree with your analaysis Alan, there is indeed a lot of really cool >work done on other platforms, but to pretend that the first generation of real time >3d acellerators will make the PC comparable with an SGI box is I feel a bit >premature. I will use a case study to show my point. My former employment was with a phone company who wants to be in the interactive content market. Right now they aretrying to make a MYST style game. So the play does not need real time acceleration. They bought 4 Indigo 2's, Alias Power Animator, and hired a Unix guy just to help them get arround. At the time the investment made sense, because they want to develop quickly to get to market. But after being at least $180,000 in the hole, they better hope for a few good games to recoup thier investment. So far thats not happening (no fault of the SGI's. Dumb iedas and bad at making deadlines) But does it not make sense that for the same price a company could buy three times as a many pentiums. Model scenes in realtime. And while they are rendering, you could work on another machine. All for less than $180,000. Now what about real time games. According to Lockhead Martin games like Virtura Fighter, Racer, Cop, are rendered at 300,000 polyons real time. L-M calims their Real 3D chip can do 700,000 textured poygons real time. Almost half of what you quoted me about an SGI. If thats what a pc can do, then why should it not be able to invade territory formilly dominated by the SGI boxes. Yea, SGI will always out preform, but you alway have to justify your cost and SGI's may not always be cost effective with this kind of competition. In one sense, who ever thought 10 years ago IBM would lose market share ( maybe a bad senario but makes my point). I just think that not everyone can justify the fastest machine out. Sometimes good enough will pay even better. Thats the case I am making about approching SGI territory. Thank you for your comments, it been a good discussion so far. Alan