Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!hsdndev!wuarchive!julius.cs.uiuc.edu!apple!vsi1!zorch!xanthian From: xanthian@zorch.SF-Bay.ORG (Kent Paul Dolan) Newsgroups: comp.sys.amiga Subject: Re: videomaster 32 (was Re: 24 bit color output) Message-ID: <1990Dec11.053730.17869@zorch.SF-Bay.ORG> Date: 11 Dec 90 05:37:30 GMT References: <9613@pasteur.Berkeley.EDU> <1990Dec10.082400.16227@zorch.SF-Bay.ORG> <14873@sdcc6.ucsd.edu> Distribution: usa Organization: SF-Bay Public-Access Unix Lines: 79 rblewitt@sdcc6.ucsd.edu (Richard Blewitt) writes: > xanthian@zorch.SF-Bay.ORG (Kent Paul Dolan) writes: [stuff] > I called PP&S this morning for some information on the videomaster 32 > board, and got this information: > True 32 bit double buffering system, with its own vram (80ns) 2 or 8 > MB configurations Hmm. That makes the previous "you only get one buffer with 1 MB" datum even more muddled. > TI graphics processor and math coprocessor 40Mhz, 0 wait Maximum > output 1024x1024 (up from previous estimates) So perhaps with a compatible monitor, one could actually _view_ the whole buffer? A shame they didn't go for 1280x1024, then, to be a little closer to usual screen form factors. > Real-time video manipulations (rotation, resizing, mapping to a surface) Sounds like you could get a pretty nice animation running in real time, then, within the onboard cpu computing limitations. Yum! > TIGL Texas Instraments Graphics Language I suspect this is not too far from one of the international standards, GKS, PHIGS, or PHIGS+, or perhaps the non-standard but popular Dore; but it would be better to know than guess. Anybody know? > The graphics speed was not official, but they said 32,000 > polygons/sec. I have asked them to post an official number when they > get it. OK, so you could render a 1000 polygon image all moving in real time; you can do a _lot_ with a 1000 polygons. Not bad. This assumes that other overheads don't throw most of that raw speed away. > The bad news, not available until march, and the price will be about > 2500 - 3000 for an 8 meg configuration. Ouch! That's going to have a lot of trouble competing with the Toaster at half the price, which will have quite an installed base by then (whatever the relative merits of the two systems). Amiga slots are still a scarce resource, so few will buy both, and pricing piece parts that high, you end up with a $20K - $30K system, which is not going to compete too well with the established $50K TV station equipment these Amiga systems are trying to replace. Those systems have big installed bases, trained operators, (I presume) good software and aftermarket accessories, and so on. A two to one price advantage is going to have trouble overcoming some of Commodore's image problems in the commercial world; ten to one was a lot more effective. >> ... no Amiga display has 16.4 million _pixels_, so no Amiga (monitor) >> display can now put that many colors on the screen at once. > I have not seen any system with a 4000x4000 screen. Ok, I haven't seen > pixar, but I have seen Silicon Graphics VGX system. I'm not sure I have either, but some of the film recorders do 4K x 3K, so that much storage capability does have a place in the multimedia world, even if it can't all be seen at once on the glass eye. Given a place to store the data for manipulation, there is a mechanism to display it. > BTW 32 bit double buffering is significantly better than most of > Silicon Graphics options, which tend to use 24 bit, split into 2 12- > bit screens for double buffering. Yeah, that part was extremely impressive. Thanks for taking the trouble to learn and share the additional information. Did you get any clarification on the question of whether the video memory and the local program memory are separate or shared? Kent, the man from xanth.