Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!usc!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!crdgw1!millerjv From: millerjv@rigel.crd.ge.com (Jim V Miller) Newsgroups: comp.sys.amiga Subject: Re: Commodore & Universities Message-ID: Date: 27 Jul 90 20:45:14 GMT References: <25432@snow-white.udel.EDU> <46200106@uxa.cso.uiuc.edu> Sender: news@crdgw1.crd.ge.com Reply-To: (Jimmy Miller) Organization: GE Corporate Research & Development Lines: 73 In-reply-to: ragg0270@uxa.cso.uiuc.edu's message of 27 Jul 90 14:58:00 GMT In article <46200106@uxa.cso.uiuc.edu> ragg0270@uxa.cso.uiuc.edu writes: From: ragg0270@uxa.cso.uiuc.edu to on the whole campus. Faculty use the machines they used as students (be it Macs or IBMs) and are only looking to move up to Suns or SGIs. You will have a hard time selling them a different machine that does not offer a significant increase in computing power. Richard I don't know about your professors, but almost all of mine had finished school before IBMs, Macs, and Amigas saw the light of day. If they wanted to use the computers they used as students, we'd still be using punch-cards.:):):) I think in general they will pick the machine that a) Can do the job. b) Is in the right price bracket. c) Is a stable machine (hardware, software). . . . z) Is the neatest thing around. I would love to see labs full of Amigas, but IBMs and Macs are "safe" purchases. There are few risks and they can usually do the job for the price. (Obviously this is not the case for some fields of study like High Performance Graphics). I ask alot of my computers, whether it be my Amiga or the lab's $100,000 SGI. I can bring any computer to its knees with a little imagination. I also know a fair bit about building computers so I am perfectly comfortable with the fact that computers do crash (in fact I have yet to meet a computer I couldn't crash, not merely core dump, but CRASH). But I have to admit, the Amiga is one of the easiest machines to crash. Of course this is because to have the total freedom to do what you want. Personally, I like this idea. It doesn't place unnecessary restrictions on my productivity. But I know some professors that won't touch a machine unless it is extremely stable. If you really want to get Amigas into hi-tech colleges (and maybe colleges in general), you are going to need virtual memory and a memory protection scheme. Of course I have an alternative motive for suggesting this, my research involves accessing huge volumes of data (a 13 meg data set is very very small). Now we have a Stardent GS2000 in the lab with 128 meg of RAM plus disk swap available so my research runs fine on this machine. However the GS2000 is extremely slow at allocating memory (why?, don't know). Anyway I would rate this machine at 5 MIPS. My A2500/30 is probably around 6 MIPS. I would therefore love to perform my number crunching on my Amiga but I don't think I want to spend over $1000 for RAM. Plus I don't think my 2500/30 can access this amount of memory. My solution: run my research on the SGI (~20 MIPS). Of course the other thing keeping my research off the Amiga is that after I access this data for a few minutes, I generate from 100K to 10 Meg of geometry that I would really like to visualize. Unforunately, I need a 24 bit system to determine if my geometry (between 4000 and 24000 polygons) is "acceptable". So I have to keep using these $100,000 machines, even though my Amiga has plenty of horsepower to do the work I need. Such is life. (Did I have a point?) -- Jimmy Miller General Electric Corporate Research and Developement: millerjv@crd.ge.com Rensselaer Design Research Center (RPI): jvmiller@rdrc.rpi.edu "All I need is room to play."