Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!iuvax!purdue!decwrl!ogicse!blake!milton!pezely@cis.udel.edu From: pezely@cis.udel.edu (102SMI) Newsgroups: sci.virtual-worlds Subject: Re: Applications of VR Message-ID: <13927@nigel.udel.EDU> Date: 15 Mar 90 02:31:08 GMT References: <16481@well.sf.ca.us> <2375@milton.acs.washington.edu> Sender: hlab@milton.acs.washington.edu Reply-To: Daniel Pezely Organization: University of Delaware Lines: 40 Approved: hitl@hardy.u.washington.edu In article <2375@milton.acs.washington.edu> jeffj@cbnewsm.att.com writes: >Daniel; > I was wondering what you are going to do your VR engine on? I heard >that it is possible to do it on a PC (probably not well 8-) ) with a >specialized video boards at 5 frames a second. Is this what you are >looking at? Hopefully I will be able to do the same thing down the road. >Thanks for any info! > > Jeff Jones For the compute engine, I was going to use Suns (Sun 3's and SparcStations). Since just one machine is not powerful enough, I was going to implement a distributed processing system. At UDel, we have lots of machines which can be used. The time of day/night might be restricted, but I'm not kepting normal hours anyway. For the display machine, it will be a sparcstation initially. (When's the sparcstation-2, -3 coming out...?) Eventually, I'd like to port this to a 486 and/or an 040 machine. There are a few IRIS's around campus, so... maybe. The evolution of the project is: a modified (for lack of a better term) CAD system, distributed/multiuser CAD, rendering system, simulator, VR system. The language will be C++, and I will make this as modular and atomic as possible. (Send mail only if you disagree or wish to flame.) I have access to gnu's g++ and the PC zortech c++ compiler, so I'll try to make most of the code compatible across a few platforms. For *my* needs as an undergraduate, the development of the tools and libraries is more important than the final application. The app will be improved upon in grad school or with my future employer (AutoDesk would be nice... (-: ), whichever route I decide to go. -Daniel -- Daniel Pezely (NSFnet) 728 Bent Ln, Newark, DE 19711 Comp Sci Lab, 102 Smith Hall, U of Delaware, Newark, DE 19716; 302/451-6339