Received: from casbah.acns.nwu.edu by karazm.math.UH.EDU with SMTP id AA29233 (5.65c/IDA-1.4.4 for ); Wed, 23 Oct 1991 19:53:13 -0500 Received: from neworder.ils.nwu.edu by casbah.acns.nwu.edu (4.1/SMI-ACNS-1.03) id AA08131; Wed, 23 Oct 91 19:49:05 CDT Return-Path: Received: by neworder.ils.nwu.edu (4.0/SMI-4.0) id AA01148; Wed, 23 Oct 91 19:39:34 CDT From: newton@neworder.ils.nwu.edu (Dave Newton) Message-Id: <9110240039.AA01148@neworder.ils.nwu.edu> Subject: Re: Transputers... To: glove-list@karazm.math.uh.edu Date: Wed, 23 Oct 91 19:39:33 CDT In-Reply-To: <9110231931.AA19069@watserv1.uwaterloo.ca>; from "Dave Stampe-Psy+Eng" at Oct 23, 91 3:31 pm X-Mailer: ELM [version 2.3 PL11] In a previous message, Dave Stampe-Psy+Eng said: > Actually, VPL is just sending world database updates by Ethernet, which > is OK. But we're talking about sending rendered pixels to the video board, > which, for a 500x500x24 bit picture, needs 6 Mbits/sec times the frame rate! > Can the video board handle this data rate??? If ypu use 4 transputers, can > they SEND this rate??? For our purposes, I think the world database > could reside on one transputer, which does preliminary clipping and > sends polygon and attribute lists to the other transputers. IF the > video board can handle the input speed, this idea will work. But not > if you have to design a custom video buffer... Or, how about using > 4 video boards, genlocking them, and switching between them as their > video segments occur??? Each set of transputers that is doing rendering would have a copy of the world in it, so theoretically all it would have to do is blob-move that memory into whatever yo uwere using for output. If I'm going the cheap eye-phone route, I'll probably not have 500x500x24 in the near future.