home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!orca!mesa!rthomson
- From: rthomson@mesa.dsd.es.com (Rich Thomson)
- Subject: Re: Graphics Systems
- Message-ID: <1992Jul22.212524.6845@dsd.es.com>
- Sender: usenet@dsd.es.com
- Nntp-Posting-Host: 130.187.85.21
- Reply-To: rthomson@dsd.es.com (Rich Thomson)
- Organization: Design Systems Division, Evans & Sutherland, SLC, UT
- References: <GLEW.92Jul12214745@pdx117.intel.com> <1992Jul13.215541.23179@adobe.com> <1992Jul20.200401.14637@awdprime.austin.ibm.com>
- Date: Wed, 22 Jul 92 21:25:24 GMT
- Lines: 34
-
- In article <1992Jul20.200401.14637@awdprime.austin.ibm.com>
- repass@futserv.austin.ibm.com (James Repass) writes:
- >CPU <-> Bus <-> Display Adapter (DA)
- >CPU <-> Bus <-> DA with memory
- >CPU <-> Bus <-> DA with memory and processing (CPU)
- >CPU <-> Bus <-> DA with memory and CPU
- >CPU <-> Bus <-> DA with memory and CPU <-> Bus <-> Display Adapter
- > / /
- > /------------------------------/ /
- > / /
- >/ /------------------------/
- >CPU <-> Bus <-> Display Adapter (DA)
- >
- >Where will it END?
- >Where WILL it end?
-
- How about this:
-
- CPU <-> Bus <-> 44 DSPs with memory <-> 4 pixel processors <-> memory
-
- Graphics pipelines are becoming more parallel in their implementation
- since graphics problems have lots of parallelism in them. Many
- graphics algorithms are "embarrassingly parallel" because they don't
- require any communication between the separate sub-problems. An
- example of this is screen-space decomposition of polygons for
- rendering. Each screen "bucket" (rectangular area) touched by the
- polgyon can be rendered independently.
-
- -- Rich
- --
- Repeal the personal income tax; vote Libertarian in 1992.
- Disclaimer: I speak for myself, except as noted; Copyright 1992 Rich Thomson
- UUCP: ...!uunet!dsd.es.com!rthomson Rich Thomson
- Internet: rthomson@dsd.es.com IRC: _Rich_ PEXt Programmer
-