home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.video.production
- Path: sparky!uunet!gatech!usenet.ins.cwru.edu!agate!spool.mu.edu!umn.edu!umeecs!quip.eecs.umich.edu!gilgalad
- From: gilgalad@quip.eecs.umich.edu (Ralph Seguin)
- Subject: Re: Vivid-24 and SGI
- Message-ID: <1992Dec30.001954.26789@zip.eecs.umich.edu>
- Sender: news@zip.eecs.umich.edu (Mr. News)
- Organization: University of Michigan EECS Dept., Ann Arbor
- References: <1992Dec29.034858.9390@mnemosyne.cs.du.edu> <gregpen.725621307@crash.cts.com> <1992Dec29.215045.20844@ludd.luth.se>
- Distribution: na
- Date: Wed, 30 Dec 1992 00:19:54 GMT
- Lines: 22
-
- In article <1992Dec29.215045.20844@ludd.luth.se> pjotr@ludd.luth.se (Peter Sj|str|m) writes:
- > assuming the mentioned 200.000 USD... that will give you
- >60 Vivid-24 or so (use cheap Amiga 2000s). 60*160MFLOPS*80%efficiency =
- >7680 MFLOPS. Will that not be enough (using decent software) to outperform
- >the SGI?
-
- Bullshit. Actual real world performance will come nowhere near this peak
- theoretical performance. I notice the 80% efficiency. I would guess something
- more like 15%-20%. If I'm wrong, I would be a VERY happy camper indeed,
- and buy a couple of these boards. As things stand, I'd much rather
- have an SGI machine.
-
- > I seem to remember a quote from 84 or so: rendering time per movie
- >(Cinema quality) is approximatly 15 minutes. When computing power increase
- >rendering time is still 15 minutes. Instead we make more complex scenes... :)
-
- Excellent quote. Short movies I guess? :>
-
- -Ralph
-
-
-
-