home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!att!fang!tarpit!tous!bilver!jwt!rjck!rob
- From: rob@rjck.UUCP (Robert J.C. Kyanko)
- Newsgroups: comp.arch
- Subject: Re: Which would give better floating point performance?
- Distribution: world
- Message-ID: <724554629snx@rjck.UUCP>
- References: <1992Dec16.063018.6239@ryptyde.cts.com>
- Date: Thu, 17 Dec 92 01:10:29 GMT
- Organization: RJK Enterprises
- Lines: 21
-
- scott@ryptyde.cts.com writes in article <1992Dec16.063018.6239@ryptyde.cts.com>:
- >
- > Hi all,
- >
- > I'm trying to figure out which one of these configurations would
- > yield the best floating point performance.
- >
- >
- > An i486DX/33 + Weitek 4167, or i486DX2/66 with no additional processor.
- >
- > Any experiences, references, and opinions are welcome.
-
- IMHO, it depends upon what you're doing. The Weitek is only going to help you
- with math intensive applications that also take advantage of this chip. I
- think the 486DX2/66 will give you almost equal performance in a real
- application anyway; plus you get double the integer performance with this part.
- I'd go with the 486DX2/66. But other opinions will vary based upon their
- needs.
- --
- I am not responsible for anything I do or say -- I'm just an opinion.
- Robert J.C. Kyanko (rob@rjck.UUCP)
-