home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.sys.ibm.pc.hardware:22397 comp.sys.intel:1551
- Path: sparky!uunet!nntp1.radiomail.net!fernwood!mpr.com!mslater
- From: mslater@mpr.com (Michael Slater)
- Newsgroups: comp.sys.ibm.pc.hardware,comp.sys.intel
- Subject: Re: Help - Cyrix processors, anyone know for sure?
- Date: Thu, 20 Aug 92 23:21:57 PST
- Organization: Microprocessor Report
- Message-ID: <0105010F.bjmclc@mprnews.mpr.com>
- Reply-To: mslater@mpr.com
- X-Mailer: uAccess - Macintosh Release: 1.6v1
- Lines: 12
-
- In article <Bt91A8.212@nntp-sc.Intel.COM> (comp.sys.ibm.pc.hardware,comp.sys.intel), cpurkis@gomez.intel.com (Clif Purkiser) writes:
- > My understanding is that the performance gain from the Cyrix
- > chip is due largely from the cache.
-
- Well, yes and no. The cache is necessary, but not sufficient, for
- the performance gain. The CPU core in the Cyrix part is much faster
- (in clocks per instruction) than the Intel 386, though not quite as
- fast as the 486. The cache is required, however, to keep the CPU fed,
- since the 386 bus doesn't have enough bandwidth to keep the core running
- at full speed.
-
- Michael Slater, Microprocessor Report mslater@mpr.com
- P.O. Box 2438, Sebastopol CA 95473 707/823-4004 fax 707/823-0504
-