home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.hardware
- Path: sparky!uunet!munnari.oz.au!uniwa!cujo!jim
- From: jim@cs.curtin.edu.au (Jim McGrady)
- Subject: Re: A quick alpha question...
- Message-ID: <jim.727695529@marsh>
- Sender: news@cujo.curtin.edu.au (News Manager)
- Organization: Curtin University of Technology
- References: <C12856.90y@mach1.wlu.ca> <1993Jan19.120539.91007@vaxc.cc.monash.edu.au>
- Date: Fri, 22 Jan 1993 09:38:49 GMT
- Lines: 22
-
- bull@vaxc.cc.monash.edu.au writes:
-
- >mart4678@mach1.wlu.ca (Phil Martin u) writes:
- >> I have heard that there is an "Alpha" processor coming out from someone
- >> that runs at 100Mhz, and being fully-RISC, produces 100 MIPS. I would like
- >> to know how the megaflop rating of a RISC processor compares to its MIPS
- >> rating...
-
- > Alpha is made by Digital (DEC).
-
- > As for Mflops ratings, it's not a imple matter of saying X MIPS =
- >Y MFLOPS. MIPS measures individual instructions, MFLOPS measure a floating
- >point multiply and an add. Two processors may have very similar MIPS
- >ratings, but if one has a much more efficient floating point unit it will
- >have a higher MFLOPS rating.
-
- Pardon my ignorance, but can someone explain how a SPECmark is calculated.
- MIPS seem to be in the bad books now, but its hard to tell if 50 million
- SPECmarks is good without a point of reference :)
-
- Jim
-
-