home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!stanford.edu!CSD-NewsHost.Stanford.EDU!news
- From: philip@ziggy.stanford.edu (Philip Machanick)
- Subject: Re: Specmarks for R4000's
- Message-ID: <1992Jul27.174415.4233@CSD-NewsHost.Stanford.EDU>
- Sender: news@CSD-NewsHost.Stanford.EDU
- Reply-To: philip@ziggy.stanford.edu (Philip Machanick)
- Organization: CS Department, Stanford University, California, USA
- References: <22775@oasys.dt.navy.mil>
- Date: Mon, 27 Jul 1992 17:44:15 GMT
- Lines: 40
-
- In article <22775@oasys.dt.navy.mil> monicaw@oasys.dt.navy.mil (Monica Walker)
- writes:
- > I was asked how the new R4000's, the Crimson, and 4D/35's compare in
- > SPECmarks with VAX- and DEC-stations. Could someone please post the
- > numbers (or mail them to me)? I also have to admit that I'm not quite
- > sure what a SPECmark is... does any one have a simple explanation?
-
- SPECmark used to be a single figure (now called Spec89) that was meant to more
- accuractely characterize the performance of a complete system than MIPS. As
- with MIPS, the figure is relative to a VAX 11/780 (= 1 SPECmark).
-
- It is based on a range of benchmarks meant to be characteristic of a wide range
- of real workloads, and is based on runs of real programs (not synthetic or toy
- benchmarks). Recently, some manufacturers found a "trick" optimization that
- sped up one of the benchmarks so much that the figure became meaningless, so
- that specific program was deleted and the numbers were split into integer and
- floating point (SPECint92 and SPECfp92). Take care not to compare Spec89 and
- Spec92 numbers - some manufacturers sitll quote SPECmark meaning Spec89.
-
- For more discussion of this see comp.benchmarks.
-
- Here are some figures that were posted to comp.benchmarks (I added in R4000
- Indigo figures as posted to this group):
-
- System clock cache SPECint92 SPECfp92
- ------ ----- ----- --------- --------
- Sun SS-10/30 36 0+36k 44.2 52.9
- Sun SS-10/41 40 1M+36k 52.6 64.7
- Sun SS-2 40 0+64k 21.8 22.8
- MIPS Magnum 50/100 0+16k 36.8 40.0
- SGI R4000 Indigo 50/100 1M+16k 57 61
- SGI Crimson 50/100 1M+16k 61.7 63.4
- HP 710 50 0+96k 31.6 47.6
- HP 750 66 0+512k 48.1 75.0
- IBM 340 33 0+40k 27.7 51.9
- IBM 560 50 0+72k 42.0 85.6
- Intel Xpress (486) 50 256k+8k 30.1 14.0
- --
- Philip Machanick
- philip@pescadero.stanford.edu
-