home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!ub!niktow!pavlov
- From: pavlov@niktow.canisius.edu (Greg Pavlov)
- Newsgroups: comp.sys.sun.hardware
- Subject: Re: Is Sparc out of Gas?
- Message-ID: <1732@niktow.canisius.edu>
- Date: 19 Nov 92 13:45:02 GMT
- References: <3864@key.COM> <lg6j1iINN5dj@exodus.Eng.Sun.COM> <1992Nov17.160019@siisun.epfl.ch>
- Organization: Canisius College, Buffalo NY. 14208
- Lines: 40
-
- In article <1992Nov17.160019@siisun.epfl.ch>, brossard@siisun.epfl.ch (Alain Brossard EPFL-SIC/SII) writes:
- > In article <1712@niktow.canisius.edu>, pavlov@niktow.canisius.edu (Greg Pavlov) writes:
- > |> In article <lg6j1iINN5dj@exodus.Eng.Sun.COM>, ram@shukra.Eng.Sun.COM (Renu Raman) writes:
- > |> >
- > |> >Most of the current crop of RISC architectures are fundamentally similar
- > |> >and if one were to apply the same compiler and implementation
- > |> >techniques to all of them - its possible to find all of them within
- > |> >a few % points of each other in terms of performance, cost etc.
- > |> >
- > |>If this is true, then SUN is having VERY serious problems with its compilers
- > |>and implementation techniques. Comparing the latest announcements:
- > |>
- > |> System # processors SPECrate_int92 SPECrate_fp92 Availability
- > |>
- > |> SUN 2000 8 8047 10,600 April, 93
- > |>
- > |> DEC 10000 4 9108 17,187 Jan/Feb, 93
- >
- > What kind of data is this? I'm sure the 2 processors NEC we have in
- > Switzerland is faster than the 4 processors DEC 10'000, but what does that
- > prove? nothing!
-
- It "proves" that NEC has processors capable of supporting considerably great-
- er throughput than either DEC or SUN. If you are alluding to price differ-
- ences, you certainly have a point. But just as performance is influenced
- by price, price is also influenced by performance. E.g., the lower price
- of the SUN 2000 (relative to that of the DEC 10000) may be the RESULT of the
- lower performance as much as the cause.
-
- The above "data" comes from SUN and DEC, respectively. I have no reason to
- assume that either company "lied" or "doctored" the results. If it is
- "honest" data, then it shows that SUN's best effort with the SPARC falls
- far short in performance per processor when compared with DEC's best first-
- generation ALPHA effort. Because of that, I don't buy SUN's proposition
- that the SPARC is equal to other RISC processors on the market. That is
- why it is trying to build a 20-head system......
-
-
- greg pavlov
- pavlov@fstrf.org
-