home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.sys.mac.hardware:14511 comp.unix.aux:3038
- Newsgroups: comp.sys.mac.hardware,comp.unix.aux
- Path: sparky!uunet!elroy.jpl.nasa.gov!ames!nsisrv!jagubox!jim
- From: jim@jagubox.gsfc.nasa.gov (Jim Jagielski)
- Subject: Re: Quadra SPECmarks (well... kinda)
- Message-ID: <925@jagubox.gsfc.nasa.gov>
- Sender: usenet@nsisrv.gsfc.nasa.gov (Usenet)
- Nntp-Posting-Host: jagubox.gsfc.nasa.gov
- Reply-To: jim@jagubox.gsfc.nasa.gov (Jim Jagielski)
- Organization: NASA/Goddard Space Flight Center
- References: <912@jagubox.gsfc.nasa.gov> <1992Jul31.063520.29297@panix.com>
- Date: Fri, 31 Jul 1992 10:23:20 GMT
- Lines: 34
-
- alexis@panix.com (Alexis Rosen) writes:
-
- >jim@jagubox.gsfc.nasa.gov (Jim Jagielski) writes:
- >>I would suggest that, with the right compilier (such as gcc), the Quadras
- >>would fall quite nicely in this catagory...
-
- >>Funny how compilers can _really_ affect "standard" benchmarks yet don't
- >>affect SPECmarks that much (as it should be)...
-
- >Um... I'm pretty sure that's not what you meant to say...
-
- >Compiler differences (c89, gcc, cc) will make _noticeable_ changes in SPECs
- >for the Quadras. Or any other machine. I guarantee it.
-
- Of course... I guess what I meant to say is that compilers that tune themselves
- for such "benchmarks" as dhrystones and the like (and thus show inflated
- performance levels for just these benchmarks) don't make such a corresponding
- _major_ change in SPECmarks...
-
- No, that didn't come out right either... I guess the basic gist is that
- dhrystones and the like aren't a very good indication of SPECmarks simply
- because it's easy for a compiler to be tuned for a specific benchmark but
- not so easy for an entire suite.
-
- I think that's it :)
- --
- ==============================================================================
- #include <std/disclaimer.h>
-
- Jim Jagielski NASA/GSFC, Code 734.4
- jim@jagubox.gsfc.nasa.gov Greenbelt, MD 20771
-
- "The great scallop... this tatty, scrofulous old rapist,
- is second in depravity only to the common clam."
-