brando@uicsl.csl.uiuc.edu (Brandon Brown) (02/09/91)
I have heard of benchmarks before, even landmarks, but what are SPEC marks? My boss asked me to find out what the SPEC mark ratings for an Intel 386 and an Intel 486 were. Do any of you happen to have that information? I would really appreciate it! If you don't know the SPECmark for the 386/486, could you send me a message with your machine, and its SPECmark? Thanks! Brandon +-----------------------------------------------------------------------------+ | Brandon Brown | Internet: brando@uicsl.csl.uiuc.edu | | Coordinated Science Laboratory | UUCP: uiucuxc!addamax!brando!brown | | University of Illinois | CompuServe: 73040,447 | | Urbana, IL 61801 | GEnie: xmg23356, macbrando | +-----------------------------------------------------------------------------+
newbery@stout.atd.ucar.edu (Santiago Newbery) (02/09/91)
In article <1991Feb8.201336.24388@ux1.cso.uiuc.edu>, brando@uicsl.csl.uiuc.edu (Brandon Brown) writes: |> |>I have heard of benchmarks before, even landmarks, but what are SPEC marks? |> |>My boss asked me to find out what the SPEC mark ratings for an Intel 386 |>and an Intel 486 were. Do any of you happen to have that information? |> |>I would really appreciate it! |> |>If you don't know the SPECmark for the 386/486, could you send me a message |>with your machine, and its SPECmark? |> | The Specmark comes from an industry consortium which includes workstation vendors such as Sun, IBM, HP and Intel. This benchmark is a composite of the results of 10 different tests. The idea is that the Specmark suite is less susceptible to benchmark "cheating" than the Dhrystone-based MIPS benchmark. Others complain however, that is more weighted in favor of floating-pt. performance vs. integer performance, so Intel branched off and created the I Spec or Integer Specmark. In any case here is some data: Workstation MHz Specmark I Specmark ============== === ======== ========== Compaq DeskPro 486 25 8.7 12.9 Sun Sparc IPC 25 10.0 11.2 Sun Sparc SLC 20 7.6 9.5
sauer@chs.dell.com (Charlie Sauer) (02/11/91)
In article <1991Feb11.073342.381@ico.isc.com> rcd@ico.isc.com (Dick Dunn) writes: >> ...Others complain however, that is more weighted in favor of >> floating-pt. performance vs. integer performance, so Intel branched off >> and created the I Spec or Integer Specmark. > >"Others" is approximately "intel" - to the extent that pundits say that "I >SPECmark" really means "Intel SPECmark". > >The serious political point in this is that SPEC hasn't blessed anything >called an "Integer SPECmark" (as far as I know). The serious technical >point is that, if the SPEC test suite overemphasizes floating-point (which >I personally think it does, but what do I know?), the "I Spec" ignores it >completely, which is rather worse. Maybe I didn't see the first Intel documents which separated the 4 nominally integer benchmarks from the 6 nominally floating point components of SPEC 1, but the first public use of this distinction that I remember came from John Mashey of MIPS, shortly after the RS/6000 announcement. Of course, John properly presented the individual results, the geometric mean of the floating components and the overall geometric mean, as well. -- Charlie Sauer Dell Computer Corp. !'s:uunet!dell!sauer (512) 343-3310 9505 Arboretum Blvd @'s:sauer@dell.com Austin, TX 78759-7299
garyt@ios.Convergent.COM (Gary Tse) (02/12/91)
rcd@ico.isc.com (Dick Dunn) writes: |newbery@stout.atd.ucar.edu (Santiago Newbery) writes in response to a |general "what's a SPECmark?" question: |> ...Others complain however, that is more weighted in favor of |> floating-pt. performance vs. integer performance, so Intel branched off |> and created the I Spec or Integer Specmark. | |"Others" is approximately "intel" - to the extent that pundits say that "I |SPECmark" really means "Intel SPECmark". Shucks, the SPEC number is supposed to be reported with a breakdown on the results of all the component benchmarks of the suite anyway. Given the breakdown, if someone wants to generate an "integer SPEC", or a "fp SPEC", or an "everything-but-spice SPEC", that's cool. Gives the marketing critters something to do, ya know. Besides, the rest of us can calculate geometric means too. Oh, or do you mean Intel is quoting their "I SPEC" number without giving a breakdown? Well, heck, yet another null data point in their marketing stuff is no big deal. (Honestly, though, anything, even this "I SPECmark", is an improvement over Dhrystone mips or Landmark MHz or whatever they use nowadays. Just MHO, of course. I am not interested in a benchmark flame, no sir. :-) -- Gary Tse, garyt@ios.Convergent.COM || tse@soda.Berkeley.EDU Live free or die.