🚀 go-pugleaf

In article <6156@npri6.npri.com>, richard@npri6.npri.com (Richard Head) writes:
> 
> DR's "reviews" (and I use the word loosely), are essentially an advertisement.
  
  Advertisement for ?????  

> >So?  Is it any worse a comparison point than the VAX-11/780?
> 
> "Worse"?  The point is, you academic flunky, that MicroVUPs are MACHINE
> SPECIFIC.  SPECmarks are machine and platform independent.  If you think
> MicroVUPs are so great, why not have MicroVAX3100VUPS?  How about
> VAXstationVUPs?  
> 
  How is a SPECmark "machine and platform independent" while MVUPS is "MACHINE 
  SPECIFIC" ?   How is it different from a MVUP, other than that the
  index for the former is an 11/780 and the index for the latter is a
  uVAX II ?  Because "VAX11/780" doesn't appear in the name ???

  What is an "academic flunky" ?  I haven't spent much time in a university
  in some years, so I am not sure what that term is supposed to mean.  Some-
  one less intelligent, discerning, and analytical than .... ?


  greg pavlov,
  pavlov@fstrf.org


  "Note that Digital Review gives relative performance using the Micro Vax II
   as a basis for comparison (MVUPS).  For consistency with the rest of this
   document, we use the VAX 11/780, which significantly affects the ratios.
    ........
   We applaud DR for several reasons.  First, they try to offer some useful 
   benchmarks in place of empty mips-ratings, which is more than most magaz-
   ines do.  Second, they are willing to listen to input and improve the
   usefulness of their benchmarks."

                              - "Performance Brief CPU Benchmarks",
                                                         John Mashey, Editor

   P.S.:

   Our own experience is this: over the past 7 years, we have run three
   major series of benchmarks based on our own applications prior to (for
   us) substantial hardware purchases.  We found that, for our applications
   at least, the DR benchmark results always correlated closely with ours,
   in terms of relative performance of the systems we were comparing.  All
   the systems we tested were UNIX-based.