🚀 go-pugleaf

In article <FISCHER.92Nov4044225@steinhaus.iesd.auc.dk> fischer@iesd.auc.dk (Lars Peter Fischer) writes:
>
>>>>>> "Eric" == Eric Scott Boltz (eboltz@jhunix.hcf.jhu.edu)
>
>Eric> 3. The NRW will be AS FAST AS A HIGH-END CLONE.
>
>Not true. Go take a look at the Road-runner (aka Sun 386i). Compare it
>to a 386 PClown. The Road-runner is *so* much faster, shows real
>smooth performance. It was a nice box, actually faster than the Sun-3
>line at the time. The SPARC killed it, but hey, they showed what you
>can do with an Intel chip if you have the guts and skills. We've even
>had Intel folks say that the Road-runner is the only machine ever
>designed to really use the 386.

  So true, so true.  Just look at Sequent.  Their box is all 486es and
  it screams, partially because of multi-processing but also because
  they have all the junk to make it a REAL computer instead of a dumb
  pc-clone (sorry but I have too much respect to call it a PClown).   
  Also Teradata used all 486s (and originally 386s, 286s and eve 8086s)
  in their database machine.  Another hummer in many ways.  Ditto for
  the Sun 386i.  We used them at Teradata extensively, mostly because of
  the compatibility with our own database machine, but they were great
  at running full SunOS with Sunview.  Much faster than any pc-clone
  I ever saw.  Of course, it was occasionally slow, but then it had
  umpteen processes running on it at the same time.  

  For my money, the P5 looks like a winner.  As a software type, I am
  not particularly interested in the underlying chip architecture EXCEPT
  insofar as it effects the cost and the speed.  The only reason RISC
  had any real purpose at all is that it was thought that you could get
  better speed at a lower cost in a shorter development time because of
  the simplifications of the instruction set.  But there are tradeoffs
  even in this regard.  RISC chips are NOT known for blazing character
  manipulation speed.  Application processors need good support for
  moving characters around and the x86 architecture is much better at
  this than most RISC chips.  NeXT is principally becoming an application
  processor.  Intel has enough experience now with the x86 architecture
  to continue to extend its performance envelope to be competitive (or  
  at least price competitive) with RISC chips for the foreseeable future.
  The market demand for the P5 (yes folks, here you CAN give some thanks
  to the pc-clone market) will guarantee that low-cost P5s will be readily
  available both from Intel and ultimately from second sources, something
  that can NOT be said for RISC chips like the 88110 and the Rx000 chips.
  We ALL program in Objective-C, right?  Why should the chip that we use
  matter that much to us, huh?  
   
  Even Steve Jobs was willing to get on the same stage at NeXTWorld Expo
  with the guy from Intel :-)  Sooner or later, everyone grows up :-)

  "pc-clone" implies an "intel chip"              
  "intel chip" does NOT imply a "pc-clone"

  Jon Rosen