Hello Mike!
Sunday April 05 1998 23:00, Mike Ruskai wrote to Jonathan Michaels:
 MR > Some senseless babbling from Jonathan Michaels to Mike Ruskai
 MR > on 04-04-98  00:31 about Motherboards and os/2...
 MR >> For example, with Java benchmarks I've done myself, my
 MR >> speed is very much faster than 200MHz Pentium Pro machines
 MR >> (using Windows NT).
 JM>> i can find easier ways to cripple a machine than that.
 MR > You're suggesting that Intel is somehow conspiring with the
 MR > writer of the JVM to make Pentium Pro machines appear slower
 MR > than Pentium II machines?
its not the forst time that it has happened, nore will it be the last, what 
s
the cost of truth when you are talking of loosing millions and millions on 
each
production run.
 MR >> I searched high and low for all speed information I could
 MR >> find, and nothing showed the Pentium Pro faster than a
 MR >> Pentium II at the same clock rate, much less twice as
 MR >> fast.  In fact, a normal Pentium with MMX (which isn't
 MR >> important - the larger internal cache is) at 200MHz is
 MR >> only marginally slower than a Pentium Pro at 180MHz.  A
 MR >> Pentium/233 with MMX is faster.
 JM>> whic intel publication did ths come from .. certain not any
 JM>> of the processor specific documentation i have seen,
 JM>> recently.
 MR > The only Intel information I referred to was their iSPEC 2.0
 MR > rating.  Everything else was independant of Intel.
same here, but mine is several year old and also a good bit of interpolation.
though i place absolutely np faith or trust in any benchmarks produced by the
manufacturers own 'benchmark suite', thier are ways and means to make things
look good or bad and all it takes is a few scant lines of code.
been thier, experienced that as wellas being responsible for authorising such
things to be one in a large organisation, the statut of linitations has run
out, so i am safe.
 MR >> I am currently using a Pentium II/233 overclocked to
 MR >> 266MHz, with a 75MHz bus.
 JM>> and this is supposed to prove, what ?
 JM>> you might be a bit stupid to do that but then again its
 JM>> your processor, not mine. or that it shows how conservative
 JM>> intel are about thier rating.
 MR > I'm well aware of the pitfalls of overclocking.  I've kept it at
 MR > this speed because it is operating perfectly reliably.  Not a
 MR > single problem.
other that raised enegry consumption, that is outside of its rated
specification .. i have been shown test and environmental results concerning
over clocking ppro ..
you get far less reserve capacity, life shortens significantly (more the 25
percent) and the processor become exceedingly sensative to transient
fluctuations. this is a sticking point in two areas, on being the typical
startup surges on initial system turnon and heatup from cold. then thier is 
the
unexpected one ..
os/2 is a real multi user time sharing operating system. this means that 
hier
are tresnient load fluctuations that are directly realted to processs and 
ser
task loading casing the processor to 'work' harder, get hotter, dig into its
'reserve capacity' shorten its life, all of this happens at a logorithmically
increased rate at each over clocking step.
 JM>> i'm stilll waiting to find out the real story behind the
 JM>> damning reports that were authored by nasa schuttle
 JM>> engineers about the arm twisith intel used to get nasa to
 JM>> swap from the far more reliable 6800 seris processors to
 JM>> the less reliable 80xxx seris 'consumer grade' procesors.
 MR > This I find very difficult to believe.
which bit, thatt he nasa engineers were 'silenced' or that intel has a
historical record of producing less reliable less 'powerfull' process control
devices on a platform specific device for device comparison with the
acknowledged market leader . motorola.
for motorola, the nasa contracts were just another contract .. for intel it 
was
life and death .. and for nasa hwo learned too late that it awas alos a life
and death issue for its personell.
i've spent the majority of my working life (tradgically cut short) in the
process control 'industry', not the much less important (sorry guys) 'home
computer' market place. a computer for the home desktop requires no real
reliabilty or longevity .. as so eminently 'shoved' down oer throats by intel
each time it makes a more fragile more cost effective processor more 
integrated
chip and motherboard system .. one chip breaks . replace the motherboard, ala
ast .. not thatt his policy is one of the reasons ast is no longer. also why
ast had 'problems' with os/2.
 MR >   I've never heard of any
 MR > arm-twisting done by Intel.  Doesn't make any sense.  What
 MR > leverage did Intel have over NASA to do that?
like microsoft has done in thier efforts to get a corporate user to shift 
rom
ibm to microsoft . they used predatory pricing policies, based upon the 
irect
value that intel/microsoft see as being provided by that particular account.
we are now drifting into the semanitis of market place stratgy that has 
ittle
to do with os2hw, yes ?
 MR >   Sure, Intel has
 MR > tried to stifle competition with lawsuits, but they're just not
 MR > in a position to use leveraging tactics.
if your hurting for funds, as nasa were at the time (and still are) an
organisation comes along and says we can provide a processr that costs one
quarter the price of your current processor, you would jump ship too.
based on the information supplied the chioce was good value .. problem was 
that
the reality varied somewhat radically to the 'facts' as presented on paper. 
t
has effectly cost nasa some 15 percent more than the original costs to use
intel based subsystems .. from memory .. every system in a capsule needed a
second redundant systen in case of failure, in the case of envoroment 
ritical
systems a thired level of redundancy .. lets face it dead austronausts don't
make good copy. the level of contingancy had to be doubled by making the jump
form motorola to intel. then in some aplications the 32 bit intel 80386 dx 33
mhz didn;t have the same grunt as the 20 year older mc 68b09e runing at some
1.7 mhz and being a 16 bit internal buss and 8 bit external data path.
given thatt he on board systems were optimised for the way the motorola
processor did its basic thing .. thier was a lot of very costly head 
craching
when the faster, less relibale, and much less powerfull intel processor hit 
the
design desks at nasa.
anyway this is where i call it quits, for those who want to contine this
netmail is a far mor aprotraite place. .. sorry for this bandwidtn.
 MR >> Granted, there are other factors which would show a
 MR >> Pentium Pro to do better, probably, due to its superior
 MR >> branching, prediction, etc.  But none of these internal
 MR >> features are absent in the Pentium II, and in fact
 MR >> enhanced.
 JM>> for ms win95 .. problem is that ms win 95 is a ms dos shell
 JM>> and ms dos is, at best a 16 bit system with 8 bit
 JM>> internals, os/2 is 32 bit largely as is nt and most if not
 JM>> all unicies.
 MR > The Pentium II is *NOT* optimized for Windows 95.  It is a
 MR > departure from the Pentium Pro in that it does not sacrifice
 MR > 16-bit performance for 32-bit performance.  It performs equally
 MR > well with both.  Furthermore, DOS has never been 8-bit.  It was
 MR > first written for the Intel 8086 16-bit processor by Seattle
 MR > Computer Products.
the intel 8086 was a competitor for the much bettter and costly 68b09e that 
was
a 16 bit internal and 8 bit EXTERNAL data path, this being the liniting 
actor
of its usefullness and processing ability. the intel chip is also a 16 bit 
chip
internally abd a 8 bit 
--- WtrGate+ v0.93.p4 sn 501
(3:712/808)
---------------