Some senseless babbling from Jonathan Michaels to Mike Ruskai
on 04-04-98  00:31 about Motherboards and os/2...
 MR > For example, with Java benchmarks I've done myself, my speed is
 MR > very much faster than 200MHz Pentium Pro machines (using Windows
 MR > NT).
 JM> i can find easier ways to cripple a machine than that.
You're suggesting that Intel is somehow conspiring with the writer of the 
JVM to make Pentium Pro machines appear slower than Pentium II machines?  
 MR > I searched high and low for all speed information I could find,
 MR > and nothing showed the Pentium Pro faster than a Pentium II at
 MR > the same clock rate, much less twice as fast.  In fact, a normal
 MR > Pentium with MMX (which isn't important - the larger internal
 MR > cache is) at 200MHz is only marginally slower than a Pentium Pro
 MR > at 180MHz.  A Pentium/233 with MMX is faster.
 JM> whic intel publication did ths come from .. certain not any of the
 JM> processor specific documentation i have seen, recently.
The only Intel information I referred to was their iSPEC 2.0 rating.  
Everything else was independant of Intel.
 
 MR > I am currently using a Pentium II/233 overclocked to 266MHz,
 MR > 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 your
 JM> processor, not mine. or that it shows how conservative intel are about
 JM> thier rating. 
I'm well aware of the pitfalls of overclocking.  I've kept it at this speed 
because it is operating perfectly reliably.  Not a single problem. 
 JM> i'm stilll waiting to find out the real story behind the damning
 JM> reports that were authored by nasa schuttle engineers about the arm
 JM> twisith intel used to get nasa to swap from the far more reliable 6800
 JM> seris processors to the less reliable 80xxx seris 'consumer grade'
 JM> procesors. 
This I find very difficult to believe.  I've never heard of any 
arm-twisting done by Intel.  Doesn't make any sense.  What leverage did 
Intel have over NASA to do that?  Sure, Intel has tried to stifle 
competition with lawsuits, but they're just not in a position to use 
leveraging tactics.  
 MR > Granted, there are other factors which would show a Pentium Pro
 MR > to do better, probably, due to its superior branching,
 MR > prediction, etc.  But none of these internal features are absent
 MR > in the Pentium II, and in fact enhanced.
 JM> for ms win95 .. problem is that ms win 95 is a ms dos shell and ms dos
 JM> is, at best a 16 bit system with 8 bit internals, os/2 is 32 bit
 JM> largely as is nt and most if not all unicies.
The Pentium II is *NOT* optimized for Windows 95.  It is a departure from 
the Pentium Pro in that it does not sacrifice 16-bit performance for 32-bit 
performance.  It performs equally well with both.  Furthermore, DOS has 
never been 8-bit.  It was first written for the Intel 8086 16-bit processor 
by Seattle Computer Products.
 JM> because they are thier is noting to write home aboutm what has been
 JM> done to them to make thems more or less usable in 32 bit mode is what
 JM> concerns me from an engineering point of view .. alas thes days only as
 JM> a source of interest. even when i was working i'd nver recommend any
 JM> project the was responsible for the saftey and lives of people be
 JM> entrusted to intel silicon. 
 JM> sure intel has goten more reliable over the last twent years, but it
 JM> is still not as reliable or predictable as motorola was twenty years
 JM> ago. 
 JM> inspite of apearences to the contrary i am open to 'reeducation' as
 JM> the chinesse call it, but you would have to put up a good showing.
I've never heard any complaints about reliability with Intel processors.  
The only problem they've ever had is that Pentium lookup table problem a 
couple years back.  I don't know what you are talking about.
 
 MR > I upgraded to this chip from a Pentium/200 (non-MMX, with the
 MR > normal internal cache), and clocked at 266MHz with the 75MHz
 MR > bus, it's right about twice as fast in raw processing power.
 MR > Encoding a MPEG-1 Layer-III audio stream with a 44.1KHz sampling
 MR > rate, joint-stereo, and a 128KBps stream speed took about 2.6
 MR > minutes per minute of audio on the Pentium/200, and takes about
 MR > 1.3 minutes per minute of audio on this machine.
 JM> ok, now you have me at a disadvantage, i am on new turf, i know very
 JM> little how precicely mpeg is processed.
 JM> these numbers are impressive, but what do they realy mean and what
 JM> sort of assistance is the 'raw' data stream getting form peripheral
 JM> circuit processing and off buss processor emulation in the operating
 JM> systems itself, or more precicely the applications, that are handling
 JM> the tasks .. whch are buss mastering and whic are slving the processor
 JM> .. thier is a lot of task specific questions that need to be answered
 JM> before you can make the statement that you have dome, by just
 JM> presinting the 'statistics' as an fait acompli. 
It's the same application.  There are no DSP's handling any of the work.  
It's done solely by the processor and the relevant subsystems.  The 
application does not use MMX, so that can't account for anything, either.  
It's using the same exact instructions to do the job, and the Pentium II at 
266MHz is twice as fast as the Pentium at 200MHz. 
 JM> over my years i have don this as well, particilarly when i knew the
 JM> client wouldn't understand the technical details and it was important
 JM> that the right hardware be used .. i'd presnet the case for the
 JM> prefered end run and let the 'statistics' speak for themselves.
Well, there's a point when statistics *do* speak for themselves.  That 
point is when everything you do on the computer has become part of the 
statistics.  There can be no more meaningful rating.  Everything I do on 
this computer is about twice as fast as on the other chip.  Everything.  
That statistic most certainly does speak for itself.
 MR > The memory subsystem on this one is also better, using SDRAM at
 MR > the 75MHz bus speed, as opposed to 60ns 72-pin DRAM SIMMs on the
 MR > other machine, but given the synchronous nature of a streamed
 MR > file encoding, I'd wager than the L2 cache of each processor
 MR > essentially nullified the effect on performance of the memory.
 JM> of cousre .. but the fact that thier is less l2 cache and that it runs
 JM> at half buss speed, that is it runs at the speed the motherbaprd is
 JM> running, in your case that woult be atclck/2 with would be some 75/2 ..
 JM> .37.5 mhz 
You have it all wrong.  The L2 cache with Pentium II chips is on the 
processor module.  It's a big bulky thing that goes into the slot on the 
motherboard.  
The Pentium II has two separate buses just like the Pentium Pro.  One L2 
cache bus, and one system bus.  The L2 cache bus operates at half the 
internal clock rate.  So my L2 cache is operating at 133MHz.  It is 
unaffected by the system bus speed, which is 75MHz.  
The Pentium Pro ran its L2 cache at processor speed, of course.  All else 
being equal, that would make the Pentium Pro the faster processor at a 
given clock rate.  But all else is *NOT* equal.  The Pentium II is the 
faster processor at a given clock rate.  And as Lynn pointed out, the 
333MHz model now runs the L2 cache at processor speed, making it that much 
faster than a Pentium Pro.
 JM> lets get one thing straigt you do not have a 75 mhz motherboard, at
 JM> best you have a system that runs at 37.5 mhz and parts of the pci bus
 JM> run at 75 mhz 
You seriously need to reevaluate what you think you know about PC 
architecture.  The system bus is running at 75MHz.  The PCI bus is running 
at half that speed, or 37.5MHz.
 JM> ther is currently a great deal of debate about the efficiencies of
 JM> this sort of design structure .. the jury is out and not expected to be
 JM> back in for at least 4 or so years .. that is how long it is going to
 JM> take to 'prove' the differences tween the intel ideas embodyied in
 JM> thier merced line, dec's alpha structures and the sun microsystems
 JM> sparc architectures. 
 JM> untill then yor guess is a s good as mine, as the okld saying goes.
Well, until you get your facts straight about the Intel processors, my 
guess is better than yours.
 
 MR > That's my say.  I'm very interested to here what kind of
 MR > calculations you did.  And if there's any kind of rough speed
 MR > testing I can do for you here (I've got XFree86 installed here,
 MR > and am using a Matrox Millenium card), let me know.
 JM> this sounds good to me .. i;ve been looking for a way to do some
 JM> definitieve do comparisons. sofar i have looked at this issue with an
 JM> old engineers eye. as i put at the foot of one of my earlier posts
 JM> regarding this, or a similar issue. i am a besoted motorola bigot, have
 JM> been for twenty someting years, i like the clean lines and clear
 JM> architecural delineations in the 6800 seris processor family
I think you've made that obvious :)
 JM> i also like to discuss processor theory, thoug as you may readily see
 JM> i am getting rusty, i am not making excuses for my self, god only knows
 JM> i need to cut myself some slack, i have been on full social security
 JM> benefits for some 10 year and i'm not ecpected to recover enough to
 JM> work again .. so far i';ve regaind the ability to talk and to walk,
 JM> much to my freinds disgust .. tey wanted a life sized doll to play with
 JM> (sort of grin ok). i got up one morning and found my brain shut down ..
 JM> it eventually took me 6 years to get out of bed, showered and dressed
 JM> the longest moring i've ever had .. grin. 
I can't say I can imagine what you've had to go through.  
 JM> sorry, i am easly distracted these days, back to the system .. what
 JM> sort of a matrox do you have, mine, ok as ordered is a millenium II
 JM> with 16 mb . yours being ? how do you plan to account for the
 JM> differences in cache size and spped. also the 440fx chipset dosent
 JM> support dimms (any body want 128 mb of dimms ?) so i had to get 128 mb
 JM> of ecc simms, well the reseller did and im not hs faverite customer at
 JM> the moment .. he forgot the differences in the chipsets and ordered the
 JM> wrong stuff.. 
Was a Millenium with 4MB, now a Millenium II with 8MB, AGP.  The Acclerated 
Graphics Port is based on a 66MHz PCI specification.  With the way it 
works, it can deliver up to four times the speed of PCI for the graphics 
card.
The Pentium II's L2 cache is 512K, and runs at 133MHz.  Most Pentium Pro 
chips only have a 256K chip, and runs at the chip speed (which isn't enough 
of an advantage to make up for the other improvements to the Pentium II).
My motherboard uses the 440LX chipset, which supports DIMMs.  As a result, 
I have two 32MB SDRAM DIMMs which have a clocked interface, running at the 
75MHz system bus speed.  ECC was too expensive, so they have no parity.  
I'm not doing mission critical work on this machine, so the odds of that 
being a problem are staggeringly low.  But ECC is available.
 JM> can you thinkof any other differences that mich have to be accounted
 JM> for ? 
Let me summarize the differences between the two systems:
1)  Processor:  
    a)  the Pentium runs at 200MHz, and has a 16K L1 cache (shared as 
        data and instruction cache).
    b)  the Pentium II runs at 266MHz, has a 32K L1 cache (16K each for
        data and instruction - a Pentium Pro only has a 16K L1 cache).
2)  Cache:
    a)  the Pentium has a 512K L2 cache on the system bus, running at
        66MHz.
    b)  the Pentium II has a 512K L2 cache on a dedicated bus inside the
        processor module, running at 133MHz (runs at half the processor
        speed).
3)  Memory:
    a)  the Pentium has 128MB of 60ns DRAM memory.
    b)  the Pentium II has 64MB of SDRAM memory, with a clocked interface
        that runs at the 75MHz system bus speed.
The other hardware from the old system was put in this machine.  The only 
change since then was a new graphics card, as mentioned above (the other 
machine is sporting a $23 PCI Cirrus Logic card in VGA mode with a 9" 
monitor that has its case off because I'm trying to figure out where the 
jittering is coming from).  Both machines were using IDE bus mastering.  
I haven't noticed any difference in hard drive speed.  
 JM> you also mention xfree86, is this a unix xfree86 or the german os/2
 JM> port ? 
It's the English version of XFree86 for OS/2.  
 JM> ps, mike, i am tired and hurting more than usual, i may have been a
 JM> bit more jonathanesque than usuall, i've just reread my post and it
 JM> strikes me as more that usuall .. please read between the lines if you
 JM> find the going tough. 
 JM> maybe i should have put this at th front .. ummm
I've got pretty thick skin, so don't worry about it.  
If I can find someone with a Pentium Pro machine (i'm going by what I found 
about other people's Pentium Pro machines), I'll do some more direct 
testing, but I don't think I'm in error.
Mike Ruskai SA/AG #1106
thanny@home.com
... Beavis & Butthead: MST3000 for morons.
___ Blue Wave/QWK v2.20
--- Renegade v5-11 Exp
---------------