Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Introducing "No-Moore's Law"

Author: Robert Hyatt

Date: 19:50:07 03/12/03

Go up one level in this thread


On March 12, 2003 at 19:20:40, Jeremiah Penery wrote:

>On March 11, 2003 at 23:29:48, Robert Hyatt wrote:
>
>>On March 11, 2003 at 20:17:57, Jeremiah Penery wrote:
>>
>>>On March 11, 2003 at 12:22:09, Robert Hyatt wrote:
>>>
>>>>On March 10, 2003 at 19:56:12, Jeremiah Penery wrote:
>>>>
>>>>>On March 09, 2003 at 22:10:06, Robert Hyatt wrote:
>>>>>
>>>>>>Did you see an advertisement where someone was trying to convince you to
>>>>>>buy over the other?  I doubt it.
>>>>>
>>>>>When P3 and P4 existed concurrently, every single ad where Intel advertised the
>>>>>P4, they were trying to get you to buy P4 over P3.  Of course they didn't
>>>>>explicitly say so, but I don't ever remember seeing an AMD ad where they
>>>>>explicitly mentioned an Intel chip either.  The P3 was, for many things, faster
>>>>>than the higher clocked P4.  Intel prematurely killed the P3 because they wanted
>>>>>to sell P4s, and P3 could have made P4 look bad.
>>>>
>>>>You are arguing this from the _wrong_ side.
>>>
>>>??? You asked if I'd ever seen an ad where Intel tried to convice me to buy a P4
>>>over a P3.  I answered that question - every single ad Intel made for the P4 was
>>>trying to get you to buy it over the P3.
>>
>>
>>Again, that isn't the point.  I asked where Intel had tried to get you to buy
>>a PIII instead of a PIV.
>
>Must I quote _your_ words in every one of _my_ posts, to show that you can't
>remember what you wrote?  Read the paragraphs above.  You never asked any such
>question.  You said "one over the other".  Again I must apologize for not
>reading your mind and somehow figuring out that you meant "P3 over P4".


I'm going to explain this _once_ more and I am not going to waste further
time:

1.  AMD marketed the K6 as a faster/cheaper replacement for the PII.  That's
simple enough to understand.  Except that it was _not_ an exact replacement,
as I have pointed out.

2.  Intel _never_ marketed the PIII as a replacement for the PIV.  _never_.
They _always_ marketed the PIV as a replacement for the PIII and said it was
faster/etc.

Now as to what you are trying to prove with your nonsensical twisting of
things, I have no idea.  But there is _no_ simularity between AMD saying "buy
me (K6 over PII)" and Intel saying "buy me (PIV over PIII)."  The PIV _will_
execute anything the PIII will execute.  The K6 will _not_ execute everything
the PII will.

So _what_ are you trying to prove with this ducking and dodging nonsense???

I have _not_ been unclear in what I wrote.   I have not changed my opinion.
I've been 100% consistent from the first post in this thread.


>
>>>  No, they didn't specifically mention
>>>the P3, but the AMD ads never specifically mentioned an Intel processor either.
>>>
>>
>>The magazine ads I saw _did_ mention the Intel chip by name.  The headline
>>was "why pay more?" (again, not an exact quote but but that was the gist)
>>and later "the K7 is faster and cheaper."  (another non-literal quote).
>
>Where is the Intel chip mentioned there?  Why don't you find a real ad and point
>me to it (or copy it here if it's in some printed source).  Otherwise, I'll
>consider your argument hand-waving.

And where exactly would I find a real add today?  The K6 and PII are _old_
news.  You might find something on the web, I have no idea.  But the very
_idea_ that AMD was trying to market their chip without saying it was "Intel-
compatible" is something barely short of absurd.  Intel _is_ the PC market.
It has been since IBM made that fateful choice 20 years ago.




>
>>>>The PIV is 100% compatible with the
>>>>PIII.  The inverse is _not_ true, and this is often referred to as "backward
>>>>compatibility".
>>>
>>>I gave one example earlier of where P4 is not 100% backward compatible.
>>
>>Where?  IE what can I do on a PIII that will fail on a PIV?
>
>When an FP instruction throws an exception, every x86 processor except P4 can
>tell you where the exception happened and what instruction caused it.
>
>Whether this is actually used in any programs is irrelevant.  It does show P4 is
>not 100% backward compatible, and that behavior _could_ be relied upon in some
>code somewhere.


OK.  And in that case it _might_ be considered to be incompatible, although by
the time it happens, the program is already DOA.  That's a bit different than
simply failing to execute certain instructions.



>
>>>>That is _different_ than their trying to sell a PIII and saying it is equal to
>>>>the PIV but
>>>>much cheaper.  And that case is closest to the K6 vs PII point.
>>>
>>>I'm not sure what exactly you're trying to say.
>>
>>
>>Trying to sell the PIV instead of the PIII is logical.  The PIV will
>>execute all PIII code.  It is faster.
>
>When the P4 was first released, it was not necessarily faster than the P3.
>Sometime around July 2001 (I can't find exact dates), the P3 was at 1.4GHz and
>apparently scaling pretty well, while the P4 was only at 1.8GHz.  I'm sure that
>P3 was faster on a LOT of stuff than the P4.
>
>Intel froze the clockspeed of the P3 at that time and raised the price a lot, so
>as to not compete with the P4.
>

And that has exactly what to do with the current disagreement?


>>>>>>Again, it depends.  If I do a target=P3, I would expect that to run on a PIV.
>>>>>
>>>>>If I do target=Pentium, I'd expect it to run on a K6.  That's basically the
>>>>>comparison you're making there.
>>>>
>>>>No, I bought a PII and did "target=pentiumII" because _ knew_ that the PII had
>>>>different optimization isues than the original pentium.  The PII had an OOE core
>>>>for example, while the original pentium depended on the compiler to pair
>>>>instructions
>>>>for the super-scalar execution units.  At the time I bought a PII, AMD was
>>>>advertising
>>>>their chip as faster and cheaper.  And it did _not_ say (but not 100%
>>>>compatible) and
>>>>that led to the confusion I mentioned.
>>>
>>>You said, "If I do a target=P3, I would expect that to run on a PIV."
>>>I replied "If I do target=Pentium, I'd expect it to run on a K6."
>>>
>>>Is that not a fair analogy?  I'm not sure where the rest of your paragraph came
>>>from, but it's pointlessly obfuscating.
>>
>>Yes.  But that isn't what we are doing.
>
>That is what I said the first time, and you said "No."  Now you say, "Yes,"
>while somehow trying to claim that it's not pertinent to the discussion?


Please _read_.  I _clearly_ said I compiled with "target=PII" because _that_
was the current architecture being produced.  It failed on the K6.

I did _not_ say "target=pentium" as it would make no sense.  Optimizations for
the pentium were _far_ different than optimizations for the PII.  So what does
target=pentium have to do with anything?  Or, while we are there, what about
"target=486" which also has _nothing_ to do with the current context.

Once again:  I used target=pentiumII.  That _was_ the current processor I was
using.  AMD convinced many to buy K6's rather than PII's.  And then we found
out that the K6 was not a pentiumII.

Why are we changing the topic back and forth?  Pentium II _was_ the issue at
the time.  It was _the_ processor of choice.




>
>>  I'm interested in buying a PII.
>>I buy one.  I compile for it and say target=pentiumII.  A friend buys a
>>K6 and assumes it is compatible with the PII since it is marketed as a
>>faster/cheaper replacement.  The code doesn't work.
>>
>>I have a PIII.  A friend buys a PIV.  My target=PIII program will work
>>perfectly for him.  I wouldn't assume a PIII will do everything a PIV
>>would do, but the inverse is logical as Intel makes that claim.  I'm used
>>to the idea that a newer version will be compatible with the older version,
>>plus adding some new features.
>
>K6 was compatible with the older version (Pentium) plus adding new features
>(3dNow at least).  K6 is slightly older than the P2.


Unfortunately it was being marketed _against_ the PII.  Which is where
this story started.

And new versions came out _after_ the introduction of the PII.  But with
the incompatibility issue still there.




>
>>>>>>And most likely if I do a target=p4 it would run on a P3 although I can
>>>>>>certainly think of reasons why it wouldn't.
>>>>>
>>>>>Most of the time it would, just as most of the time K6 would run P2 executables.
>>>>> The only difference I'm aware of (for K6/P2) is CMOV.  For P3/P4 the difference
>>>>>is SSE2.
>>>>
>>>>But _if_ you compile for the PIII it will absolutely run on the PIV.  If you
>>>>compile for
>>>>the PIV it will run on the PIV but perhaps not on the PIII depending on the
>>>>instructions
>>>>the compiler is capable of producing.  However, _nowhere_ did you see Intel say
>>>>"The
>>>>PIII is 100% compatible with the PIV".  You will find "The PIV is 100% backward
>>>>compatible with the Pentium processor product line."
>>>
>>>Point me to something where AMD specifically mentioned compatibility in terms of
>>>the _P2_.  Nowhere did you see AMD say "The K6 is 100% compatible with the P2".
>>
>>
>>No, they just said "we are faster and cheaper than the PII."  With the
>>implication that all else is "equal".  It isn't.  (or wasn't).
>
>I ask again if they specifically mentioned the P2.  I still haven't seen any
>evidence.

Got any evidence about how Kennedy was killed?  This was not last week.
But I'm waiting for any logical explanation for why AMD would _not_ claim
to be "Intel compatible".  And that means "current Intel compatible" by
reasonable inference.




>
>>>>>>But Intel doesn't market the P3 to be compatible with the P4.
>>>>>
>>>>>No.  They market P4 to be compatible with P3.
>>>>
>>>>Playing word games.  "backward compatible" is not "compatible".
>>>
>>>Just a couple paragraphs ago, you wrote, "The PIV is 100% compatible with the
>>>PIII."  Before you accuse me of playing 'word games', don't play them with your
>>>own writing.
>>
>>I believe I also used the term "backward-compatible" pretty clearly.  I'm
>
>And I didn't?

Not to me.  Talking about PIII not executing PIV code is not exactly
"backward compatible".


>
>>not playing any word games.  _anything_ that works on a PIII will execute on
>>a PIV, unless instruction timing is critical.  That _is_ compatible in the
>>definition of "backward compatible".
>
>I said the same thing, and you accused me of playing word games.  Somehow you're
>not playing word games when you say the same thing.  Funny how that works.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.