Computer Chess Club Archives


Search

Terms

Messages

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

Author: Jeremiah Penery

Date: 15:05:34 03/06/03

Go up one level in this thread


On March 05, 2003 at 22:56:50, Robert Hyatt wrote:

>On March 04, 2003 at 21:55:55, Jeremiah Penery wrote:
>
>>On March 04, 2003 at 11:23:17, Robert Hyatt wrote:
>
>The precise specification is that written by Intel, which is my point.  A
>program with timing loops is bad programming, but it still works on Intel
>boxes if it is developed there.  It won't work on something that behaves

It wouldn't work on the same Intel CPU with twice the clockspeed.

>differently.  But, back to the _real_ point.  A cpu that supposedly competes
>with the PII should be _compatible_ with the PII at any level of abstraction
>a user might want.  All the way up to the output from a compiler that targets
>the PII processor with an executable.  The K6 failed that test and led to some
>bad vibes.

I asked already whether they claimed specific P2 compatibility (which they could
fail), or just general compatibility (which they meet, because they properly
report the CPU capability in CPUID).

>>If you made patented advances in Crafty, which required licenses for the
>>opponents to use, your window would be eternity if you wanted it to be.  That's
>>the situation Intel is in.
>
>And???  The window is there for the leader whether there are licensing issues
>or not.

My point is that the window would be FOREVER if Intel chooses not to license.

>>>When Intel announces something "new" they have a window that stretches for
>>>however long it
>>>takes the opposition to implement the changes.  Or if they choose not to, they
>>>can choose to
>>>make those "changes" a big deal in advertising which will hurt the competition.
>>>So the
>>>followers have to follow for the most part.
>>
>>AMD _can't_ implement SSE (or whatever) unless Intel licenses it to them.  If it
>>was that big of a deal, Intel wouldn't license it at all.
>
>
>Why are we off onto SSE?  It hasn't been a problem.  Cmov _was_ a problem
>that was documented and mentioned in many discussion groups.

We were talking about any general changes Intel makes to the ISA.  I would
certainly include SSE in that category.

>>>Right.  Don't you think if they were _exact_ clones, just cheaper, they would be
>>>selling
>>>_more_?
>>
>>If that were so, WalMart brand tissues would outsell Kleenex, but I doubt they
>>do.  It's about brand name.  You would buy Intel anyway, and so would a whole
>>lot of other people, just because they're Intel.
>
>Absolutely _wrong_.  We buy them because the name is "intel" _and_ we have had
>good success using Intel.  They could be called stinkems instead of Intel and
>if the reputation had been built over the 25-30 years of microprocessors, then
>we'd all be buying stinkems instead.  The name is tied to success, and _that_
>is what leads me back to them.  I will occasionally try new things, but when
>things are critical, I want what works _first_ and I'll try something new when
>I'm playing, to see if it works.

That's the point.  Even if AMD made identical clones, you would probably still
buy Intel, because they have a good reputation.

Again, I give the example about Kleenex.  Why do you suppose Kleenex probably
sells far more than WalMart tissues, even though WalMart tissues are identical,
and a lot cheaper?

>>They were once follower who became market leader - the thing you said was
>>impossible.  That's the point of this statement.  I wasn't saying anything about
>>compatibility.
>
>
>Again you put words in my mouth.  A "follower" _can_ become a leader.  In the

_I_ put words in _your_ mouth?  Ha ha ha.  Let me quote you once again:

"And 2nd place is all that a follower can _ever_ reach..."

>1950's and early 1960's, Univac was the number one data processing vendor
>around.  They were displaced by IBM in the middle to late 1960's.  So it
>happens.  But it happend (in the case of IBM) because they were _not_
>following trying to make Univac boxes.  They made their own architecture,
>proved it was better and cheaper, and dominated the market.
>
>AMD apparently thinks it is more profitable to follow rather than to design
>a _good_ architecture that would catch on.  Heaven knows the X86 ISA is a piece
>of trash.  Kludge piled on top of kludge.  But it is what we have right now.

AMD is smart enough to know that designing a new ISA will fail.  It has always
failed.  Alpha failed on the desktop, despite being far technically superior.
PPC has pretty much failed.  MIPS, HP-PA, SPARC never even really tried.

It would take a miracle to overcome the x86 user base and supplant it with
another ISA.  Why do you think even _Intel_ hasn't tried it yet?

>How can they innovate when they have to faithfully follow the intel ISA?  3dnow
>is _dead_, for example.  Its dead because the ISA is intel's and nobody wants to
>write a piece of code that is AMD-only, because the intel side of the market is
>_way_ bigger.

I already mentioned x86-64 having lots of support, but you dismissed it and
claimed we were only talking about IA32.



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.