Computer Chess Club Archives




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

Author: Robert Hyatt

Date: 07:11:13 03/01/03

Go up one level in this thread

On March 01, 2003 at 01:53:11, Jeremiah Penery wrote:

>Please go read this:
>I'm not going to try arguing the economic points anymore.  The post above may
>help you understand what I may have failed to explain well.

That is not exactly a place I would cite in a technical paper, any more than
I would cite a post here. :)

>On March 01, 2003 at 00:07:01, Robert Hyatt wrote:
>>On February 28, 2003 at 20:41:59, Jeremiah Penery wrote:
>>Sparc isn't selling "so well".  IN fact, if you talk to Sun insiders, it is
>>doomed and they are moving to the PC world quickly.  They already use a PC
>>type chassis with IDE disks and the like.  Care to guess why?  Processor sucks.
>>Everybody knows it sucks.  Only the "sun loyalty" keeps a few coming back.  We
>>used to be 100% sun, for example.  We now have 5 out of 250 computers here.
>>The rest are mostly PCs with a few others (SGI, etc) thrown in for good measure.
> and
>"In 2002, Sun was No. 1 in sales of non-Intel servers.
>It had 37% of all non-Intel server revenue, according
>to researcher International Data Corp.

So?  37% of 5% is what?  PCs dominate the computer market, reaching near
95% of all computers.  So Sun is fighting to compete for some fraction of
5%?  And they are going down the hole.  Sparc is doomed, and sun has already
said this.  They will be X86-based within two years...  although it might be
Intel or AMD of course...

>Hewlett-Packard was No. 2, with 31% of those sales.
>IBM Corp. came in No. 3 with 21%. And Fujitsu was No.
>4, with just 2.6%"
>So by revenue, RISC server market share breakdown is
>40% SPARC
>31% PA-RISC + Alpha
>21% Power/PowerPC
>The 31% HP share is likely about two thirds PA-RISC
>and one third Alpha. SGI probably takes a good chunk
>of the 8% unaccounted for.

But that entire segment is 5% or less of the total sales.  Probably less as
mainframes are in that 5% category as well...

>>The alpha was too expensive for the desktop.  256 bit bus was a killer.  But
>They planned to release it for the desktop.  It wasn't all that expensive,
>either.  With FX!32, they could even run native Intel binaries, so software
>compatibility shouldn't have been a barrier.

There were some cheapo versions released.  Polywell was one example.  But it
was a dog because it used the PC 64 bit bus so it could use the PC memory and
PCI cards.  And it was a flop.

>>that is _all_ that was wrong with it and give things another 3-5 years and
>>64 bit chips _will_ be the norm.
>What does this statement have anything to do with the discussion?  Of course, I
>agree with this, especially because of Hammer's imminent release.
>>>_You_ were the one who said it first, not me.  "But reducing the die size has
>>>other advantages, including lower voltage and lower heat, at the same clock
>>>frequency, so there is a reason for going there.."  I just tried to respond to
>>What's wrong with that statement?
>Nothing's wrong with that statement, except that it wasn't the subject of the
>discussion.  You made the point, I replied to it, and now you say _I'm_ the one
>getting off-topic?  Sheesh.
>>  There is _no_ reason to go to reduced die
>>size except to (a) shorten distances, (b) shorten switching times, (c) reduce
>>power requirements, and (d) reduce heat dissipation.  You can go to a smaller
>>fab without going faster, if your only goal is reduced heat, for example.  But
>>we _were_ talking about speed, and nothing else.
>So why did _YOU_ bring up the heat/power argument?

I didn't.  I responded to it.

>>>The biggest potential number for headroom I may have given was 25%.  That may be
>>>a bit high, but not implausibly so.
>>And if that is all you are saying, then why are you arguing, because I haven't
>>said anything different.  which leads me to wonder who is _really_ skimming.  :)
>I'm arguing it because you keep claiming I said something else.
>>>So what are you harping on about here?  _Nobody_ ever claimed that they build
>>>first and then test.  Nobody disputed that the engineers should have a good idea
>>>of the maximum clock speed.  All I've said is that they may be able to produce
>>>something, but that doesn't mean they will sell it.
>>I'm only "harping" on how engineering works.  It is _not_ as haphazard as you
>>and others would suggest.  The engineers know very well what a particular fab
>Please give me one example of someone saying anything remotely implying
>something like, "engineering is haphazard."

Just read thru this thread.  The discussion about "max speed".  I claim they
know before the first chip rolls off the fab.  Others claim that they don't know
until they test.

>>process and design will do.  It might take a while to get there, but there is
>>no guesswork at all...
>>ANd that is what I have been saying over and over.  Steve agreed.  As did any
>>other engineer I have talked to over the years...
>You act like engineers determine everything about product releases, though.
>Unfortunately for the rest of the world, this isn't how it works.  Marketing
>determines far more than they probably should.
>>>You've already bought the program that's 200 points better than the competitors.
>>> The question is, "When do you 'need' to buy a replacement program?"  If you
>>>already have something twice as good as any competing product, it's very likely
>>>that your buying cycle time will dramatically increase.
>>Not if they come out next year with something 200 points better.
>If you suddenly came across an idea that improved the strength of Crafty by 300
>SSDF points overnight, what makes you think you could repeat the improvement
>next year, or ever?
>>Will I buy a 3ghz machine today and buy a 3.2ghz machine in 6 months?  No.
>>But the risk is will I buy that 3.0ghz if your competition is right behind
>>you in speed and significantly below you in cost (AMD vs Intel for example).
>It seems to work for Intel so far, in that they still have almost 90% of the
>>>The better off you are _today_.  If companies focused only on today, they would
>>>fail tomorrow.  In your business model, the companies sacrifice long term sales
>>>and revenue for a quick injection of cash.  That won't sustain anyone.
>>I still don't see how this is an issue.  How will producing a slower product
>>today help me tomorrow?  Once I lose a customer to a competitor, it is _much_
>>harder to get them _back_.  I'd want to offer the best that I could offer, to
>>drain _their_ customers that need more performance.
>>IE Cray _never_ played these games, _ever_.
>If we were talking about the high-performance, server-oriented RISC market, or
>supercomputer market, your comments about marketing may well be correct.  We're
>talking about the commodity x86 market, however, which is completely different.

I believe it is different _only_ in the volume of sales..

>>>If he's designing a several-million square foot, 100 story office tower,
>>>compared to the architect 30 years ago designing a 1500 square foot, 2 story
>>>house, the job is certainly NOT easier for the human today, no matter what tools
>>>he has.
>>But he isn't.  He is designing a small city, made up of one of these, one of
>>those, two of those, etc.  Divide and conquer and all that.  It isn't just one
>That doesn't make it any easier for the individuals.  They not only have to
>design their own building now, they have to make sure it integrates well with
>all the other buildings.
>>>That doesn't mean the silicon isn't capable of running at the same speed as the
>>>desktop part.  That's the crux of this entire argument.
>>No it isn't.  The crux of the argument is "can the desktop processor, with the
>>full setup for heatsink and fan and power supply and so forth run faster than
>>the engineers say?"
>Define 'full setup for heatsink and fan and ...'  I would guess that desktop
>parts are probably often thermally limited, rather than silicon limited.
>They're released at certain speed grades taking into account near worst-case
>thermal conditions.  With better cooling, they can be run closer to the limit of
>the silicon, rather than simply what the thermal characteristics allow with
>mediocre cooling.
>>The answer seems to be "no" according to popular (engineering) opinion.
>You've heard from a couple engineering friends of yours, and asked one guy on
>this forum who gave a somewhat non-commital response.  Hardly an overwhelming
>opinion, considering there are opposing viewpoints from people also in
>engineering and chip design and marketing.
>>Once again, I do not _care_ about the intentionally slowed down processors
>>and whether they will overclock or not.  I care about the front-line fastest
>>chips being produced _only_.  All my comments are addressed to that specific
>>segment of the chip market.  Not the low-head (mobile) processors.  Not the
>You were the one who brought up the laptop processors.  I only tried to respond
>to the point.

Once again I did _not_ bring up the laptop processors.  I responded to something
about "heat dissipation".  I've _never_ worried about low-end or mid-range
chips. Only top-end.

>>re-marked slower-clocked processors made for an economy niche.  The best of
>>the processors _only_ is what I have been talking about, and I have _not_
>>been vague in that position whatsoever...
>Except that you brought up things like laptop processors, Cray, chess programs,
>house building, etc.  All I've done is try to respond to your points in kind.
>>All attempts to change the chip topic will be returned to the main idea,
>>time after time.  :)
>Then maybe you should stop changing the topic, so I don't keep sounding
>off-topic to you. :P

This page took 0.02 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

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