Computer Chess Club Archives




Subject: Re: WMCCC Hardware

Author: Robert Hyatt

Date: 20:21:27 10/17/97

Go up one level in this thread

On October 17, 1997 at 18:23:47, Amir Ban wrote:

>Would like to add my six cents here:
>Some may be limited in hardware because they write assembler, but the
>issue for others is simply availability and logistics. I have a pure C++
>program that runs on Win32, which makes it probable that I could run on
>an Alpha quite easily. When I learned that I was going to compete
>against several Alphas, I put extra effort in my program and actually
>managed to shave some percents off, no small achievement for a mature
>program. But this overclocking business completely defeats me and makes
>me aware that I have been wasting my time investing in my program
>instead of going all out for some hardware.
>Again the issue of fairness is brought up instead of the right one. Is
>it fair ? Well, it's within the rules (not sure about the overclocking
>though), and everybody does it, but it is not good for the event. Sure,
>as Bruce points out, the ICCA events are very much about logistics and
>organization. The point is  that they shouldn't be. Another point is
>that this hurts the sponsor since everyone who is someone brings a
>machine that is not the sponsor's.

there are many more issues than this however.  This has *never* been the
"PC World MicroComputer Chess Championship"...  we've had overclocked
MC68000-based machines (mephisto), bitsliced (homebrew) versions of the
6502, and overclocked X86 (Fidelity) machines for years.  Each of these
architectures offers different features.  Now on to the alpha...

If you go back to the beginning of my work on Crafty, which dates back
exactly 3 years ago, now, you know that one of my specific goals for
"experiment" was to target development for a 64bit architecture, such as
Cray *or* the alpha.  Most of you have written your programs optimized
16/32 bit words.  I wrote mine optimized for 64 bit words.  It doesn't
reasonable to me that Crafty should have to be handicapped by having to
on a non-optimal architecture, "just because everyone else is."  So that
point 1:  a program ought to run on the architecture it is designed for.
don't know of any way to "equalize" the alpha and the X86...  it depends
the program.  Bruce gets 1.8x faster on his 533 than on the P6/200,
seems to get exactly 2.5X faster on a 500mhz machine than on the P6
on recent tests.  That's because I'm designed to pump 64bit words
while Bruce is not.

Second, alphas are *not* expensive.  They can be had for under $5000,
is less than I paid for my P6/200 18 months ago when I got it.  So this
*not* exotic hardware.  point 2.  It is readily available...

Next, the overclocking issue.  It was OK for Lang to overclock Mephisto,
was ok for Fedility to overclock chess challenger, and for Dave to use a
bit-slice (think it was him but I'm not sure) for what was back then
various things such as "superconny"...  So I don't buy the fact that
Bruce or Ernst have done something unconscionable here.  Crafty *might*
get a
766 machine.  It will definitely get a 500mhz machine.

next the concept of "carrying a custom machine under our arms."  If you
at the last WMCCC, *many* toted their own machines and it raised no
I think part of this is caused by the "shrinking gap" between commercial
amateur programs.  But, in crafty's case, we aren't carrying *anything*
to the
tournament.  DEC is supplying us a machine.  Kryotech will if possible.
this sounds *just like* what AMD is doing as well.  I'm not buying an
at present.  I'm using one supplied by DEC.  Others are using PC's
supplied by
AMD or whomever.  Somehow I get a sense of "those damned rich amateurs
vs us
poor commercial programmers" and I simply don't "get it."  I'm not rich.
Operator Jason Deines simply spent hours on the phone getting us an
alpha.  It
was hard work, not amateur/professional status that made this happen.
programmer *could* have done this, excepting those .asm special cases.
that was a poor choice they made.  I did that for the Cray, and suffered
it couldn't be ported to something else like a Fujitsu or Hitachi when
were available and a Cray wasn't.

>What do I suggest ? When the ICCA polled about the WMCCC invitations I
>suggested that the rules will contain a mild deterrent: Have anyone who
>brings his own machine, and can run on the supplied hardware, pay a fee
>for entering external hardware. Nothing too high. If this does not deter
>many, at least it will generate some revenue for the ICCA, and we may
>not have to see the ridiculous sight of "professionals" staying out
>because they cannot afford the entry fee while "amateurs" haul custom
>machines halfway around the world.

this "deterrent" is a stupid idea.  Just as the fee for commercial
is a stupid idea.  The idea is *not* to deter, but to encourage
The very concept of "a deterrent" leaves a foul taste with me.  Why
would we
want to deter *anyone* from competing?


If anyone wants to host a PC-only event, I'll participate.  I probably
attend unless it is reasonably short, but I'll make arrangements to have
operator ready.  If anyone wants to host an alpha-only event, count me
there too.  Ditto for a PowerPC event.  I don't mind "equal machines".
I do
mind getting my brains beat out because I have slow hardware and someone
has something faster.  I didn't like the alphas in the last event, nor
sun Sparcs in the one before that.  But they got in.  And in my case,
alpha is the machine of choice, since the Cray doesn't fit the micro
definition (yet).

However, if enough are unhappy, why not take it up at the tournament,
take some action for next year.  Make it PC-only for all I care.  I'll
participate.  But this "tone" (not necessarily Amir's tone but a couple
others too) seems "backward."  I'd think that "commercial" programs
be good enough to give up a factor of 2x or so and still do well?  Or
the amateurs really pretty close?  I have my opinion...

This page took 0.07 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.