Computer Chess Club Archives




Subject: Re: WMCCC Hardware

Author: Bruce Moreland

Date: 10:53:24 10/18/97

Go up one level in this thread

On October 18, 1997 at 04:37:02, Amir Ban wrote:

>Not trying to deter anyone from competing, just trying to protect the
>interests of the real amateurs between us (I'll probably exclude myself)
>who may be the most talented programmers in the world for all we know
>and had done their best to bring a great program. Nobody told them this
>was not the issue. Also want to reassure those who read the clause about
>"generally available single processor" and thought to themselves that at
>least they are in the same ballpark only to learn that there are some
>tricks they hadn't heard of.

Yes, there are some tricks, like reading the July '97 issue of PC
Computing, which has an Alpha on the cover ( if you
want to see a small picture of the cover).  The 500mhz version of the
machine I bought is on page 89.  I also had the IDE version of the
machine on the cover, but I sent it back because I liked the platform so
much I decided to pay more for a 533.  I would have bought another
Enorex machine, but they had a shipping delay, so I changed to Polywell.

Another trick is talking to peers and being observant.  Dark Thought has
been bringing Alphas forever.  They are pretty quiet, but they will talk
to you a bit if you ask directly.  Also, there was for a long period of
time a Crafty on ICC running on an Alpha (the one on the cover), and he
kibitzes search stats including node rate sometimes.  I noticed that
this number was rather large.

Nobody pays attention to Dark Thought, because they haven't won the
event, and they are quiet.  But they did well twice in a row.  They will
eventually win this event.

>Not suggesting a PC event at all. SOS has always run on Sun. DarkThought
>was born on Alpha. Let them bring their native machines with no penalty
>(I would even consider supporting them to make the competition fair).
>You make some case that Alpha is your natural environment. Fine. Bruce
>has a much weaker argument. Guys who bring Pentium 120 to Pentium 90
>events have no case at all and the ICCA should blast them, or at least
>make them pay something so that Hiarcs may also have a chance to compete
>once in a while.

This is ludicrous.

I first ran on Alpha in approximately 1994, and I ran on a Power PC in
1995 (I have macros that handle byte-ordering issues).  You have not
heard about these versions because neither of them was as fast as the
PC's I had at the time.  I also have a command-line version that will
run independently of Windows.  There was a time I thought that I would
need this, but I never did.  When I do need it, it is there.

I can make an argument that anything I run on is "planned", since I have
written my application to be portable.

Also, this is another effort to distinguish based purely upon
instruction set.
There are at least three manufacturers of chips that run x86 natively.
Nobody has questioned the legality of using the non-Intel chips (and in
fact, one of them is sponsoring the event).  The architecture of any of
these chips (including Intel's) is probably quite different from the
architecture of that 486/33 you used to use on ICC, and they are
probably quite different from each other.  Nobody questions the legality
of your moving to the newer versions of these chips, because there is
binary compatibility between these chips.  But if I change instruction
set, while running the same software I was running before (I have to
look at the light on the switch above my monitor in order to know
whether I am on the Alpha or the P6), there is outcry, and there
shouldn't be.


PS.  Perhaps next year, I will be on a Power Mac.  I haven't
investigated them at all yet.

This page took 0.01 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.