Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: best chess programmers

Author: Robert Hyatt

Date: 18:25:39 07/20/00

Go up one level in this thread


On July 20, 2000 at 20:42:21, Tom Kerrigan wrote:

>On July 20, 2000 at 18:05:19, Bruce Moreland wrote:
>
>>On July 20, 2000 at 13:44:12, Tom Kerrigan wrote:
>>
>>>On July 19, 2000 at 21:24:27, Wayne Lowrance wrote:
>>>
>>>>On July 19, 2000 at 21:05:18, walter irvin wrote:
>>>>
>>>>>my best 5 chess programmers
>>>>>1.deep blue team (deep blue)
>>>>>2.richard lang (all were good + 8 titles!!!!!!)
>>>>>3.amir ban (deep junior)
>>>>>4.frans morsch (fritz)
>>>>>5.ed  (chess machine  and rebel)
>>>>>
>>>>>there are a few that get left off the list ,either they did not win a title or
>>>>>they just could not keep pace with the better programs .
>>>>
>>>>Dont you have a spot for Dr Rober Hyatt on that list ? I could not begin to give
>>>>you the correct order except Dr Hyatt has got to receive major
>>>>attention/consideration !
>>>
>>>Hsu, Lang, Morsch, and Ed all have incredible history behind them. Hyatt
>>>doesn't. I don't really consider Amir to be a legend (yet) and his name is
>>>easily replaced with a number of others. Kittinger, Stanback, Bruce Moreland,
>>>Christophe, Stefan, Uniakle, de Koening; sorry if I left anybody out. But I
>>>consider any of these guys more impressive than Hyatt.
>>>
>>>-Tom
>>
>>Glad to see there's nothing personal going on on your end of the Bob - Tom
>>equation.
>>
>>Bob gets credit for a lot of stuff:
>>
>>1) Writing Cray Blitz.  Was it the best program ever written?  Would it have
>>performed against modern micros?  Who cares!  It was there, when it was there,
>>it won two championships, it got into the news, it promoted its sponsor, and Bob
>>gets credit for putting it all together.
>
>Two points on this:
>1) Was Bob the only person responsible for CB? I thought he had partners. Who
>knows how much he really contributed to its strength. Lang, Morsch, etc. were on
>their own.


This has been documented many times, assuming you read any computer chess
literature that discusses Cray Blitz.  I wrote _all_ the code.  Harry's
contribution was rewriting serious parts of the FORTRAN into assembly to make
them faster.  in the last 3-4 years of CB's life I took over this as Harry
had gotten a little older, and had burned out as it was hard work.  Bert
Gower was responsible for most of the opening book.

CB's "code" was mine.  And mine alone...




>2) CB ran on a Cray. If I'm not mistaken, "Blitz" was not extremely impressive.
>If CB's competition was also running on Crays, who knows how it would have done.



Just like if everybody ran on DB's hardware they would be better than DB?  This
is such a childish argument.  Do you know anything about supercomputers and
vector processing and the Cray machine in particular?  I didn't think so.  You
don't just drop a piece of code onto a Cray and it goes fast, unless it is a
matrix-type code.  Chess is not.  Harry and I spent years looking at ways to
make various parts of the program vectorize, from move generation to evaluation.
It took thousands of hours of effort.




>
>>2) Writing Crafty.  Crafty isn't the world champion, but who cares, it's
>>obviously a high-end program and it's open source!  It's been downloaded by a
>>zillion people who either want to play against it or learn from the source, and
>>thousands of people have played against it on the Internet and are playing
>>against it right now.
>
>Making your program open source is not a way to be a "great programmer." It
>takes exactly zero effort to make a program open source.
>


It does take more effort to "beat it" though, doesn't it?  It is open source
_and_ very strong.  With plenty of innovations from rotated bitmaps to being
the first SMP (PC) program running...

Not particularly spectacular as I have done SMP programming forever.  But
also not 'chickenfeed'...




>>3) Being an Internet authority.  He has something to say about essentially
>>everything technical.  He says it not to show how smart he is, or to put others
>>down, but because he wants to help people solve problems and make their chess
>>programs better.  If you ask Bob a question you get an answer, and it's the best
>>answer he can give you, and he'll do work to get you the answer.  And this is
>>not just a recent thing, he's been doing this since the Internet came of age and
>>before.
>
>I've seen Bob misunderstand/misread questions and post unrelated answers
>(sometimes with bad data) so often that I think this argument is bogus too.
>Anyway, what does answering questions have to do with being a great programmer,
>either?

So.  I have seen you do the same thing.  I have seen you post arcane insults
rather than answer simple and direct questions...  I don't see your point...




>
>>4) Researching and publishing.  He's published useful articles on Cray Blitz, in
>>a field where most published articles are not useful, especially early articles.
>> He's also published several articles about Crafty and about general computer
>>chess topics such as diminishing returns in search and parallel search.  This is
>>stuff that anyone can learn from and many have.  Any computer chess library will
>>contain articles written by Bob.
>
>I'd say this is a gray area between being a good chess programmer and being a
>good person. Sure, okay, publishing papers might get him in the running for a
>top-5 position. But when you compare that to some of Lang's achievements, it
>looks pretty weak.
>
>-Tom




Which WCCC events did Richard win, again??? I seem to have missed them.  Winning
that event has _never_ been easy.  Only one program has done it twice in a row.

:)



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.