Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: WMCCC - may the best man at getting the fastest hardware win :(

Author: Robert Hyatt

Date: 20:59:09 10/19/97

Go up one level in this thread


On October 19, 1997 at 17:53:19, Thorsten Czub wrote:

>On October 19, 1997 at 16:19:55, Robert Hyatt wrote:
>
>>After posting one response this morning, something kept gnawing at the
>>back of my mind, "something is wrong here."  But I couldn't put my
>>finger
>>on it.  Later it hit me.  I went to DejaNews, and did a search on
>>subjects
>>with "NPS and match" in them.  And found the long thread where Ed and I
>>were debating the importance of NPS.  Ed was on his "tactical
>>sufficiency"
>>bandwagon, and lo and behold, I found posts from Chris and Thorsten
>>right
>>in the middle, all saying the same thing.  NPS doesn't matter.
>
>When I have program A on a pentium120 and the same program A on a
>p2/200, and B beats A of course, what is to be said about this ??
>That NPS is unimportant ?
>That B is stronger than B ?
>
>From my point of view NPS in unimportant. And we will show this with
>CSTal.
>And others show this with THEIR programs that will have less NPS than
>yours.
>
>I followed Chris threads here in CCC.
>
>I think he is not right.
>
>It doesn't matter which machines you have.
>It does not help you in Den Haag , and it will not help much anyway.
>In a few weeks or months AFTER Paris the opponent programs WOULD develop
>strategies against YOUR programs IF your programs would be sold
>somewhere.
>

my program is "sold" *everywhere*.  To this date in time, there have
over
170,000 *different* host machines that have downloaded one or more
copies/
versions of crafty.  That's 170,000 different people at least, and quite
likely more since many use some pretty common ISP's...



>It is easy to create ANTI-Ferret programs or ANTI-crafty programs, I
>guess.
>The only fact this does not happen (like it happens against all other
>commercial programs Hiarcs, Mchess, Rebel, Genius, Fritz, CSTal etc.)
>is:
>Your programs do not compete anywhere !


You should re-state this.  I competed at Jakarta last year, will be at
Paris
this year, and played in at least two good human events this last year,
winning
the Pan Am, and playing in a round robin with 5 GM players...

Maybe I don't compete where *you* compete.  But I do compete.  I have
played
more games on the chess servers than all the commercial program added
together
and multiplied by 10X.  So I'm pretty visible, and I know how I (crafty)
plays.


>
>This is - from my point of view - one main reason you are so impressed
>by your NPS. Because you don't compete.
>If you would compete, you would lose like all others lose.

Care to lets post a list of opponents and number of games played over
this
past year?  I'd start with over 150 games against GM Larry Christiansen.
 Plus
Shabalov, Shirov, Karpov, Ivanov, Roman, Shaked, and a dozen other GM's
I have
forgotten...


>
>
>
>>
>>Now, suddenly, NPS does matter.  And I suppose I am confused as to what
>>has
>>really happened:
>>
>>1.  You didn't really believe any of the things you were saying in the
>>NPS debate, but was really only saying them to keep the debate raging.
>
>
>You are confused. Indeed.
>
>>
>>2.  You did believe the NPS argument (your side of it) but later changed
>>your mind.
>
>
>What is the sense of an olympic event when some american or german guys
>are doped with anabolica or Dec Alpha hardware ?
>It only proofs: not any female is really female. Some have a dongle,
>where there should be no dongle. But thats dope.
>
>>
>>I don't see any other explanation.  In light of your comments on the NPS
>>argument, and those of others, this alpha-issue seems to be a non-issue.
>>*if* you believe what everyone but *me* was saying...
>>
>>
>>Next topic...
>>
>>I don't know why Hiarcs withdrew.  I have an opinion.  If you want to
>>hear
>>it, here goes:
>>
>>  Hiarcs was withdrawn due to fear of losing.  It is currently on top of
>>the SSDF list.  It is probably beginning to sell decently.  A disaster
>>in
>>Paris could reverse that quickly.  I strongly believe that any
>>commercial
>>program that chooses to avoid the WMCCC is doing so solely for
>>commercial
>>reasons.
>
>I think you are wrong concerning Hiarcs. I don't like these: he is
>feared or he is feared discussion. I have seen hiarcs and rebel and
>mchess and others compete in many tournaments and championships. And
>they did a good job. I have not seen crafty or ferret or other programs
>there too. Were you afraid to lose ??

kind of hard for us to compete.  The programs didn't exist.  Crafty is
not yet
three years old.  Bruce is a little older, but only maybe a year or so.
So that
question is pretty stupid.  But *I* have competed with a computer chess
program
for almost 30 years now.  I *never* ran and hid.  *never*...



>
>
> I don't like to lose either.  However, there is one difference
>>between the way I operate and the way the commercial programmers
>>operate:
>>
>>Cray Blitz won the WCCC in 1983 and 1986.  By the time 1989 rolled
>>around,
>>however, I *knew* we wouldn't win again, unless there was some
>>unprecedented
>>luck involved, because of Deep Thought.  I had trouble finding machine
>>time.
>>I nearly decided to forget about it.  But after thinking about this, it
>>didn't
>>seem fair to Hsu and company, because if they won without Cray Blitz
>>being
>>there, they would have had to endure the "If CB had been there this
>>would
>>have been different."  I kept after Cray, they finally found a customer
>>that would let us use their machine.  We did, DT won, and there was no
>>doubt that they were best.  Contrast that to the commercial mind-set for
>>a
>>minute.  Do *everything* based on the bottom line of sales and public
>>opinion.  They didn't go last year.  For the same reason they aren't
>>going
>>this year.  And it doesn't have a thing to do with new versions or
>>anything
>>else other than not wanting to take a chance.  Perhaps in another year
>>this
>>will be amateur only.  Doesn't matter to me...  I can think of a couple
>>of
>>amateur programs I fear more than any of the commercial programs already
>>and this is going to get better (or worse depending on perspective.)
>>
>>So I don't believe for a minute that the "alpha issue" ran Hiarcs off,
>>nor
>>kept Genius away, nor Rebel, nor any other program.  I think it is a
>>matter
>>of "fear" more than anything else.
>
>
>I don't think Mark is feared to lose, nor do I think Ed is afraid or
>Marty or somebody else. I can only laugh about your statement.
>maybe you are afraid, but I don't believe Mark would be afraid to play
>against Ferret or crafty. In fact hiarcs played against ferret once ...
>

and against crafty many times.  I'd be happy to post the results if
you'd
like to see them.  They are probably not quite what you are expecting
however.


>
>>Unfortunate, because I am just as
>>afraid
>>of doing badly as anyone else.  But *not* so afraid that I won't play.
>
>And this is right for all the other you call feared.
>

hmmm... then *where are they?*???



>
>>If
>>Crafty loses every game, I'll get to enjoy a year of razzing.   But it
>>will
>>at least be there.  I'm willing to put my money where my mouth is...
>
>The money question was not the reason for mark, and not your
>fear-claims.

right...  and not for Richard...  and not for Ed, even though he
explicitly
stated that he was afraid of these events because amateurs cooked his
book.
I use a different book for tournaments than I do for casual play.  I'll
keep
my opinion for the present...



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.