Author: Christophe Theron
Date: 15:30:21 05/02/01
Go up one level in this thread
On May 02, 2001 at 02:32:27, Chessfun wrote:
>On May 01, 2001 at 14:12:44, Christophe Theron wrote:
>
>>On April 30, 2001 at 23:11:05, Chessfun wrote:
>>
>>>On April 30, 2001 at 02:55:13, Christophe Theron wrote:
>>>
>>>>On April 29, 2001 at 23:58:58, Chessfun wrote:
>>>>
>>>>>On April 29, 2001 at 13:16:26, Christophe Theron wrote:
>>>>>
>>>>>>I said several times I had no plan to release a SMP version of Tiger. I did NOT
>>>>>>say I did not have one. I did NOT say I have never worked on SMP.
>>>>>>
>>>>>>I also said that I could change my mind if the market demands it. Last time I
>>>>>>said so was in a live interview on the CSS forum. Go to to this forum, go to the
>>>>>>page where the 16 hours interview has been reported, and you will find my words
>>>>>>about this.
>>>>>>
>>>>>>It was BTW before I heard about the Kramnik match.
>>>>>
>>>>>
>>>>>Christophe,
>>>>> Following is cut and pasted from the CSS Sprech-Stunde
>>>>>this was April 8.
>>>>>
>>>>>At this time it was known of the upcoming match-up with Kramnik.
>>>>>Not in detail but it was known.
>>>>
>>>>
>>>>
>>>>The amusing thing about this is that on April 8 Enrique told Ed about the SMP
>>>>requirement, but asked Ed to keep the information hidden to me.
>>>>
>>>>That's exactly what Ed did, because he had agreed to keep the information
>>>>secret.
>>>
>>>>So on April 8 I knew close to nothing about this.
>>>>
>>>>So you are right, I knew about it, but not in details. I did not know about the
>>>>SMP requirement.
>>>
>>>
>>>That was a question I asked you. It was already written and known that
>>>it would be SMP otherwise how would I have known it was 8 cpu's?
>>>See below.
>>>
>>>
>>>>We have asked Enrique why he asked Ed to keep the SMP information secret, even
>>>>for me, and he has not answered.
>>>>
>>>>
>>>>
>>>>
>>>>>If you had a SMP version why did you not answer this question in
>>>>>that manner?. Or assuming secrecy ;-) then contact BGN as suggested
>>>>>in a follow-up about having an SMP Tiger.
>>>>
>>>>
>>>>
>>>>I had no idea the deadline was so close. We had just learned about it, and I had
>>>>not even talked with Ed about it.
>>>>
>>>>I did not know about the SMP requirement. Actually I knew very little about this
>>>>match.
>>>
>>>
>>>At the time, I myself never knew very much, only what was posted here at CCC.
>>>The question I asked below, was directly relevant to the tournament being run
>>>to decide Kramniks challenger. As it was already known it would be an SMP
>>>capable program.
>>>
>>>
>>>>>Sarah Bird: Do you feel that either Tiger 14.0 or Gambit 2.0 running on say a TB
>>>>>1.5 Ghz, would compete better against the World Champion than Deep Fritz running
>>>>>on 8 processors.? Naturally assuming your answer is yes :-) then what benefit
>>>>>would there be to organizers who promote multiple cpu's in using either of your
>>>>>programs?
>>>>>
>>>>>CT: If it is about promoting multiple CPUs, then... Well in short I do not
>>>>>understand why Gambit Tiger has not even been considered as a potential
>>>>>candidate.
>>>
>>>>Once again the basic problem is that we have not been asked if we could enter
>>>>the match or not.
>>>>
>>>>It has been assumed that we could not enter the match, and that's all.
>>>
>>>
>>>It would seem a natural assumption.
>>>Even now I have seen nothing that makes me think that a Tiger SMP
>>>version actually exists. Has been tested and fully ready to play.
>>>
>>>All I have seen is we did some work on it a while back.
>>>To say we can produce one is different than we have one. I cannot see
>>>how it could be possible to delay the tournament while Rebel produce an
>>>SMP capable version.
>>>
>>>
>>>>The same applies for Nimzo, Hiarcs, Gandalf... Did Enrique ask them if they had
>>>>a SMP version that could compete?
>>>
>>>
>>>That statement by implication says Rebel do.
>>>Again IMO it makes no sense to ask ever program in say the top ten
>>>of the SSDF if they have a SMP version available to compete in a
>>>tournament for the right to play Kramnik. Would any say NO I think not.
>>>They would all attempt to produce such a thing within a few days ;-)
>>>
>>>
>>>>I know the answer because Enrique told me in email.
>>>>
>>>>HE DID NOT ASK THEM EITHER.
>>>
>>>As I said I see no reason he should ask them.
>>>
>>>>What other evidence do you need that the selection process has been arbitrary?
>>>
>>>The selection process, actually to me seems pretty natural.
>>>
>>>>>------------------
>>>>>Sarah Bird: Personally at this point why would Gambit be a candidate? It hasn't
>>>>>yet proved to be stronger than any of the main competition. The organizers
>>>>>whoever they are, if they are interested in promoting dual processors what is to
>>>>>be gained for them with a single cpu playing?
>>>>>
>>>>>CT: Playing videos with the other CPU? ;)
>>>>>------------------
>>>>
>>>>
>>>>
>>>>Please note that it was a joke...
>>>
>>>I knew it was a joke, but the point is simply either you have an SMP Tiger
>>>or you don't. There is no middle ground here for we can produce one, or
>>>we will have it in a few days or weeks, or we was working on one and it
>>>was kinda almost finished. Either it exists or not.
>>
>>
>>
>>Where Deep Fritz and Deep Junior the commercial versions?
>>If not I should be allowed to enter a version that is not commercial, i.e. a
>>prototype.
>
>
>Of course you know they weren't. However they did have existing versions.
>
>
>>I would have entered a prototype version of SMP Tiger if I had been given a
>>chance to.
>
>
>I am not sure that I would consider either of the Deep's to have be a prototype.
>Both had versions on the market. Deep Junior the commercial version is 1-1/2
>so it's natural to assume they have a modified version.
>
>
>>>You never answered any of those questions as if one existed, and since I put
>>>the questions directly relevant to the Kramnik match and tournament it was
>>>clear that SMP was a requirement.
>
>
>>A SMP version of Tiger exists. Actually existed last year.
>>A friend of mine went on a long hollyday last year, and instead of leaving his
>>dual system unused at home, he lent it to me.
>>I developped a SMP version on top of Chess Tiger 12.2.
>>But as I already told several times SMP was not a priority for me. So once I had
>>SMP working, I disabled the code in Tiger's source code. I did not erase it of
>>course, I just deactivated the relevant parts of code.
>>Since then, Tiger has evolved on many points, including a different internal
>>move coding and a different hash table structure.
>>So now I cannot just reactivate the SMP code. It will not work directly. I need
>>to reactivate it and adapt it to what has changed in the latest Tiger engines.
>
>>It's not a lot of work. The principle of the SMP algorithm does not have to be
>>rewritten, it's just a matter of adapting the code.
>
>>I just asked for a little delay in order to do that and to double check that
>>everything was working.
>
>>You can have your doubts about this SMP version, but after all if I send a SMP
>>Tiger that crashes in Cadaques, I will be the one to look stupid, not the
>>organizers of the Kramnik match.
>
>Actually I have seen to many of your ideas these past few months actually come
>to pass to doubt either your word that one existed, or that you could have got
>it working with the additional changes since 12.2 within a stated timeframe.
>
>It is a shame if one had existed that after April 8 at CSS the organizers were
>not contacted immediately.
I guess that can be called a mistake, but you must realize that at that time we
had no idea about the tight schedule and the SMP requirement.
>>Completing the DF-DJ match has taken a full week, and that was the amount of
>>time I have asked for. It was possible to run the DF-DJ match, and then let the
>>winner play against SMP Tiger.
>
>That really don't sound right as that implies Tiger gets a bye into the final.
>Naturally you could have run a tourney of 24 games then insert Tiger to play
>both the Deeps.
What I mean is that they would have get the program after the DF-DJ match, and
they could have done whatever they wanted in order to achieve a fair match.
Now that we know a little bit more about this, I realize that anyway their
schedule was already fixed, and that the offer from Enrique to include SMP Tiger
(if he could get it on Monday 23rd) was a fake. Tiger would have never been
allowed to compete anyway.
>>Some people have said that I have kept a "secret" around the SMP Tiger. That's
>>not really the spirit. I have no plan to commercialize the SMP Tiger in the near
>>future, so I was not trying to keep a secret in order to create a surprise.
>
>
>No I understand what your saying. You have always said SMP wasn't something you
>were in a rush to get done. But now what we actually have is a working 12.2 that
>you say you could have got ready in 2 weeks. One week prior to the match between
>the Deeps then it would be ready.
Again, on the day we learned about the match I had no idea about the tight
schedule and the SMP requirement.
And I also have another project running at this time, so I wasn't going to stop
working on it and start working on SMP Tiger again when I did not know what this
match was about.
I'm not sitting there waiting for match opportunities all the time, wondering
what's the next requirement I'll have to fulfill in order to be allowed in, you
know.
>With you previous statements on SMP what made you originally spend the time
>implementing it in 12.2?
After the release of 12.0 and the good result on the SSDF I considered several
options for the future. One of them was SMP.
>>Actually I am currently working on another project, which has a higher priority
>>for me, and this one can be called a secret. It will not be a secret anymore in
>>a few days. Actually I think a few people can already guess what I am working
>>on.
>
>My guess. A Tiger for Palm !
:)
No comment.
Christophe
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.