Author: Vincent Diepeveen
Date: 03:03:53 10/14/02
Go up one level in this thread
On October 14, 2002 at 04:19:46, Odd Gunnar Malin wrote:
>On October 14, 2002 at 04:04:30, Ingo Althofer wrote:
>
>>In the log-files of the ICC hour with Feng Hsiung Hsu
>>the following passages can be found:
>>
>>> EeEk(* DM) kibitzes: I heard that Fritz did
>>> play a match against Deep Blue in Hong Kong
>>> 1995,according to one of the Fritz programmers,
>>> is this not true?
>>> CrazyBird(DM) kibitzes:
>>> false advertisement. deep blue does not exist
>>> until 1996. the new chip was not completed
>>> until january 1996...
>>>
>>> ...
>>>
>>> EeEk(* DM) kibitzes: question from oddg: Back to
>>> the WC 1995, There was an entry with the name
>>> Deep Blue (Fritz won against DB), did it not have
>>> any relations to your Deep Blue? (EeEk: any idea
>>> how Deep Blue's name got in there, is this
>>> completely false?)
>>> CrazyBird(DM) kibitzes: deep blue did not play in
>>> 1995, since it did not exist yet.
>>> it was just deliberate relabeling on the part of
>>> commercial vendors, for obvious reasons.
>>
>
>You cliped a bit to early, here are the rest of his answer:
But he said it wasn't deep blue. that's nonsense it was deep blue.
whether it was the old dtii is logical. every year you
have new versions of your thing of course. 95,96,97 each year new
chips. but it WAS called deep blue. calling it something else now
is not nice. If it had won, it would be said different: "deep blue
dominated always and also against kasparov". That the level of the
games played by computers in 1995-1997 was still very low, the audience
has forgotten of course.
Relevant is that it was called deep blue. Not deep shit.
>"
>CrazyBird(DM) kibitzes: the program played was deep thought ii. which
>was vastly inferior to deep blue in chess knowledge as well search
>speed (1000 to 1 ratio in effective speed, 100 to 1 in raw speed)
>CrazyBird(DM) kibitzes: and we were as unlucky as kramnik is today:).
>"
the speed projections are not interesting. The number of nodes is not
the important factor here basically. There is so many things where you
lose to that the number of nodes a second is irrelevant.
Relevant is that the 1997 version got 12.2 ply on average. This was
WITH no progress pruning and WITH very dubious pruning last plies
(which however tactically doesn't miss something soon).
So it was seeing a lot less than DIEP is doing at 12 ply, with exception
of a few tactical jokes.
Diep doesn't forward prune positional moves at all. It's
only nullmoving.
Extensions are positional irrelevant. Sometimes they help to avoid losing
tactical lines. But at these depths even Hsu says he feels it doesn't matter.
I agree with him there.
You can definitely write down a statement from me that i am amazed they
forward pruned last few plies.
Of course that explains why they got 12.2 ply and in experiments i never
came above 10 ply much.
The real number of nodes deep blue 1997 got was estimated at 126 MLN nodes
a second on average by Hsu.
The effective speedup was not so high. I would estimate it on about 5%.
Hsu initially estimated 20% but later a percentile of 10 was quoted here.
It's less than that with so many chips doing so many things without hashtable,
killermoves etc.
hardware cpu1 doesn't help the next cpu either. So it really is a big loss
in efficiency.
that leaves for deep blue 1997 only a couple of millions of nodes a second
left. With a fullwidth search that's not so impressive.
I do not doubt however that the 97 machine was way better than the 96
machine. The only difference said Hsu was the evaluation. Yet that's
a very important difference.
Still we must see it in the time. For 1997, Hsu did an excellent job.
For 2002 standards, his thing is outdated everywhere.
- openings book a factor 50 smaller (hand tuned part) then
todays books. Also there is new openingstheory which would
kill it anyway.
- middlegame was 1997 standards. So many mistakes many positional
errors. 2002 programs do this a lot better. See world champs 2002.
- endgame was not so good. However we know from kramnik-fritz that
todays programs have major problems here too in technical respect,
yet do not forget that deep blue didn't have EGTBs in the hardware
search at all. So last few plies, also forward pruning added,
it didn't see any transposition to won EGTBs at all.
Best regards,
Vincent
>Odd Gunnar
>>
>>I had a look back into the June 1995 issue of the
>>ICCA Journal and found the following data which
>>might clearify things:
>>
>>* On p.97ff the contestants of the Hong Kong World
>>Championship are described. One of them is
>>"Deep Blue Prototype by Feng-Hsiung Hsu, ..."
>>
>>* On p.102 the Final Standings are given, with
>>"Pos.3 DEEP BLUE USA ...."
>>
>>* On p.130 of the same issue there is an announcement:
>> "The ACM Chess Challenge
>> World Champon Kasparov to play IBM's DEEP BLUE"
>>In this announcement it reads
>>"DEEP BLUE is being developed at the IBM ... by
>>Feng-Hsiung Hsu, Murray Campbell and Joe Hoane,
>>under the supervision of Chung-Jen Tan. An earlier
>>version of the system won the 1994 ACM International
>>Computer Chess Championship at Cape May, New Jersey."
>>
>>
>>So, the name "Deep Blue (Prototype)" was not a
>>"deliberate relabeling on the part of commercial
>>vendors" but the official name in 1995.
>>
>>Ingo Althofer.
This page took 0 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.