Deep Blue

This forum is for general discussions and questions, including Collectors Corner and anything to do with Computer chess.

Moderators: Harvey Williamson, Steve B, Watchman

Forum rules
This textbox is used to restore diagrams posted with the fen tag before the upgrade.
Post Reply
ThatJimGuy
Member
Posts: 2
Joined: Wed Mar 18, 2015 12:15 am

Deep Blue

Post by ThatJimGuy »

Although Deep Blue was years ago, I would like to know if any of the newer computer programs have taken it on?
Reinfeld
Member
Posts: 486
Joined: Thu Feb 17, 2011 3:54 am
Location: Tacoma, WA

Post by Reinfeld »

Technically, it can't take any machine on - Deep Blue was a one-off piece of hardware, not a program in the way we think of it now. I explored this a while back, wondering if it was possible to find a Deep Blue "port," so to speak, but as far as I know, nothing of that nature exists.

I'll defer to other experts here who may have more knowledge, but that's my understanding of it. So measuring Deep Blue's strength against modern software becomes a matter of taking Deep Blue's games and testing them against modern software.

Worth noting that one of the Deep Blue versions (1995) lost a game to Fritz 3 and drew a game with WChess, a Kittinger program, perhaps equivalent in strength to a Novag Diamond, which plays at a 2150 ELO clip.

- R.
"You have, let us say, a promising politician, a rising artist that you wish to destroy. Dagger or bomb are archaic and unreliable - but teach him, inoculate him with chess."
– H.G. Wells
User avatar
spacious_mind
Senior Member
Posts: 3999
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Reinfeld wrote:Technically, it can't take any machine on - Deep Blue was a one-off piece of hardware, not a program in the way we think of it now. I explored this a while back, wondering if it was possible to find a Deep Blue "port," so to speak, but as far as I know, nothing of that nature exists.

I'll defer to other experts here who may have more knowledge, but that's my understanding of it. So measuring Deep Blue's strength against modern software becomes a matter of taking Deep Blue's games and testing them against modern software.

Worth noting that one of the Deep Blue versions (1995) lost a game to Fritz 3 and drew a game with WChess, a Kittinger program, perhaps equivalent in strength to a Novag Diamond, which plays at a 2150 ELO clip.

- R.
Hi R,
WChess while if it is the same as Novag Diamond (which I don't know) on a PC is a lot stronger than 2150 ELO. More like 2500+.

Best regards
Nick
Carl Bicknell
Member
Posts: 252
Joined: Mon May 04, 2009 10:06 pm

Post by Carl Bicknell »

spacious_mind wrote:
Reinfeld wrote:Technically, it can't take any machine on - Deep Blue was a one-off piece of hardware, not a program in the way we think of it now. I explored this a while back, wondering if it was possible to find a Deep Blue "port," so to speak, but as far as I know, nothing of that nature exists.

I'll defer to other experts here who may have more knowledge, but that's my understanding of it. So measuring Deep Blue's strength against modern software becomes a matter of taking Deep Blue's games and testing them against modern software.

Worth noting that one of the Deep Blue versions (1995) lost a game to Fritz 3 and drew a game with WChess, a Kittinger program, perhaps equivalent in strength to a Novag Diamond, which plays at a 2150 ELO clip.

- R.
Hi R,
WChess while if it is the same as Novag Diamond (which I don't know) on a PC is a lot stronger than 2150 ELO. More like 2500+.

Best regards
Yeah, I was thinking the same thing. WChess would trounce the Diamond.

Also the tournament was just a tiny game sample. If WChess & Fritz 3 were forced to play a match against Deep Blue, they would surely lose.

IIRC one of the objections the Deep Blue team had to those events was that there were only something like 5 rounds making it into a bit of a lottery. They were consistently claiming (for both Deep Blue I & II I believe) 90% against PC programs, something which I believe.

That still leaves even the 1997 Deep Blue II miles behind today's engines, of course.
Reinfeld
Member
Posts: 486
Joined: Thu Feb 17, 2011 3:54 am
Location: Tacoma, WA

Post by Reinfeld »

Just a footnote here - I suggested a rating test (using Nick's methods) based on a game between Deep Thought II (early version of Deep Blue) in 1991. The link is here:

http://hiarcs.net/forums/viewtopic.php?t=7029

As Nick's continued testing efforts have shown, one way to measure the strength of engines is to measure their moves against modern software (and our dedicated collections), using a scoring system. In this way, it becomes possible to derive comparisons between the Deep Blue/Deep Thought iterations and other systems.

- R.
"You have, let us say, a promising politician, a rising artist that you wish to destroy. Dagger or bomb are archaic and unreliable - but teach him, inoculate him with chess."
– H.G. Wells
User avatar
spacious_mind
Senior Member
Posts: 3999
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Hi R,

That would work. The game in the other post, is that the game you want to see tested?

Thanks
Nick
Reinfeld
Member
Posts: 486
Joined: Thu Feb 17, 2011 3:54 am
Location: Tacoma, WA

Post by Reinfeld »

Nick -

Correct. The Deep Thought game in the other post comes from the Daniel King book. It has the benefit of suggested scores (by King), but your method augments that and finds errors.

- R.
"You have, let us say, a promising politician, a rising artist that you wish to destroy. Dagger or bomb are archaic and unreliable - but teach him, inoculate him with chess."
– H.G. Wells
User avatar
ricard60
Senior Member
Posts: 1285
Joined: Thu Aug 09, 2007 2:46 pm
Location: Puerto Ordaz

Post by ricard60 »

Deep Blue version from 1997 won over the strongest human player by that time and world champion Garry Kasparov. Kasparov el in 1997 i believe was over 2800 elo
Post Reply