TalkChess.com

How much Hash tables memory did Deep Blue use?

hammerklavier wrote:How much Hash tables memory did Deep Blue use?

I sent your question on to Murray Campbell. While we wait, I know that part of the search in Deep Blue was done in software. The software had access to Hash tables. At the end noes of the software search a hardware search would be sent to the special chips. The hardware part of the search did not have Hash tables:

http://ac.els-cdn.com/S0004370201001291 ... 47f33639d8

mjlef wrote:

hammerklavier wrote:How much Hash tables memory did Deep Blue use?

I sent your question on to Murray Campbell. While we wait, I know that part of the search in Deep Blue was done in software. The software had access to Hash tables. At the end noes of the software search a hardware search would be sent to the special chips. The hardware part of the search did not have Hash tables:

http://ac.els-cdn.com/S0004370201001291 ... 47f33639d8

I really wonder what techniques the top engines like stockfish/Komodo use which weren`t used by Deep Blue. Did DB use Null-Move/LMR/ProbCut ? Would be really intresting to know

If you are interested, there are several good reads on the subject, like "Behind Deep Blue".

I can't remember all the details. They implemented some cutting edge things, like conspiracy number search, but skipped on others because they didn't like that they introduced potential inaccuracies. I don't think they used null move at all for this reason, and this was well before LMR became popular.

It seemed like they had their hands full getting things stable and ready, and never got to the point of methodical incremental testing that is commonplace today. They depended on speed to cover some of those shortfalls.

I really wonder what techniques the top engines like stockfish/Komodo use which weren`t used by Deep Blue. Did DB use Null-Move/LMR/ProbCut ? Would be really intresting to know Razz

DB didn't use Nul Move, and probably didn't use LMR and ProbCut.
DB's pruning techniques were less intensive that the ones we have today.
In a speech Murray said DB's branching factor was 3 or 4, so as you can see there was not much intense pruning going on on DB's search.

Going a little off topic, I think it was so sad that team's work was ended so abruptly. Hsu (the chess chips designer) intended to design a much more powerfull version of the chip wich he hoped would defeat the world champion using it on desktop pc, so no need for a super computer. This means he intended to have a much more deeper search in the chip than 4 plies.
But IBM dropped funding and it was the end of the deep blue project.
Some say IBM was after it's stock value improvement.

Alvaro

Yeah. I don`t really like that IBM stopped the entire Deep Blue project. They couldn`t even claim that DB was the best chess player in the world since one match doesn`t really prove anything and many top chess players said, to put it mildly, that those games werent Kasparov finest.
They should have had more matches against Kasparov or other top chess players.

However, if they had NullMove and LMR it would have been a slaughter Amazes me how far we have come by not relying on hardware improvements !!!

CheckersGuy wrote:Yeah. I don`t really like that IBM stopped the entire Deep Blue project. They couldn`t even claim that DB was the best chess player in the world since one match doesn`t really prove anything and many top chess players said, to put it mildly, that those games werent Kasparov finest.
They should have had more matches against Kasparov or other top chess players.

At the very least, a tie-breaking match would have been sporting.

Murray Campbell reported to me that each SP had 1 GB or ram (A lot for its day). There were 32 nodes, so a total of 32 GB.

Mark

Murray Campbell reported to me that each SP had 1 GB or ram (A lot for its day). There were 32 nodes, so a total of 32 GB.

Thanks Mark!
32Gb is really fantastic ! in 1998 the most common configuration was 64Mb ram in Pentium MMX/Pentium II

CheckersGuy wrote:Yeah. I don`t really like that IBM stopped the entire Deep Blue project. They couldn`t even claim that DB was the best chess player in the world since one match doesn`t really prove anything and many top chess players said, to put it mildly, that those games werent Kasparov finest.
They should have had more matches against Kasparov or other top chess players.

However, if they had NullMove and LMR it would have been a slaughter Amazes me how far we have come by not relying on hardware improvements !!!

IBM did not completely stop the project, they did continue with Deep Blue Jnr for a while https://www.research.ibm.com/deepblue/p ... .6.4.shtml