Author: Robert Hyatt
Date: 22:42:29 05/11/01
Go up one level in this thread
On May 12, 2001 at 00:58:28, Uri Blass wrote: >On May 11, 2001 at 22:19:51, Robert Hyatt wrote: > >>On May 11, 2001 at 18:09:42, Joshua Lee wrote: >> >>>On May 11, 2001 at 10:54:04, Robert Hyatt wrote: >>> >>>>On May 11, 2001 at 09:47:17, Joshua Lee wrote: >>>> >>>>>Kasparov,G - Comp Deep Blue Game 6 >>>>> >>>>>[D] rbr5/1p1q1p1k/pPP1pBp1/P2pP2p/3Q3P/6P1/4RP2/2R3K1 b - - 0 1 >>>>> >>>>>I've looked at this position every so often and wonder how come no program wants >>>>>to play pawn takes? I have looked at the analyzed game and either way Black is >>>>>lost much earlier. This i believe can only show how far programs are away from >>>>>being on this level as some people have made the arguement of "Well if this >>>>>program could search that many nodes or more" I think i asked about this as well >>>>>but this should clearly show that even if your computer searched twice as many >>>>>nodes as Deep Blue or even Deeper Blue it may find deeper tactics but it is >>>>>still looking at the position and evaluating it differently and not as well. The >>>>>same goes for any of us Compared to a GM or IM they Think about each position >>>>>Differently than we do. >>>> >>>> >>>>One obvious reason is it gives white a protected passed pawn on the 6th rank. >>>> >>>>Another is there are obvious tactical problems after bxc6. Qc5 is one >>>>issue to deal with. Rec2 is another... Etc. !!!!! DB played pawn takes, the alternative analysis given in the Fritz6 database is for Rxc6. >>> >>>This would just hint that Deep Blue is NOT better than commercial programs >>>positionally and that is not true. In fact if i remember correctly the move made >>>is thought to be worse than the alternative atleast by fritz and hiarcs so does >>>this mean that the depth shows something the programs don't see or is it >>>something else. >>> >>> >>>The point i was trying to make is Deep Blue Made this move but most software >>>available will not. I guess i didn't explain myself well enough, I did say >>>however that Deep Blue was lost prior but i want to know why Commercial Programs >>>will not play the same move as DB shouldn't they after some time? There has to >>>be some reason that shows the move played slightly better than the other so is >>>it that besides speed the program is better or what? >> >> >>Was this from the first match or the second? > >This was from the first match but I do not see the point because both moves are >losing. > >The fact that deep blue played a different move does not prove nothing. >Both Rxc6 and bxc6 are losing moves and different extensions can encourage a >program to play a different move. > >We do not have the logfile of deep blue from these games so we can know nothing >about the line that Deep blue saw and if it changed it's mind from Rxc6 to bxc6 >so it seem that using this position to compare between the first Deep Blue and >the commercial programs is useless. > >Uri That's why I asked. First, DB1 had several shortcomings that were addressed in the new chip used in DB2. Second we had no output from DB1 as you said, so there is no way to know what it saw and what the eval was (of course, for DB2 it is not easy to know what it saw either since it can not display all (or even most) of the PVs.
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.