Author: blass uri
Date: 04:55:24 01/24/00
Go up one level in this thread
On January 24, 2000 at 07:36:41, Alvaro Polo wrote: >On January 24, 2000 at 06:32:40, blass uri wrote: > >>On January 24, 2000 at 05:37:21, Jeremiah Penery wrote: >> >>>On January 23, 2000 at 19:40:47, Tom Kerrigan wrote: >>> >>>>On January 23, 2000 at 19:29:31, Jeremiah Penery wrote: >>>> >>>>>>No, the only reason you think DB has a bunch more evaluation is because Hyatt >>>>>It has nothing to do with Hyatt. Please try to construct better arguments in >>>>>the future. The reason I think DB has a better evaluation is that I've seen the >>>>>games, and analyzed them. Kasparov and other GMs have said that DB was clearly >>>>>superior to anything else they've seen. >>>> >>>>Of course it's superior, it searched 200M NPS. Searching has the property of >>>>"adding knowledge" to a program. How do you know that you were seeing evaluation >>>>function terms in those games, and not tactics that are so deep that they're >>>>hidden to humans? >>> >>>So why didn't he think this of DB-1? It did about 100M NPS, but he ended up >>>crushing it (see game 6 of the first match). The NPS didn't seem to help so >>>much there. Obviously, there was a lot more knowledge in DB-2. >> >>Obviously 200M is bigger than 100M. >>It is also possible that the evaluation of DB-2 was better than DB-1 but it does >>not prove that it was better than the commercial programs. >> >>Uri > >I have a comment here. Let's suppose that DB evaluation is, in fact, not better >than the commercials'. I believe that state of the art commercials, such as Deep >Junior, are doing around 1,000,000 NPS. So, DB2 has an advantage of 200 times >the speed. It would be possible to set an experiment to see if Deep Junior >thinking at 10 hours a move (200x3 mins) could be better than Kasparov. Just >take some good games from K and simulate that Deep Junior is the opponent and >look at Deep Junior answers at 10h/move. This way you could see if DJ would have >beaten him, catching Kasparov in a blunder, or if DJ would produce uncorrect >replies that would have lead to a defeat from Kasparov. The point is, 200 times >is not such a huge factor (it is big but simulable). > >Alvaro It is not possible to do it because deep Junior is going to analyze positions not from the opening book of it and even if it does mistakes it prove nothing because deep junior may be in different positions in games. Uri
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.