Author: Robert Hyatt
Date: 20:27:33 08/21/02
Go up one level in this thread
On August 21, 2002 at 10:27:53, Vincent Diepeveen wrote: >On August 21, 2002 at 09:28:03, Uri Blass wrote: > >Read section 'dual threat extensions' Uri in the document. > >the cpu's searched 4 to 5 ply. *not* 6 ply ever in hardware. >note that 4 to 5 ply is already too inefficient. > >what i really wonder about already for many years about is that >in contradiction to the Brutus team, who are printing out exactly >how many searches they do from software to the hardware (so >how many searches are done in hardware) the deep blue team never >mentions in any log file any number regarding it. So? I once asked Ken Thompson "Ken, how many nodes are you searching?" He responded "I don't know. It would take as long to increment a counter as it would to search a position. I don't want to waste that much time." Having the chips count nodes would simply be another issue in the design process. Hsu had instrumentation to measure the duty cycle of each chess processor. From that he could directly compute the number of nodes each chess processor searched. Summing would give him that. The uncut log files had some of that buried in it. That is how he came up with the 350M peak number among others. I only downloaded and saved the "clean" logs as the raw logs had way too much crap in them for me. > >All their nodes a second figures are a big guess simply IMHO. Fortunately, YHO is meaningless here. > >They nowhere measured it! Sure they did. As I mentioned above. > >Despite that IBM sold the thing as the fastest machine on the planet >(read well: not the deepest searching). they nowhere print it in the >logfiles. Their 126 million nodes a second estimate is weird in itself. You simply don't have the raw logs. I used to have one of them but I don't think I kept it... It was on the web site for a while but I think the raw files were removed due to lack of interest by anyone. > >It is a guess in the air. Actually, that is more of what you like to do than them... at least so it seems to this observer... > >>On August 21, 2002 at 09:00:19, Terry McCracken wrote: >> >>>On August 21, 2002 at 08:22:06, Vincent Diepeveen wrote: >>> >>>>On August 21, 2002 at 08:13:47, emerson tan wrote: >>>> >>>>>wHATA IS DEEP BLUES AVERAGE PLY LEVEL ON TOURNAMENT TIME CONTROL, I UNDERSTAND >>>>>THAT IT CALCULATES 200 MILLION POSITIONS PER SECOND BUT NEVER HEARD OF ITS >>>>>AVERAGE PLY. THANKS >>>> >>>>they estimate it at 126 million nodes a seconda gainst deep blue. >>>> >>>>It had a nominal search depth (depth limit) of 12 ply. With a lot of >>>>tactical extensions a lot of lines were searched at 17 ply though. >>>> >>>>that's however true for all chess programs. I averaged in DIEP way >>>>deeper. It depends basically whether you count hashtable cutoffs with >>>>the depth or not. I tend to not count them. In hardware there was not >>>>a hashtable. >>>> >>>>Average search depth says nothing. The nominal search depth is more important. >>>>This was 12 ply. So the weakest link were lines of 12 ply simply. >>>> >>>>Best regards, >>>>Vincent >>> >>> >>>Deep Blue II did searches with extensions of 17 ply + at 3 min. per move and >>>wasn't uncommon to actually hit 22 plys. >>> >>>Dr. Robert Hyatt watched this with his own eyes, so I'm sure he'll have >>>something relevant to say on this matter. >> >>Nobody watched the tree of deep blue. >> >>I remember from previous post that people in the deep blue team >>said that the brute force depth was only 12 plies. >> >>The output of deeper blue also does not say 17 plies. >>There are people who claim that 12(6) means 18 plies >>brute force but the deep blue team never claimed it. >> >>I do not know what it means. >> >>It may mean 12 plies brute force with some selective search >>of the next 6 plies(of course there were extensions but >>the number 12 6 are not about extensions because >>it is clear that the longest lines that they searched >>were clearly more than 18 plies). >> >>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.