Author: Brian Katz
Date: 02:45:03 06/08/04
Go up one level in this thread
On June 08, 2004 at 03:48:43, GuyHaworth wrote: > >KRPKQ has maxDTMs = 68 (1-0, wtm) and 104 (0-1, btm) which is deep but not in >'16-bit EGT-entry' territory. > >The 'P' gives 3.8x the number of positions because only one symmetry is 'on'. > >Neither stat should account for the long time you are running. > > >I think Eugene has upgraded his use of RAM to a bit-mode rather like Ken >Thompson's use of RAM. This cuts RAM requirements by a factor of at least 4. > >In 'bit-mode', you should only need 200MB of RAM for your program. Otherwise, >you will need at least 800MB which may not be available. > >If the latter, you will be bottlenecked by the disc I/O activity. Have you run >any 5-man P-endgame EGT generation with no 'like men' successfully? > > >Others know the details better than me but I think the above is in the ballpark. > >g At the moment, I am running the Tablebase Generator to create files that were not loaded from the 9 disc set, because I reinstalled my O/S and the copies of Generated Tablebase files that I had on DVD were no longer on the DVD I had copied to months ago. At that time, I also Generated the missing files. I don't remember if one particular file took as long as I have been referring to. Perhaps I just did not realize that one file in particular took so long, thinking I was just waiting for a number of files. This time around, I just happened to notice that there is one file taking this lond. Obviously, it Generated last time. The only other possibility that I can think of.......Is that for some reason this particular file did not load off of the CD......And last time it did. Perhaps I should go back to my CD Set and check to see if the KRPKQ files are on one of the 9 dics I have. I have 2 sets of 9 discs due to an error of a double mailing when purchased. Perhaps one of the 2 sets is missing something due to a defective CD. I remember last year that one of the 2 sets did not load properly from particular CD, so I used the other, not thinking anything of it. I will have to look into that. At the moment, it has gone beyond 53 hours. It is now upto Starting Major Iteration 35. I know in the past the Generations took about 4 or 5 days to complete the missing files. Perhaps not realizing that one set of files took more than half of the time. Brian
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.