Author: Uri Blass
Date: 06:37:10 12/25/02
Go up one level in this thread
On December 25, 2002 at 09:00:09, John Lowe wrote: >On December 25, 2002 at 07:21:03, Uri Blass wrote: > >>On December 25, 2002 at 07:05:43, Richard Bean wrote: >> >>>movei: >>> >>>>>perft 5 13,853,661(0.922 seconds) >>>>>perft 6 115,892,741(38.505 seconds) >>>>>perft 7 17,650,572,059(time 1182.761 seconds on 1000 mhz) >>> >>>Crafty and my program output 470,702,875 and 17,650,572,059 - 470,702,875 = >>>2^34. I must forgotten a "long long" somewhere in the version I still have. >>> >>>Also chest gets the above perft result. >>> >>>Can someone verify my perft 10 result from the initial position - >>>69,353,847,899,771, or verify the number of unique nodes at plies 4 to 8? >> >>movei on 1000 mhz >>perft 4 197281 time 0.02 seconds >>perft 5 4865609 time 0.601 seconds >>perft 6 119060324 time 15.222 seconds >>perft 7 3,195,901,860 time 385.494 seconds >> >>You can see that for perft 8 I may need more than 2 hours >>I need some days for perft 9 and I need some months for perft 10 and many years >>for perft 11. >> >>Uri > >Forgive an observation from one who knows 0 (+/-6), but although doing the Full >Monty at ply 1 may usually be irrelevant it must be the cheapest irrelevancy in >computer chess. > >You would have needed a different position for timing your beasties. > >Cheers > >John He asked for the initial position It can work also for other positions. It can be faster but I use my full chess program and it does irrelevant calculations of scores that will be used for order of moves. 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.