Author: Uri Blass
Date: 20:39:18 05/05/01
Go up one level in this thread
On May 05, 2001 at 21:00:39, Robert Hyatt wrote: >On May 05, 2001 at 11:33:45, Uri Blass wrote: > >>On May 05, 2001 at 10:36:31, Robert Hyatt wrote: >> >>>On May 05, 2001 at 08:25:56, Uri Blass wrote: >>> >>>>There is a possibility to set selectivity of Deep Fritz to 0 and it means that >>>>Fritz avoid null move pruning. >>>> >>>>The interesting question is how many effective plies are lost from null move >>>>pruning. >>>> >>>>Some results suggest that not very much >>>> >>>>Tiger14(depth3)-Deep Fritz(selectivity=0,depth 3)35-15(6 draws) >>>>Tiger14(depth4)-Deep Fritz(selectivity=0,depth 4) 31-19(8 draws) >>>>Tiger14(depth 8)-Deep Fritz(selectivity=0,depth 8) 20-23(12 draws) is going to >>>>be finished in an hour. >>>> >>>>The results without null pruning were not significantly better and at depth 3 it >>>>is not clear if selectivity=0 is productive because tiger could win only >>>>34.5-15.5 against selectivity=2. >>>> >>> >>>Null-move gains at least 2 plies. But it loses some as well, as a 3 ply >>>search with null move is not as good as a 3 ply search without... >> >>The results that I got do not give significant difference for Deep Fritz at >>depth 3 but maybe I did not do enough games. >> >>The results clearly suggest that 5 or 8 plies without null move are better than >>5 or 8 plies with null move. > >That makes perfect sense. Just watch the times. The null-move search will >finish in seconds while it takes the non-null search minutes... Yes, but it is still interesting to know how much better in order to translate depth without null move pruning to depth with null move pruning. > > > >> >>Some new results for Fritz *without null move pruning* in the nunn2 match >>The difference seems to be bigger when the depth is bigger >> >>Every match is 50 games >> >>Deep Fritz(depth 5)-Tiger14(depth 5) 25-25(14 draws) >>Deep Fritz (depth 8)-Tiger14(Depth 8) 27.5-22.5(13 draws) >> >>Deep Fritz(depth 6) and Deep Fritz (depth 9) without null move did better >>results against the same oponents so it seems that Fritz does not lose a full >>ply even at depth 9 from pruning but again I did not do enough games. >> >>Uri > > >I'm not sure it makes sense to compare null-move and non-null-move at the same >depth. That is a _huge_ time handicap match... I do not compare it in order to check if null move is good fro Deep Fritz but in order to find some formula to translate depth with null move pruning to depth without null move pruning. For some reason for Fritz there it no big difference at depth 3 and the difference starts to be significant at bigger depthes and still the handicap is probably less than a full ply at depth 9. My result for Deep Fritz(depth 3) without null move pruning is the following against Tiger(depth 3,4,5) Tiger14 depth 3-Deep fritz depth 3 35-15 Tiger14 depth 4-Deep Fritz depth 3 43-7 Tiger14 depth 5-Deep Fritz depth 3 46-4 For comaprison I got for the default Fritz the following results(it scored only 1/2 out of 150 less than Fritz without null move pruning): Tiger14 depth 3-Deep fritz depth 3 34.5-15.5 Tiger14 depth 4-Deep Fritz depth 3 42.5-7.5 Tiger14 depth 5-Deep Fritz depth 3 47.5-2.5 I found that the default Deep Fritz at depth 3 won more games when Fritz without null move pruning got more points from drawing games. 18 draws out of 150 games for Deep Fritz without null move pruning. 9 draws out of 150 games for Deep Fritz with null move pruning. 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.