Author: Uri Blass
Date: 09:43:19 06/24/05
Go up one level in this thread
On June 24, 2005 at 12:07:08, Andreas Guettinger wrote: >On June 24, 2005 at 10:16:01, Uri Blass wrote: > >>On June 24, 2005 at 09:34:43, Andreas Guettinger wrote: >> >>>On June 24, 2005 at 09:05:27, Volker Pittlik wrote: >>> >>>>On June 24, 2005 at 08:19:33, M Hurd wrote: >>>> >>>>... >>>>>I think this speed is to fast to determine which is best. A longer time control >>>>>would be better. >>>> >>>>There was also a black cat crossing my way during the test was running :-). >>>> >>>>Volker >>> >>>Introducing pruning into an engine (which makes it faster) and testing it at >>>blitz speed makes as much sense at selling poket calculators to mountain goats. >>> >>> >>>regrads >>>Andy >>> >>>PS: No offence meant, just my opinion. >> >>I disagree. >>Testing pruning at blitz speed can say something about the level of the engine >>in blitz. >> >>Of course it proves nothing about long time control but >>people are interested in blitz result and not only in long time control results. >> >>If something helps in blitz and does not help at long time control it is also >>interesting information. >> >>If you test only at long time control and not at blitz you may miss this >>information. >> >>Uri >>Uri > > >Ok, then I ask you directly. >Would you implement code into Movei that makes it 20 Elo stronger in Blitz but >20 Elo weaker at longer time controls? > >- Andy No but I think that I may learn from this information. If some change make the program weaker at long time control and stronger at blitz then it suggests that some change to a different direction may make the program stronger at long time control and weaker at blitz and maybe some tuning of the idea may do the program stronger at all time controls. 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.