Author: Sune Fischer
Date: 10:24:16 01/19/04
Go up one level in this thread
On January 19, 2004 at 12:16:47, Daniel Clausen wrote: >On January 19, 2004 at 11:43:21, Sune Fischer wrote: > >[snip] > >>IMO the PV is a non-essential feature, having to do it "correctly" messes >>up the code and slows down the engine. > >The few lines (like Bruce suggests on his site) mess up the code? I don't see >how.. it's seems to be very clear and straightforward to me. I could never get it to work correctly, there were problems with terminating the PVs. E.g. when you return repetition draw or a mate score you need to terminate the PV if you are on a PV, otherwise you are going to update and copy a too long PV. It can probably be solved, but IMO it is messy. > Not sure about the >speed at the moment, but I'd be surprised if it's that much. It shouldn't be that much overhead, I think I measured it to less than 1%. It might depend on how good the move ordering is. >IMHO, >xboard-related code or checks whether someone tapped something on the keyboard >in the middle of search() mess up the code a bit more. :) Yes, there is already enough to struggle with, no need to add increased complexity :) > >>That's just more than I want to spend on a fun-to-have feature. > >I can understand this POV, but if I'd buy a commercial product and want to use >it for analyzing, I wouldn't be too happy if the PV is useless (except for the >1st move) I understand your POV as well, but like I said I tested it carefully and saw only very rarely problems. The longer the search the bigger the strain on the hash and the bigger problems (the shorter the PV), but for standard time controls I saw no problems with it. -S. >YMMV-IMO (tm Dann Corbit :) >Sargon
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.