Author: Bo Persson
Date: 02:54:53 08/21/03
Go up one level in this thread
On August 20, 2003 at 19:45:52, Sune Fischer wrote: >On August 20, 2003 at 06:15:14, Bo Persson wrote: > >>On August 20, 2003 at 05:28:59, Omid David Tabibi wrote: >> >>>On August 19, 2003 at 22:11:14, Robert Hyatt wrote: >>> >>>>On August 19, 2003 at 20:06:58, Mathieu Pagé wrote: >>>> >>>>>Hi, >>>>> >>>>>The fact: >>>>> >>>>>I have this question i read at some place that it is faster to unmake a move >>>>>than to save the state of the game before moving then restoring it when we want >>>>>to unmake the move. >>>>> >>>>>For the moment my engines did not implement unmake() (it is still buggy). >>>>> >>>>>My thougth: >>>>> >>>>>Since bitboard computation are slow (on 32 hardware) i think that it can be >>>>>slower to unmake the move than to save the state. I friend of me that is lot >>>>>better than me at optimizing code also think that. >>>>> >>>>>My questions: >>>>> >>>>>Are you all using unmake() function or there is some of you that found that >>>>>saving the state is better ? >>>> >>>> >>>> >>>>read the comments from Crafty in main.c. I started out using what is >>>>commonly called "copy/make" as that worked well in Cray Blitz. But it >>>>didn't work well in the PC. The PC has very limited memory bandwidth, >>>>when you compare the speed of memory to the speed/demands of current >>>>processors. If you keep the board in cache, and update it there, it is >>>>more efficient than to copy it from real memory to real memory... >>>> >>>>> >>>>>In the case that unmake is better. by wich factor ? >>>> >>>> >>>> >>>>For me, just faster. Not 300% or any such number, but at least 25% if >>>>not more. >>>> >>>> >>> >>>Up until now almost all my data was global, updated in make_move and takeback. >>>However, the more data structures I add, the more costly takeback becomes. Now >>>I'm trying to dynamically update my attack tables in make_move, which is a very >>>costly operation. Even though the attack tablea are about 256 bytes, I think >>>copying them will be far cheaper than resotorig in take_back. >>> >>>So, it seems that one should experiment with each piece of data to find out >>>whether resotring in takeback is cheaper than copying the structure. >>> >> >>It all depends on the size of you data. Bob has a lot of bitmaps, most of which >>do not change during a single move. Copying knight and bishop maps during a pawn >>move seems like overdoing it. >> >>Late in the game they might even be empty and can be allowed to drop out of the >>cache. > >Probably not, since they would still be a part of move generation and some eval >terms. Could be. > >I don't see an efficient way of rooting out empty boards entirely, and that is >unfortunate since it is pretty common for a large part of the game to be played >without all types of pieces. I was thinking more about how silly it is to copy the empty bitboards for each ply. If you update the boards that are active, they will stay in the cache. Those that are not used might drop out, unless they are copied once every micro second. Bo Persson
This page took 0.12 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.