Author: Vincent Diepeveen

Date: 12:04:15 11/19/02

Go up one level in this thread

On November 19, 2002 at 14:34:21, Robert Hyatt wrote: Significant is that you first slow down your thing 2 times at a few points in order to get 33% faster later at the same points you first got 2 times slower. Anyway, by the time we all have 64 bits machines writing chessknowledge in bitboards is too cryptic anyway. Programming in a neat general way is always preferred above the bitboard hacking with inline assembly and things like writing only simplistic 1 line patterns IMHO. Best regards, Vincent >On November 19, 2002 at 14:24:04, Vincent Diepeveen wrote: > >>On November 19, 2002 at 14:11:18, Robert Hyatt wrote: >> >>>On November 19, 2002 at 12:25:11, Gian-Carlo Pascutto wrote: >>> >>>>On November 19, 2002 at 11:35:24, Robert Hyatt wrote: >>>> >>>>>Bitboards have a bit of a performance advantage on 64 bit processors, >>>> >>>>Proof? >>>> >>>>-- >>>>GCP >>> >>>Counter-proof? >>> >>>Seems intuitively obvious to me. Bitmaps seem to suffer _no_ performance >>>penalty on >>>X86 with 32 bits, compared to 0x88. Seems intuitively obvious to me that they >>>will pick >>>up speed on a machine that does 64 bit operations. >>> >>>Bruce and I did this comparison when he used the alpha in the WMCCC (1997 I >>>think). >>> >>>He re-compiled ferret for the alpha, did the same for Crafty. My speed >>>improvement was >>>significantly better than his on the _same_ machine. Because of the 64 bit >>>stuff. His program >>>didn't need any 64 bit stuff so it was wasted... >>> >>>Best "proof" is to try it. I have... >> >>But at that time i also proofed that you generated moves already 2.2 times >>slower than i did. If you then get 33% faster because of getting 64 bits, >>that doesn't proof anything. > >OK... let's pick a common architecture and generate moves from a position you >choose. What processor/speed do you want to use? I have a bunch here, from >400mhz >to 550 to 700 to 750 to 800 to 1.4ghz that I can get to easily. > > >> >>If we look at the speed at specint of crafty versus specint of K7, >>then we see that a 1Ghz alpha 21264c, which is their fastest CPU, >>is performing at the OFFICIAL benchmark like a 1.33Ghz K7. >> >>So if we roughly give you 33% for bitboards going from 32 to 64 bits, >>then that's simply the maximum you can claim for it. > >33% is significant, is it not??? > >And I get it _free_... to boot... > > >> >>The argument that specint is not allowing inline assembly is not valid. >>I do not use inline assembly in my program either (with exception of >>locking at the x86). So let's keep it a fair compare. >> >>So first a slowdown of a factor 2 (used to be 2.2) then winning back >>33% because of getting 64 bits, that isn't very impressive to me. > > >I know. 10% is impressive when you want it to be. 33% is not when you don't >want >it to be... > >When we had the discussion about null-move not hurting SMP performance, it >dropped >from 3.1x faster to 3.0 faster when null-move was turned on. That was >significant to >you. When we had the speedup discussion where I said Crafty was 1.7X faster on >a dual >and you said it was 1.6X faster, that was significant. But 1.0 vs 1.3 is not >significant, >because you don't want it to be??? > > > > >> >>It sure is intuitively VERY CLEAR to me that just >>having 1 bit of info spreaded over loads of bitboards is not >>a very good plan, because for any complex chess pattern you >>simply need more instructions than non-bitboarders do. > >It might be clear to you. Fortunately, for those of us doing this stuff, we are >not quite >that rigid in our thinking, and we find solutions to problems that you say can't >be solved. > >The flood-fill stuff was one interesting discussion that you will have a hell of >a time doing >in a non-bitboard program... there are other cases where bitmaps might be >worse. But there >are just as many where they are better. As I said I don't see _any_ advantage >except that on >64 bit machines they are faster due to increased data density.

- Re: significant math
**Robert Hyatt***13:09:47 11/19/02* - Re: significant math
**Russell Reagan***12:15:56 11/19/02*- Re: significant math
**Gian-Carlo Pascutto***12:37:24 11/19/02*- Re: significant math
**Russell Reagan***13:08:58 11/19/02*- Re: significant math
**Gian-Carlo Pascutto***13:18:48 11/19/02*- Re: significant math
**Sune Fischer***02:05:25 11/20/02*- Re: significant math
**Steffan Westcott***02:37:43 11/20/02*- Re: significant math
**Sune Fischer***02:44:10 11/20/02*- Re: significant math
**Robert Hyatt***07:57:01 11/20/02*- Re: significant math
**Sune Fischer***08:07:32 11/20/02*- Re: significant math
**Robert Hyatt***08:15:17 11/20/02*- Re: significant math
**Sune Fischer***08:39:14 11/20/02*- Re: significant math
**Steffan Westcott***13:52:38 11/20/02*- Re: significant math
**Sune Fischer***14:58:33 11/20/02*- Re: significant math
**Steffan Westcott***15:58:54 11/20/02*

- Re: significant math

- Re: significant math
- Re: significant math
**Robert Hyatt***11:11:57 11/20/02*

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Steffan Westcott***04:13:51 11/20/02*- Re: significant math
**Sune Fischer***04:35:46 11/20/02*- Re: significant math
**Steffan Westcott***06:37:11 11/20/02*- Re: significant math
**Gerd Isenberg***12:17:57 11/20/02*- Re: significant math
**Steffan Westcott***15:15:49 11/20/02*- Re: significant math
**Gerd Isenberg***23:32:50 11/20/02*

- Re: significant math

- Re: significant math
- Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***09:55:07 11/20/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Steffan Westcott***14:57:12 11/20/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***02:55:59 11/21/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Steffan Westcott***03:47:24 11/21/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***12:23:55 11/21/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Steffan Westcott***16:17:46 11/21/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***23:49:05 11/21/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Gerd Isenberg***04:42:51 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***08:53:54 11/22/02* - Re: Natural move generation with bitboards (was Re:significant math)
**Sune Fischer***08:37:56 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Gerd Isenberg***10:20:35 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Sune Fischer***13:17:29 11/22/02*

- Re: Natural move generation with bitboards (was Re:significant math)
- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***08:58:34 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Sune Fischer***09:19:42 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***09:52:54 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Steffan Westcott***16:08:27 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***16:44:52 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Steffan Westcott***17:19:09 11/22/02*

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)
- Re: Natural move generation with bitboards (was Re:significant math)
**Sune Fischer***10:38:50 11/22/02*- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***11:01:19 11/22/02*

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)
- Re: Natural move generation with bitboards (was Re:significant math)
**Alessandro Damiani***00:24:24 11/22/02*

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)
- Re: Natural move generation with bitboards (was Re:significant math)
**Daniel Clausen***03:55:11 11/21/02*

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: Natural move generation with bitboards (was Re:significant math)

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Russell Reagan***13:27:31 11/19/02*- Re: significant math
**Robert Hyatt***13:30:28 11/19/02* - Re: significant math
**Gian-Carlo Pascutto***13:29:49 11/19/02*

- Re: significant math

- Re: significant math
- Re: significant math
**Dan Andersson***13:14:07 11/19/02*- Re: significant math
**Russell Reagan***13:22:14 11/19/02*- Re: significant math
**Dan Andersson***20:54:40 11/19/02*- Re: significant math
**Robert Hyatt***22:06:34 11/19/02*- Re: significant math
**Dan Andersson***01:00:20 11/20/02*- Re: significant math
**Robert Hyatt***07:59:33 11/20/02*- Re: significant math
**Dan Andersson***13:52:11 11/20/02*- Re: significant math
**Robert Hyatt***15:34:35 11/20/02* - Re: significant math
**Sune Fischer***14:37:08 11/20/02*- Re: significant math
**Vincent Diepeveen***21:00:13 11/21/02*- Re: significant math
**Sune Fischer***03:24:41 11/22/02*

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Uri Blass***22:50:38 11/19/02*- Re: significant math
**Robert Hyatt***08:01:29 11/20/02* - Re: significant math
**Uri Blass***22:52:20 11/19/02*

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Gian-Carlo Pascutto***13:24:36 11/19/02*- Re: significant math
**Robert Hyatt***13:31:33 11/19/02*- Re: significant math
**Gian-Carlo Pascutto***13:33:40 11/19/02*- Re: significant math
**Robert Hyatt***13:42:21 11/19/02*- Re: significant math
**Gian-Carlo Pascutto***13:51:35 11/19/02*- Re: significant math
**Robert Hyatt***13:53:48 11/19/02*- Re: significant math
**Eugene Nalimov***14:15:46 11/19/02* - Re: significant math
**Gian-Carlo Pascutto***13:59:29 11/19/02*- Re: significant math
**Robert Hyatt***15:03:28 11/19/02*- Re: ftp news
**Robert Hyatt***15:37:42 11/20/02* - Re: ftp news
**Robert Hyatt***15:36:58 11/20/02*

- Re: ftp news
- Re: significant math
**Eugene Nalimov***14:16:28 11/19/02*

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Russell Reagan***13:38:03 11/19/02*- Re: significant math
**Robert Hyatt***13:43:31 11/19/02*

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Russell Reagan***13:31:15 11/19/02*- Re: significant math
**Alessandro Damiani***13:55:55 11/19/02*- Re: significant math
**Vincent Diepeveen***16:13:39 11/19/02*- Re: significant math
**Jeremiah Penery***17:23:11 11/19/02*- Re: significant math
**Alessandro Damiani***00:44:20 11/20/02*- Re: significant math
**Robert Hyatt***08:02:55 11/20/02*

- Re: significant math

- Re: significant math
- Re: significant math
**Russell Reagan***17:09:11 11/19/02*- Re: significant math
**Uri Blass***17:35:18 11/19/02*- Re: significant math
**Russell Reagan***19:24:36 11/19/02*

- Re: significant math

- Re: significant math
- Re: significant math
**Robert Hyatt***17:06:00 11/19/02*

- Re: significant math

- Re: significant math
- Re: significant math
**Gian-Carlo Pascutto***13:38:37 11/19/02*- Re: significant math
**Robert Hyatt***13:51:36 11/19/02*- Re: significant math
**Gian-Carlo Pascutto***13:56:35 11/19/02*- Re: significant math
**Gerd Isenberg***14:25:26 11/19/02*

- Re: significant math

- Re: significant math

- Re: significant math
- Re: significant math
**Bob Durrett***13:37:33 11/19/02*

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

- Re: significant math

This page took 0.31 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.