Computer Chess Club Archives

Search

Messages

Subject: Re: By the way, here is an instance where crafty's node counts truncate

Author: Robert Hyatt

Date: 10:36:11 09/22/99

Go up one level in this thread

On September 22, 1999 at 02:29:00, Richard A. Fowell (fowell@netcom.com) wrote:

>On September 21, 1999 at 13:38:56, Dann Corbit wrote:
>
>>On September 21, 1999 at 13:20:59, Dann Corbit wrote:
>>[snip]
>>>              time=600:00  cpu=99%  mat=0  n=2230832056  fh=1%  nps=61952
>>>              ext-> checks=663307593 recaps=20560034 pawns=12911665
>>>1rep=21027250 thrt:296212
>>>              predicted=0  nodes=2230832056  evals=1175553882
>>>              endgame tablebase-> probes done=48311  successful=44616
>>The actual node count will be something around 6 billion.
>
>Looks like rollover of an unsigned 32-bit int. It's a common symptom - at
>least Crafty doesn't seem to be using signed 32-bit ints, which leads to the
>ever-popular negative node count. MacChess went to 64-bit ints
>a number of revs back because of this issue, which I first noted when
>running 3-day analyses.
>
>Richard A. Fowell (fowell@netcom.com)
>http://dmoz.org/Games/Board_Games/Chess/Software/Macintosh/


There is a problem doing this for me.  64 bit ints are not yet ANSI-standard.
If a compiler supports them, you can use them.  But _if_ you want to print them,
portability becomes an issue.  IE in gcc I can use %lld to print one and it
works just fine.  But that breaks other compilers.  If we can get a group that
can give me a working 'format specification' that works on each compiler that
supports this, I can perhaps #define something and switch the node counter to
a long long/_int64...

Bob

• Re: By the way, here is an instance where crafty's node counts truncate Dann Corbit 12:18:07 09/22/99
• Re: By the way, here is an instance where crafty's node counts truncate David Eppstein 11:32:00 09/22/99
  • Re: By the way, here is an instance where crafty's node counts truncate Robert Hyatt 12:25:16 09/22/99
    • Re: By the way, here is an instance where crafty's node counts truncate Andreas Stabel 02:00:04 09/23/99
      • Re: By the way, here is an instance where crafty's node counts truncate Robert Hyatt 06:53:00 09/23/99
        • Re: By the way, here is an instance where crafty's node counts truncate Heiner Marxen 11:07:03 09/26/99
          • Re: By the way, here is an instance where crafty's node counts truncate Robert Hyatt 13:25:44 09/26/99
            • Re: By the way, here is an instance where crafty's node counts truncate Eugene Nalimov 16:15:37 09/26/99
            • Re: By the way, here is an instance where crafty's node counts truncate David Eppstein 15:29:37 09/26/99
              • Re: By the way, here is an instance where crafty's node counts truncate Robert Hyatt 18:48:04 09/26/99
                • Re: By the way, here is an instance where crafty's node counts truncate David Eppstein 19:39:30 09/26/99
            • Re: By the way, here is an instance where crafty's node counts truncate Heiner Marxen 15:05:19 09/26/99
  • Re: By the way, here is an instance where crafty's node counts truncate Dave Gomboc 12:00:12 09/22/99
    • Re: By the way, here is an instance where crafty's node counts truncate Dave Gomboc 12:00:47 09/22/99
      • Re: By the way, here is an instance where crafty's node counts truncate Dann Corbit 14:24:41 09/22/99
        • Re: By the way, here is an instance where crafty's node counts truncate Robert Hyatt 17:02:46 09/22/99
          • The upcoming C9x Standard Dann Corbit 17:22:42 09/22/99
• Re: By the way, here is an instance where crafty's node counts truncate Eugene Nalimov 11:19:50 09/22/99

This page took 0.03 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.