Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: could someone benchmark my checkers program?

Author: martin fierz

Date: 08:44:53 06/11/04

Go up one level in this thread


On June 11, 2004 at 08:11:09, Anthony Cozzie wrote:

>On June 10, 2004 at 17:31:54, martin fierz wrote:
>
>>On June 10, 2004 at 12:10:48, Anthony Cozzie wrote:
>>
>>>On June 09, 2004 at 17:59:14, Dann Corbit wrote:
>>>
>>>>On June 09, 2004 at 17:32:32, martin fierz wrote:
>>>>
>>>>>aloha!
>>>>>
>>>>>there's a computer checkers world championship coming up, and i would like to
>>>>>participate. the best machine i have at home is an AMD XP2400+, which was good a
>>>>>year ago but of course is now already a bit dated. there is no real reason my
>>>>>program should run faster with an AMD64, but you never know... so:
>>>>>
>>>>>if anybody with an AMD64 has a spare minute, could you download checkerboard
>>>>>(www.fierz.ch/CB_setup.exe), install it, turn off the opening book of the
>>>>>engine, set the level to 1 minute, and start the engine and send me the result?
>>>>>the engine (cake) produces a file called cakelog.txt which i would like to have
>>>>>to see how fast it's running on modern computers compared to what i have (of
>>>>>course, if you have a P4 > 3GHz, or an XP3200+ or any other hot machine i'd also
>>>>>be interested in your results...).
>>>>
>>>>OS Name	Microsoft(R) Windows(R) Server 2003, Standard Edition
>>>>Version	5.2.3790  Build 3790
>>>>OS Manufacturer	Microsoft Corporation
>>>>System Name	DCORBIT64
>>>>System Manufacturer	To Be Filled By O.E.M.
>>>>System Model	To Be Filled By O.E.M.
>>>>System Type	X86-based PC
>>>>Processor	x86 Family 15 Model 4 Stepping 10 AuthenticAMD ~2202 Mhz
>>>>BIOS Version/Date	American Megatrends Inc. 1001.005, 2/9/2004
>>>>SMBIOS Version	2.3
>>>>Windows Directory	C:\WINDOWS
>>>>System Directory	C:\WINDOWS\system32
>>>>Boot Device	\Device\HarddiskDmVolumes\Dcorbit64Dg0\Volume1
>>>>Locale	United States
>>>>Hardware Abstraction Layer	Version = "5.2.3790.0 (srv03_rtm.030324-2048)"
>>>>User Name	CORPORATE\dcorbit
>>>>Time Zone	Pacific Daylight Time
>>>>Total Physical Memory	1,024.00 MB
>>>>Available Physical Memory	560.10 MB
>>>>Total Virtual Memory	3.41 GB
>>>>Available Virtual Memory	2.38 GB
>>>>Page File Space	2.41 GB
>>>>Page File	C:\pagefile.sys
>>>>
>>>>It was getting 92% CPU.
>>>>
>>>>book hashtable with 400000 entries allocated
>>>>93393 moves in opening book
>>>>parsing db\db2.idx
>>>>parsing db\db3.idx
>>>>parsing db\db4.idx
>>>>index files parsed
>>>>allocated 1 KB for indexing
>>>>allocated 2000 KB for DB cache
>>>>allocated 0 KB for block pointer array
>>>>allocated 23 KB for LRU linked list
>>>>dbfp[3] is null!
>>>>dbfp[4] is null!
>>>>dbfp[5] is null!
>>>>
>>>>
>>>> w w w w
>>>>w w w w
>>>> w w w w
>>>>- - - -
>>>> - - - -
>>>>b b b b
>>>> b b b b
>>>>b b b b
>>>>position hex bm bk wm wk color:
>>>>{0x00000fff, 0x00000000, 0xfff00000, 0x00000000, 2}
>>>>depth 1/1/1.0  time 0.00s  value=-2  nodes 11  0kN/s  db 0%  pv 10-15
>>>>depth 3/11/4.9  time 0.00s  value=2  nodes 495  0kN/s  db 0%  pv  9-14 24-19  5-
>>>>9 28-24
>>>>depth 5/13/6.5  time 0.00s  value=2  nodes 2144  0kN/s  db 0%  pv 11-15 22-18
>>>>15x22 25x18  8-11 18-14
>>>>depth 7/24/9.4  time 0.01s  value=0  nodes 10769  717kN/s  db 0%  pv 11-15 22-18
>>>>15x22 25x18  8-11 29-25  4- 8 25-22
>>>>depth 9/26/11.3  time 0.03s  value=2  nodes 47511  1532kN/s  db 0%  pv  9-14
>>>>22-18 11-15 18x11  8x15 25-22 15-19 24x15
>>>>depth 11/29/13.2  time 0.08s  value=0  nodes 175403  2248kN/s  db 0%  pv  9-14
>>>>22-18  5- 9 25-22 12-16 24-19  8-12 28-24
>>>>depth 13/34/15.3  time 0.33s  value=0  nodes 710957  2167kN/s  db 0%  pv  9-14
>>>>22-18  5- 9 24-19 11-15 18x11  8x24 28x19
>>>>depth 15/35/17.4  time 1.26s  value=2  nodes 2732948  2160kN/s  db 0%  pv 11-15
>>>>22-18 15x22 25x18  8-11 29-25  4- 8 18-14
>>>>depth 17/37/19.2  time 4.47s  value=0  nodes 9382548  2099kN/s  db 0%  pv 11-15
>>>>22-18 15x22 25x18  8-11 29-25  4- 8 25-22
>>>>depth 19/41/21.4  time 16.09s  value=2  nodes 32861073  2041kN/s  db 0%  pv
>>>>11-15 22-18 15x22 25x18  8-11 29-25  4- 8 18-14
>>>>depth 21/44/23.4  time 49.06s  value=2  nodes 94540500  1926kN/s  db 80%  pv
>>>>11-15 22-18 15x22 25x18  8-11 29-25  4- 8 18-14
>>>
>>>Does this mean it is probing the endgame tables in a deep search from the
>>>opening position?
>>
>>yes, and this although dann is only using the 4-piece EGTB, which comes as a
>>download with my program. you can compute the 6-piece EGTB on your own machine,
>>and i once computed the 8-piece EGTB - unfortunately, that is much too large to
>>put on the web unless you're at a university like jonathan schaeffer...
>>with the 8pc EGTB you get much more hits of course.
>>
>>cheers
>>  martin
>
>But you already getting 80% hits if i read correctly!! Why don't you just let it
>run overnight and solve the game :)

hi anthony,

the db % hit rate is something completely different: it tells me how many
accesses to the db were already in cache, and how many it had to get from disk.
so it's a measure for the cache efficiency, that's all...

cheers
  martin

>anthony



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.