Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: benchmark test for fun (and for Vincent) , that means *1.862 Speedup

Author: Vincent Diepeveen

Date: 05:54:21 09/06/02

Go up one level in this thread


On September 05, 2002 at 11:07:38, Robert Hyatt wrote:

It should affect nps. This is the diffrence between you and
me. You assume too much for data instead of doing tests
correctly.


>On September 05, 2002 at 10:49:13, Vincent Diepeveen wrote:
>
>>On September 05, 2002 at 03:56:20, Vincent Lejeune wrote:
>>
>>No it's not 1.862 speedup at all. Both searches
>>come from hashtable. It's useless compare.
>>
>>In diep it would be 0 seconds to get it all from hashtable.
>
>
>
>He ran the test wrong.  It won't affect the NPS much, however, as
>we are not measuring SMP speedup, but simply raw NPS speedup, to
>address your ridiculous claim about smp vs non-smp crafty.  The
>numbers I have checked so far didn't make this mistake, and they
>are right in line with what I expect, and far better than what you
>claim...
>
>
>
>
>>I have true values in my hashtable, crafty obviously has a bug
>>in implementation of it, otherwise getting to 12 ply wouldn't
>>take 14 seconds in the search 2.
>
>Default hash size?  we are using default hash size?  we are mixing
>apples and oranges?  I'm only looking at raw NPS.
>
>>
>>>On September 05, 2002 at 02:15:58, Volker Pittlik wrote:
>>>
>>>>System: 2*433 MHz Celeron, 128 MB RAM, NT 4.0
>>>>
>>>>Results:
>>>>
>>>>...
>>>>hash table memory = 12M bytes.
>>>>pawn hash table memory = 3M bytes.
>>>>...
>>>>EGTB cache memory = 2M bytes.
>>>>4 piece tablebase files found
>>>>1414kb of RAM used for TB indices and decompression tables
>>>>...
>>>>search time set to 180.00.
>>>>pondering disabled.
>>>>search depth set to 13.
>>>>...
>>>>Crafty v18.15 (2 cpus)
>>>>
>>>>...
>>>>White(1): mt=1
>>>>max threads set to 1
>>>>White(1): setboard 2r2rk1/1bqnbpp1/1p1ppn1p/pP6/N1P1P3/P2B1N1P/1B2QPP1/R2R2K1 b
>>>>- - 0 1
>>>>
>>>>Black(1): go
>>>>              clearing hash tables
>>>>              time surplus   0.00  time limit 3:00 (3:00)
>>>>         nss  depth   time  score   variation (1)
>>>>                8->   0.80  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>                9     0.80  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>                9->   1.64  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               10     1.64  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               10->   4.40  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               11     4.40  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               11->  13.69  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               12    13.69  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               12->  26.58  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               13     1:34  -0.60   1. ... Bxe4 2. Bxe4 Qxc4 3. Qxc4 Rxc4
>>>>                                    4. Nxb6 Rxe4 5. Nxd7 Nxd7 6. Bd4 e5
>>>>                                    7. Be3 Rb8 8. Nd2 Rh4
>>>>               13->   2:04  -0.60   1. ... Bxe4 2. Bxe4 Qxc4 3. Qxc4 Rxc4
>>>>                                    4. Nxb6 Rxe4 5. Nxd7 Nxd7 6. Bd4 e5
>>>>                                    7. Be3 Rb8 8. Nd2 Rh4
>>>>              time=2:04  cpu=99%  mat=0  n=25416120  fh=92%  nps=203k
>>>>              ext-> chk=236848 cap=77200 pp=4763 1rep=11848 mate=1281
>>>>              predicted=0  nodes=25416120  evals=6188352
>>>>              endgame tablebase-> probes done=0  successful=0
>>>>              SMP->  split=0  stop=0  data=0/32  cpu=2:04  elap=2:04
>>>>
>>>>...
>>>>White(2): new
>>>>...
>>>>White(1): mt=2
>>>>max threads set to 2
>>>>White(1): setboard 2r2rk1/1bqnbpp1/1p1ppn1p/pP6/N1P1P3/P2B1N1P/1B2QPP1/R2R2K1 b
>>>>- - 0 1
>>>>
>>>>Black(1): go
>>>>              clearing hash tables
>>>>              time surplus   0.00  time limit 3:00 (3:00)
>>>>         nss  depth   time  score   variation (1)
>>>>starting thread 1
>>>>                8->   0.46  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>                9     0.46  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>                9->   0.91  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               10     0.91  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               10->   2.39  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               11     2.39  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               11->   7.44  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               12     7.46  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               12->  14.46  -0.63   1. ... Bxe4 2. Bxe4 <HT>
>>>>               13    52.14  -0.60   1. ... Bxe4 2. Bxe4 Qxc4 3. Qxc4 Rxc4
>>>>                                    4. Nxb6 Rxe4 5. Nxd7 Nxd7 6. Bd4 e5
>>>>                                    7. Be3 Rb8 8. Nd2 Rh4
>>>>               13->   1:10  -0.60   1. ... Bxe4 2. Bxe4 Qxc4 3. Qxc4 Rxc4
>>>>                                    4. Nxb6 Rxe4 5. Nxd7 Nxd7 6. Bd4 e5
>>>>                                    7. Be3 Rb8 8. Nd2 Rh4
>>>>              time=1:10  cpu=579%  mat=0  n=26743151  fh=92%  nps=378k
>>>>              ext-> chk=252682 cap=81998 pp=5746 1rep=13038 mate=1314
>>>>              predicted=0  nodes=26743151  evals=6550687
>>>>              endgame tablebase-> probes done=0  successful=0
>>>>              SMP->  split=127  stop=8  data=5/32  cpu=6:49  elap=1:10



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.