engine and books on usb stick?

[BBauer]

Thank you for running this position.

Several programs I have tried need several minutes to solve this position. Crafty cannot solve this position in 25 sec. You only show a fail high for the move a4. It looks as my machine (AMD X2 5200+) is much faster.

Code: Select all

smpmt=2
hash=512M
hashp=128M
tbpath=C:\Users\bbauer\64\Tb
egtb
cache=64M
Crafty-21.6  64-bit:
   37	02:16	 136.244.982	1.001.801	+1,59	1. Ke8 Kc7
  37	02:32	 167.240.067	1.100.263	+1,59	1. Ke8 Kc7
  38	03:09	 245.574.634	1.299.336	+1,59	1. Ke8 Kc7
  38	04:00	 340.661.279	1.419.421	+1,59	1. Ke8 Kc7
  39	04:29	 404.731.494	1.504.578	+1,59	1. Ke8 Kc7
 

Same as above, but no table bases

Code: Select all

  34	00:13	  77.052.301	5.927.100	+3,10	1. Ke8!!
  34	00:41	 238.919.025	5.827.293	+2,30	1. Ke8 Kc7
  35	00:49	 286.519.154	5.847.329	+2,70	1. Ke8!!
  35	01:04	 374.830.422	5.856.725	+2,70	1. Ke8 Kc7
  35	01:08	 397.094.160	5.839.620	+2,70	1. Ke8 Kc7
  36	01:43	 596.170.939	5.788.067	+2,80	1. Ke8 Kc7
  36	01:48	 629.298.141	5.826.834	+2,80	1. Ke8 Kc7
  37	02:22	 834.094.116	5.873.902	+2,80	1. Ke8 Kc7
  37	02:38	 923.237.185	5.843.273	+2,80	1. Ke8 Kc7
  38	03:29	1.225.972.118	5.865.895	+2,70	1. Ke8 Kc7
  38	04:39	1.573.651.659	5.640.328	+3,20	1. a4!!
  38	05:55	2.000.871.085	5.636.256	+4,20	1. a4!!
  38	08:27	2.905.441.021	5.720	+2,70	1. a4 Kc7 2. Ke8 b5
  38	08:50	3.041.632.812	5.733	+2,70	1. a4 Kc7 2. Ke8 b5

So I get
1.5 Mnods/s with table bases and
5.7 Mnodes/s without table bases
Note that crafty may show a fail high, but the score goes down!

kind regards
Bernhard

[bob]

Thank you for running this position.

Several programs I have tried need several minutes to solve this position. Crafty cannot solve this position in 25 sec. You only show a fail high for the move a4. It looks as my machine (AMD X2 5200+) is much faster.

Code: Select all

smpmt=2
hash=512M
hashp=128M
tbpath=C:\Users\bbauer\64\Tb
egtb
cache=64M
Crafty-21.6  64-bit:
   37	02:16	 136.244.982	1.001.801	+1,59	1. Ke8 Kc7
  37	02:32	 167.240.067	1.100.263	+1,59	1. Ke8 Kc7
  38	03:09	 245.574.634	1.299.336	+1,59	1. Ke8 Kc7
  38	04:00	 340.661.279	1.419.421	+1,59	1. Ke8 Kc7
  39	04:29	 404.731.494	1.504.578	+1,59	1. Ke8 Kc7
 

Same as above, but no table bases

Code: Select all

  34	00:13	  77.052.301	5.927.100	+3,10	1. Ke8!!
  34	00:41	 238.919.025	5.827.293	+2,30	1. Ke8 Kc7
  35	00:49	 286.519.154	5.847.329	+2,70	1. Ke8!!
  35	01:04	 374.830.422	5.856.725	+2,70	1. Ke8 Kc7
  35	01:08	 397.094.160	5.839.620	+2,70	1. Ke8 Kc7
  36	01:43	 596.170.939	5.788.067	+2,80	1. Ke8 Kc7
  36	01:48	 629.298.141	5.826.834	+2,80	1. Ke8 Kc7
  37	02:22	 834.094.116	5.873.902	+2,80	1. Ke8 Kc7
  37	02:38	 923.237.185	5.843.273	+2,80	1. Ke8 Kc7
  38	03:29	1.225.972.118	5.865.895	+2,70	1. Ke8 Kc7
  38	04:39	1.573.651.659	5.640.328	+3,20	1. a4!!
  38	05:55	2.000.871.085	5.636.256	+4,20	1. a4!!
  38	08:27	2.905.441.021	5.720	+2,70	1. a4 Kc7 2. Ke8 b5
  38	08:50	3.041.632.812	5.733	+2,70	1. a4 Kc7 2. Ke8 b5

So I get
1.5 Mnods/s with table bases and
5.7 Mnodes/s without table bases
Note that crafty may show a fail high, but the score goes down!

kind regards
Bernhard

You didn't mention what kind of disk drives you have. I have 15K U320 scsi drives in raid-0...

Here's a quick test on my box. I copied krpkq.* three times, to create a file that is about 3/4 gigabyte in size. I then rebooted to make sure nothing was cached, and copied it from one raid-0 array to a second one (the second is 10K drives so it is slower. It took just over 4 seconds to do the copy... I then rebooted again, and read every 3-4-5 piece table (cat * > /dev/null) which took just under 23 seconds total...

that's why I say that fast SCSI drives in raid-0 are very tough to beat.