Author: Robert Hyatt
Date: 13:36:26 01/16/03
Go up one level in this thread
On January 16, 2003 at 15:12:06, David Rasmussen wrote: >On January 16, 2003 at 14:47:54, Robert Hyatt wrote: > >>On January 16, 2003 at 13:31:39, David Rasmussen wrote: >> >>>I do more or less the same thing as everybody else, when probing EGTB's, judging >>>from the open source programs I've inspected. But I get a severe drop in NPS >>>when the tables are probed heavily. Much more than other programs, from what I >>>can see on ICC. >>> >>>What could be the course of this? >> >>Fast disks. IE I am currently using 15K rpm, ultra-320 scsi drives, with six of >>these 36 gig screamers running in a raid-0 (striped) manner, which is _really_ >>fast. >> >>If you are using an EIDE disk, performance is going to die. Particularly if you >>are >>using 5400 rpm drives. >> > >First of all, I'm not using 5400 rpm drives. Secondly, current IDE disks are not >as bad as you think. Thirdly, all of the program I've compared with during games >on ICC, where running on IDE disks, I would think. Sorry, but maybe you are right, "they are not as bad as I think." They are _worse_. :) Believe me, compared to good SCSI drives, anything IDE simply crawls. I have some of the fastest IDE drives around on machines here and they don't even get into the same ballpark with U320 15K scsi drives. IE average latency=2ms. 320 megabytes/sec burst transfer. Very difficult to even imagine an IDE drive that fast. > >> >>>I have the same order of number of probes as other programs as I can see on ICC. >>>So my limiting of probes is not the problem. I only probe in the first couple of >>>plys and if there have just been a change of material on the board. The harddisk >>>I use is very fast. It is a Western Digital 80GB 7200 RPM with 8MB cache. So >>>that shouldn't be the problem either. I use NTFS, not FAT, I don't know if that >>>makes a difference, or file systems in general. I haven't done any serious >>>testing of this on EXT2, EXT3, ReiserFS etc. But as far as I know, people are >>>able to do this on Windows, so I must be able to do it too. >> >>that 7200RPM drive is double the rotational latency of a 15K drive. The >>bandwidth >>is a small fraction of a hot ultra-320 SCSI drive. >> > >I know it's slower, but very few people are not using disks comparable to mine, >and they don't seem to have a problem. Disks such as yours are not strictly >necesary, I would think. Maybe I'm wrong. Also, I still think you are >underestimating the performance of current IDE drives. I don't believe that >fraction is very small. We are talking a factor of 2-3-4 at _least_. And when you get an IDE drive going fast, it saturates the bus and stops everything else. SCSI doesn't. Try working on a machine with IDE vs a machine with SCSI and fire up a huge I/O copy. You can't notice with SCSI. With IDE you can't use the machine. > >>You might try upping the egtb cache size significantly to avoid some I/O. >> > >To how much, would you say? I tried going from 8MB to 64MB just now, and it >didn't help at all. In fact, it was slower. > >/David If it is slower you have something else wrong. IE you must be paging. Back off the size of the normal hash stuff as making the egtb cache bigger should not slow you down _ever_. unless it begins to cause paging.
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.