Computer Chess Club Archives

Search

Messages

Subject: Re: hash table size - is a power of 2 still an advantage these days?

Author: Robert Hyatt

Date: 13:28:57 09/24/03

Go up one level in this thread

On September 24, 2003 at 13:25:49, Leen Ammeraal wrote:

>On September 24, 2003 at 11:49:58, Robert Hyatt wrote:
>
>>On September 24, 2003 at 11:20:13, William Penn wrote:
>>
>>>For example, is 512MB hash table size really better than 513MB (512+1), or
>>>better than 528MB (512+16)?
>>>
>>>Another example, is 768MB (3x256) better than 784MB (3x256+16)?
>>>
>>>I'm also wondering if these old rules (truisms) regarding optimum numbers for
>>>hash table size only apply to the Windows 9x/Me op systems which had problems
>>>with allocation of resources? I'm doubtful that they still apply to 2000/XP op
>>>systems.
>>>
>>>Does anyone really know?
>>
>>The point is a bit of speed.   You have to convert a hash signature into a
>>hash table index.  For a tablesize that is a power of 2, you can simply
>>AND (mask) off the upper bits leaving a power-of-2 table index.  For other
>>sizes, you will end up doing a divide (mod) to get the remainder.  The divide
>>is not fast.
>>
>>How significant this is is debatable, but for some of us, "every cycle counts."
>>
>>
>I agree that masking is to be preferred to the modulo operation.
>However, what about tournaments (organized by others) that
>allow a hash table size which corresponds to, for example,
>12 MB entries in your table?
>It seems not wise to me to use only 8 MB entries in this case
>just to make the table length a power of 2.

In my case, I do it a bit differently, anyway.  I hash to a bucked of
3 entries, 48 bytes.  this lets me set my hash table size to 3/4 of a
power of 2.  IE hash=12M, 24M, 48M, etc.  That leaves 1/4 of memory
free for the OS and the other pieces of the chess engine, including
pawn hash, egtb cache, etc.

I try to use _most_ of main memory for serious games, and if you have a
1 gig machine, I generally use something like hash=784M, hashp=40M,
cache=128M, and go from there...


>
>Leen Ammeraal

• Re: hash table size - is a power of 2 still an advantage these days? Tord Romstad 06:48:33 09/25/03
  • Re: hash table size - is a power of 2 still an advantage these days? Robert Hyatt 08:28:55 09/25/03
    • Re: hash table size - is a power of 2 still an advantage these days? Francesco Di Tolla 07:42:12 09/26/03
      • Re: hash table size - is a power of 2 still an advantage these days? Robert Hyatt 19:03:45 09/27/03
    • Re: hash table size - is a power of 2 still an advantage these days? Tord Romstad 10:02:22 09/25/03
      • Re: hash table size - is a power of 2 still an advantage these days? Ricardo Gibert 04:01:49 09/26/03
        • Re: hash table size - is a power of 2 still an advantage these days? Tord Romstad 07:15:44 09/26/03
        • Re: hash table size - is a power of 2 still an advantage these days? Gerd Isenberg 06:49:49 09/26/03
          • Re: hash table size - is a power of 2 still an advantage these days? Uri Blass 08:59:10 09/26/03
      • Re: hash table size - is a power of 2 still an advantage these days? Tony Werten 22:03:33 09/25/03
        • Re: hash table size - is a power of 2 still an advantage these days? Uri Blass 01:41:33 09/26/03
  • Re: hash table size - is a power of 2 still an advantage these days? Uri Blass 07:43:56 09/25/03
• Re: hash table size - is a power of 2 still an advantage these days? Pham Hong Nguyen 18:18:02 09/24/03
  • Re: hash table size - is a power of 2 still an advantage these days? Robert Hyatt 08:30:06 09/25/03
    • Re: hash table size - is a power of 2 still an advantage these days? Pham Hong Nguyen 19:12:24 09/25/03
      • Re: hash table size - is a power of 2 still an advantage these days? Robert Hyatt 19:30:03 09/27/03

This page took 0.01 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.