[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] Patch: Caching for owl reading
|
From: |
Gunnar Farneback |
|
Subject: |
Re: [gnugo-devel] Patch: Caching for owl reading |
|
Date: |
Thu, 22 Jan 2004 23:35:21 +0100 |
|
User-agent: |
EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/20.7 (sparc-sun-solaris2.7) (with unibyte mode) |
Inge wrote:
> Ok, here is the final version of new style caching for owl reading.
> It seems to work, although I haven't yet taken the time to run all
> the regressions and time it compared to the old implementation.
I tried it and got these regression results compared to current CVS:
blunder:11 FAIL B3 [D4|E4|E5|F4|C6]
1 FAIL
Total nodes: 1488073421 2810516 10882042 (+0.24% +2.6% +0.12%)
(The node numbers are for tactical reading, owl reading, and
connection reading, in that order.)
The increase in owl nodes is somewhat worrying but maybe it will be
reduced again when code is added to give priority to owl nodes in case
of cache collisions. There are also some gains to be expected when all
caching memory can be allocated to the new cache.
/Gunnar