Article View: gnu.chess
Article #4887New TTable aging
From: Roger Uzun
Date: Sun, 05 Jun 1994 08:34
Date: Sun, 05 Jun 1994 08:34
24 lines
1439 bytes
1439 bytes
In a recent email document, from Dan Oetting, it was suggested
that new TTdepthage[n] values should be calculated using
max(0,TTDepthage[0] - ({nodes}/TTagerate));
Where {nodes} is the avg number of nodes entered in the table
for depth == n.
Should we now keep an array of #nodes for each depth,
and inside PutInTTable should we now increment the count
of each time we add a node? Zero out the node counts
for all depths when ZeroTTable(0) is called, then
increment the count where we add items to the ttable,
something inside PutInTTable like:
ptbl->depth = (UCHAR)depth;
ptbl->score = score
ptbl->mv = mv;
NodeCount[depth]++; // this is added to keep count of #nodes in this depth
Does this look like an efficient/correct implementation of the
new aging code?
-Roger
--------------------------------------------------------------
bix: ruzun
NET: uzun@crash.cts.com
Technical Details
Message-ID:
<Cqx15I.F7K@crash.cts.com>Path:
arc1-ca.pugleaf.net!bga.com!news.sprintlink.net!crash!uzunShow Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 10d71d,777c141bb3f2b041,start X-Google-Attributes: gid10d71d,public X-Google-ArrivalTime: 1994-06-05 03:25:17 PST Newsgroups: gnu.chess Path: arc1-ca.pugleaf.net!bga.com!news.sprintlink.net!crash!uzun From: uzun@crash.cts.com (Roger Uzun) Subject: New TTable aging Organization: CTS Network Services (CTSNET/crash), San Diego, CA Date: Sun, 5 Jun 1994 08:34:29 GMT Message-ID: <Cqx15I.F7K@crash.cts.com> Sender: news@crash.cts.com (news subsystem) Nntp-Posting-Host: crash.cts.com Lines: 24