Re: [gnugo-devel] code maintenance patch

[Gunnar Farneback]

Nando wrote:
> Ack. Then I guess that the function has to stay in owl.c (not a problem I
> think). Revised patch attached.

The long term solution would probably be to move all persistent cache
code to a file of its own. Arend proposed that recently in a
discussion about adding a persistent connection cache.

> Checking "harder" now... nope, apparently not related to other patches :
> 
> Regressing owl.tst with D715 uncorrected (symmetry is '8'), I have :
> 
> ---- File [owl.tst]
> Summary: 231/262 passes. 0 unexpected passes, 0 unexpected failures
> 
> With symmetry corrected (i.e. '-' instead of '8'), I get :
> 
> ---- File [owl.tst]
> 23 FAILED: Correct '0', got '1 S10'
> Summary: 230/262 passes. 0 unexpected passes, 1 unexpected failure
> 
> 
> Odd...

There's one plausible and probably also likely explanation. When
removing the redundant symmetry for these patterns the matching order
of patterns may change which in turn may lead to a different move
ordering. Since only a limited number of moves are tried this could
lead to a different result.

The way to verify this is to run --decide-dragon with and without the
patch and compare the trace output (-t) and/or the generated sgf files
(-o vars.sgf). If the point of divergence involves multiple moves with
the same value, one of which comes from the modified patterns, the
hypothesis can be assumed to be correct.

/Gunnar