ctak, branch mainAn implementation of Tak and a computer opponent in Chttps://git.l-3.space/ctak/atom?h=main2026-08-28T18:37:41Zfix longstanding negamax bug2026-08-28T18:37:41Ztsliltslil@posteo.de2026-05-05T21:06:02Zurn:sha1:c415ed2a0bbab890fc1fdd2179c8f1405940c15a
we needed to be saving the alpha before the search loop in which we modified it. At this point, though i can't remember, i expect that this is the cause of whatever "instability" the comment was talking about that would have prevented us from using >= in the TT lookup.
switch to returning an array of actions instead of a linked list2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2023-01-29T19:49:04Zurn:sha1:f0139acd1d648dd1db93ee3d3b75546fb7b24c77
- attempts to keep the same move ordering as the list method
- saves ~170msec on a depth 7 search for a given board configuration
- there is room to improve the pre-allocation size estimates, these
bounds are not obviously tight and it may or may not be faster to
have tighter bounds or even some form of estimation
might as well enable size 62026-08-28T18:37:41Ztslil clingmantslil@posteo.de2023-01-21T18:30:45Zurn:sha1:0e81096d5ecb6027814e7aae10b461e774f96407
Same network architecture, same training principle. Predictably this is
too slow.
Also statically allocate state in driver programmes.
Change transposition table lookup policy2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2023-01-19T21:40:24Zurn:sha1:cb2b78ced27fc7996ed11c3450f69138d7d1b61f
Given the approximate nature of the evaluations and truncated tree
searches, the result of negamax will always be sensitive to the
particulars of the depth bounding and the conditions for referring to
precomputed values.
The tradeoff here is a slight performance penalty (~200ms at depth 6
on my old Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz), but a greatly
increased opponent strength as compared to `>=`, and the old
convolutional network (flawed as the implementation was).
switch to explicit game state & important bug fix & clang format2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2023-01-15T20:31:00Zurn:sha1:0223a9bec5535fced1a7698b55fd42155d9b0446
Previously the code base assumed that there was a single, global game
state which was the implicit target of all actions taken. Looking
ahead at architectural improvements, this has now been (almost
entirely) made explicit and functions take tak_state_p where
necessary (and also where unnecessary).
Two important fixes to actions.c were made:
- Previously when generating the possible stack moves, stack height
overflows (> 15) were not taken into account and this resulted in the
tree search corrupting the board state. Now action search does not
list all legal actions, rather the subset of these encodeable by the
implementation.
- The check for crushing on a stack move was incorrect (too strict),
and this resulted in many legitimate moves being igonored.
Finally, in other changes, weights have also been improved by training
all games instead of some subset for chosen players, and clang-format
was run on the codebase.
new neural network arch (faster + better) & minor changes + fixes2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2023-01-15T15:03:37Zurn:sha1:ee216c008a188a9436fedb85c70ee5d1719733b1
Gone is the convolutional neural network, for it turns out not only is
it more difficult to train, but all of the extra information about
board layers didn't make much of a difference at this size.
So cnn1986 has been replaced by nn1986, a standard, two-layer, dense
nn configured as a binary classifier and (mis)used in that capacity.
Note: total number of parameters is unchanged.
HARK: this new nn exposes a bug somewhere in ctak. Run ctlm with
self-play to see the completely borked board state at the end.
--fast-math seems saves some time2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2021-10-22T19:33:06Zurn:sha1:7cf3a656d0c923dd92025c09747461f2f2d1bed0
That i can tell at a glance, the largest relevant change this
introduces is flush-to-zero for floats. Some testing is required to
ensure that the CNN isn't too sensitive to this. Preferable to that
option is to simply train it without relying on subnormal floats in
the first place.
Went for more standard unicode symbols2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2021-10-06T13:46:13Zurn:sha1:d416679e5a8ad01a74345f9efb43472f2898d9b7Corrected win screen in geminict2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2021-10-06T13:26:04Zurn:sha1:bcaa378f2dbbe73607e7201a675683f1ab0eecb6Enforce quoted ptn string for geminict2026-08-28T18:37:41Ztslil clingmantslil@posteo.de2021-10-06T12:50:27Zurn:sha1:3a4c7107d682b4219cc87a696ecfe3a6095512be