They're very similar. The 30 block net has more parameters. Unsurprisingly, it seems to persistently be about 20 Elo ahead in my internal tests with equal search.

But highly noisy, sometimes it appears behind due to measurement noise in the number of games I'm using in internal tests or because by random luck in fact not as far ahead at that point I sampled. This is more sort of an "on average" thing across many different points in training across the last month.

But it's also slower to evaluate, which cancels out that advantage. So probably on most hardware the 40 block net will be better than the 30 block. For users on weaker hardware, likely even the 20 block net will still be better than either of those - it's trailing by about 120 Elo right now, but still improving due to continued training on the big-nets data, and will get much deeper search than the big nets which will probably more-than-compensate that on weaker hardware and/or for lower numbers of playouts.

Hi,LightVector!
Just one simple question: does "kata" mean anything?
Is it from Japanese or Greek? someone told me it
could be a girl's name!

New neural nets released!
https://github.com/lightvector/KataGo/r ... 1.3.5-nets

And just for fun, a special version for bigger board sizes:
https://github.com/lightvector/KataGo/r ... 3.5%2Bbs29

lightvector it's just a holiday !!! I am very very very grateful to you !!! I almost stopped hoping for the game on big boards in the foreseeable future

Auto tuning for version 29x29 15,20 and 40 blocks does not work. but networks work with renamed files from version 19x19! 30 blocks does not work in any way (GT610)
maybe I’m lucky to choose working values for 30 blocks empirically
I don’t know how and why it works. therefore perhaps this is not worth repeating!

how to make a game 29x29 of two networks to get a sgf file? Is there a way around the 25x25 restriction for sabaki? or some kind of script to play directly using the gtp? can I launch a game from two networks on the command line?

The only way to get around the 25x25 restriction for a particular GUI program or match-playing software is to request that program or match-playing software to implement a way to handle coordinates for larger sizes compatible with the way KataGo extended the GTP protocol for coordinates. The way I had KataGo implement seemed like by far the most natural way to extend the protocol for coordinates, and is the same as a common choice that board editing GUIs or even spreadsheets like Excel/Google Sheets use to label more coordinates - namely, after Z, you get AA, AB, AC, .... but I'd also be open for adding minor hacks to do a different format if necessary to be compatible with something else.

It's slightly unfortunate that the designer(s) of GTP did not look forward to try to be future-proof and include a specification for how to go past 25x25, or choose a coordinate format in the first place that would be more general and doesn't suffer from limited range, such that we now have this problem. But I've also of course written software where a seemingly minor choice had to be made early and it turned out not to be the right one later, but was stuck with it - it's also hard to be future-proof sometimes.

I think KataGo gets to set the standards these days. I've implemented this scheme in q5go and it seems to work. Code is available on github, and I'll make a new binary release one of these days.

bernds excellent! please also add support up to 37x37. this is interesting, see for example https://senseis.xmp.net/?LargeBoards

25x25 KataGo g170 6 block s1.75 (1 visits) - leela 11 (default, about from 50000 to 100000 visits) 1:0

Relative coordinates (https://senseis.xmp.net/?Coordinates) are infinitely scalable. The symmetries of the goban are integral to this coordinate system; it may be perfect for developing go engines. However I guess that current code might have to be completely re-written to take advantage of that. All points on the goban may be written in the form ij, ij*, i*j, i*j*, ji, ji*, j*i, & j*i* (summarised as i^j^ & j^i^). This illustrates how all points described with identical numerals (that is all variants of i^j^& j^i^) are in the same position relative to their corner, taking all symmetries into account (including those across the goban's two main diagonals.)

I'm posting this on the off chance that experts such as lightvector might see something worthwhile here.

it’s logical for GUI developers to add a coordinate system selection menu. at least if there is support more than 25x25.
in MultiGo, support for large boards (up to 52x52) appeared in 2006, we can assume that "passed the test of time". in KataGo the coordinate system is used as in MultiGo http://www.ruijiang.com/multigo/

Hello,
This is my first post here.

I need some help. I have Katago cuda installed and working. And would like to test some of the other NN against Katago.
I would like to know. If there is a better GUI then Sabakai to test with when playing NN vs NN.

With Sabaki I am only able to test one game at a time. And then I must rematch.

Hardware 2950x, and RTX 2080 ti.

Thanks for your help!

Gogui (from Remi Coulom) works relatively well. Not very intuitive, but does the job.

Friday9i you can remove the restriction size 25 in gogui-twogtp?

for matches you can use the GoGui https://github.com/Remi-Coulom/gogui/releases
and q5go https://lifein19x19.com/viewtopic.php?p=256124#p256124

example bat file for GoGui:

q5go: play online - file - computer vs. computer play

Leela 0.11.0 https://www.sjeng.org/dl/Leela0110GTP.zip