SF+NNUE reach the ceiling?

[Leo]

...
If you are already convinced, these 4 weeks were all to be expected at the utmost, you'll probably make a self- fulfilling prophecy for your own use of NNUE, but as well as for LC0- like nets and engines, for PUCT and MCTS and A-B-search and all of these things, that have reached such a high elo-performance nowadays.
Let it be developed for game- playing, let the engines play and watch them draw and forget about the rest. As Chrilly Donninger used to say: "Like watching the washing machine doing the laundry".

I does not know the Japanese developers how many years worked on shogi but I am sure it was much more than four weeks. Without them there is no SF+NNUE.
I agree you in the viewpoint of gaming chess driving only for higher and higher Elo is a rather duly thing because it will only enhance the number of draws. But chess is chess: It is a draw-type game. And there is a far aim too:
Solving the chess or at least to reach the "perfect" play.

Solving chess might be impossible for us mortals.

[pohl4711]

You're missing the point. Read "Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess". One of the variants among others evaluated was no-castling chess. And since there is a change to the scoring rules with no-black castling (or no black short-castling) also, it is even more certainly a different variant. It is absolutely not classical chess. It can however replace classical chess and a lot of knowledge accumulated by playing classical chess can be reused.

Here is Kramnik making a distinction between classical chess and no-castling chess for example: "...possibly opening a game with 1. f4,
which is not a great idea in classical chess..." But I'm sure a world champion for 6 years doesn't know what "classical chess" means?

First, you need a patch for GUIs for a new scoring system. Second, you need a patch so that when you click "New Game," you get an option to choose this variant and you need a clear indicator about which type of game you're playing. The GUI should also support new opening books. So you're also wrong that you don't need any changes. The good part is that these changes are relatively minor giving this variant a higher chance of being implemented.

Of course, it would be nice, if GUIs would support Advanced Armageddon scoring. But I am not so naive, to believe, that this will happen. Because of this, I added a batch-tool ("livescoring_advanced.bat") in the NBSC-download, which does an Advanced Armageddon rescoring (using a pgn-gamefile) into a new pgn-file, starts ORDO on that new file and display the results in one step. This tool can be used to see intermediate live-results, if a testrun is still running. Only requirement is, that the GUI writes the played games in a pgn-file. Except FritzGUI, most GUIs do so (Cutechess (-cli), Arena, Shredder, LittleBlitzer). You can put a shortcut to that tool on the desktop and start it anytime, while the testrun is still running.

And, by the way, at the moment I am analyzing more than 165000 different opening lines out of normal human games (Megabase 2020) in order to find enough lines with a measureable advantage for white (without any "no castling" or other modifications). If this succeeds, Advanced Armageddon scoring can be used on these openings, too, which are definitly normal chess, because all lines are taken out of the Megabase without any modification. And, it should be possible to play these lines without Advanced Armageddon rescoring of the resulting games, too. That should give not so good statistics like Advanced Armageddon (of course), but measureable lower draw-rates than any other classical openings-set.

But to find working eval-intervals will need a lot of pre-tests. So, I will need some time, after the analyzing of the 165000 line is finished (end of September...). Hopefully this setup will raise the acceptance of my Advanced Armageddon scoring system.

[Laskos]

You're missing the point. Read "Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess". One of the variants among others evaluated was no-castling chess. And since there is a change to the scoring rules with no-black castling (or no black short-castling) also, it is even more certainly a different variant. It is absolutely not classical chess. It can however replace classical chess and a lot of knowledge accumulated by playing classical chess can be reused.

Here is Kramnik making a distinction between classical chess and no-castling chess for example: "...possibly opening a game with 1. f4,
which is not a great idea in classical chess..." But I'm sure a world champion for 6 years doesn't know what "classical chess" means?

First, you need a patch for GUIs for a new scoring system. Second, you need a patch so that when you click "New Game," you get an option to choose this variant and you need a clear indicator about which type of game you're playing. The GUI should also support new opening books. So you're also wrong that you don't need any changes. The good part is that these changes are relatively minor giving this variant a higher chance of being implemented.

Of course, it would be nice, if GUIs would support Advanced Armageddon scoring. But I am not so naive, to believe, that this will happen. Because of this, I added a batch-tool ("livescoring_advanced.bat") in the NBSC-download, which does an Advanced Armageddon rescoring (using a pgn-gamefile) into a new pgn-file, starts ORDO on that new file and display the results in one step. This tool can be used to see intermediate live-results, if a testrun is still running. Only requirement is, that the GUI writes the played games in a pgn-file. Except FritzGUI, most GUIs do so (Cutechess (-cli), Arena, Shredder, LittleBlitzer). You can put a shortcut to that tool on the desktop and start it anytime, while the testrun is still running.

And, by the way, at the moment I am analyzing more than 165000 different opening lines out of normal human games (Megabase 2020) in order to find enough lines with a measureable advantage for white (without any "no castling" or other modifications). If this succeeds, Advanced Armageddon scoring can be used on these openings, too, which are definitly normal chess, because all lines are taken out of the Megabase without any modification. And, it should be possible to play these lines without Advanced Armageddon rescoring of the resulting games, too. That should give not so good statistics like Advanced Armageddon (of course), but measureable lower draw-rates than any other classical openings-set.

But to find working eval-intervals will need a lot of pre-tests. So, I will need some time, after the analyzing of the 165000 line is finished (end of September...). Hopefully this setup will raise the acceptance of my Advanced Armageddon scoring system.

Well, I am using borderline White Win / Draw unbalanced opening suites for 4-5 years now:
http://talkchess.com/forum3/viewtopic.php?t=61245

I have built quite sensitive unbalanced opening suites like 2moves_80_100 or 3moves_Elo2200_08_10 in case the draw rate is expected to be higher than 60-70%. These openings are too about 50% Win and about 50% Draw and not very much dependent on this on TC. I don't see a reason to use Armageddon scoring, as it is possible to show that this sort of scoring doesn't enhance the resolution of the opening suite. I will post some empirical results, maybe sometime later I will show this thing more theoretically.