Balanced opening lines or advantageous opening lines?

[jefk]

there are unbalanced openings, without significant advantage
for either side. Remember the thread about complexity (of a position),
sharp unbalanced lines, preferably without too much theory (like
the Najdorf poisoned pawn), that's imo the way to go.
And then there are the Pohl openings, many of them quite useful I
suspect (although i haven't looked at it in detail).
If you want to see clear differences in engine strength
(and sometimes interesting games between the a/b and the mcts
(Lc0) engines). And ofcourse playing such lines with Black and White
to ensure equal chances.

[jefk]

more info:
as it's done now:
https://tcec-chess.com/articles/TCEC_Openings_FAQ.html

the Pohl openings:
https://www.sp-cc.de/uho_2024.htm

these UHO openings look good to me, whererby the drawkiller
openings may be a bit too artificial (to resemble 'real' chess)

[Jouni]

Today CCC is number 1 tournament. TCEC has outdated CPU. CCC uses also now Stefan Pohl's UHO_2022_8mvs_+110_+119. UHO is OK, but it don't give real elo difference. But maybe there is no difference between 3800 engines anymore?

[CornfedForever]

Today CCC is number 1 tournament. TCEC has outdated CPU. CCC uses also now Stefan Pohl's UHO_2022_8mvs_+110_+119. UHO is OK, but it don't give real elo difference. But maybe there is no difference between 3800 engines anymore?

"outdated CPU"....CPU should largely be irrelevant. An event should turn on the merits of the engine/software....not how fast you can make them go (if that is what you are getting at by the quote)

[gaard]

If you could back up your choice with reasons, that would be helpful too.

Watching and analyzing, definitely balanced. Testing and rank ordering for development, unbalanced (UHO). For establishing realistic Elos then balanced, for sure. Both are essential.

[Viz]

Today CCC is number 1 tournament. TCEC has outdated CPU. CCC uses also now Stefan Pohl's UHO_2022_8mvs_+110_+119. UHO is OK, but it don't give real elo difference. But maybe there is no difference between 3800 engines anymore?

"outdated CPU"....CPU should largely be irrelevant. An event should turn on the merits of the engine/software....not how fast you can make them go (if that is what you are getting at by the quote)

You do know that leela uses GPUs, right?
You can't just throw in any CPU and any GPU and say "it's okay".
Tcec actually is heavily GPU favourite, cccc has the same GPU setup or almost the same but instead of really old xeons it uses AMD EPYC that are much stronger - and thus more fair since GPU is also pretty new.
The fact that stockfish can win TCEC with such stability while running on relatively weak hardware is one more proof of how strong it actually is.

[Dann Corbit]

I think it depends on the lines and the goals.
I think books that are formed from a collection of draw games are a mistake, because they lead to artificially high draw rates.
I think that positions so lopsided that almost always an engine wins both with black and white is also not useful, even though it adds to the 'drama' of the contest and even though it gives the weaker engines some wins.

I think the most valuable sort of openings would be the most frequently played openings from OTB games and correspondence games from the last five years. There will be drawish lines and firecracker lines. But the collection will be useful because of the pragmatic value and because the engines may show us a weak point in a commonly played line.

I have no interest today in Grob games, even though many Grob lines will not result in 100% wins for one side. The fact is that to get to a position that is not lost, you have to pass through several lost or won positions (depending on side to move) so you will never get there in a real game.

I like the Orangutan, but like most other openings, it is drawish in nature. Other people hate it. Which just goes to show, you can't please everyone.

[gaard]

I think the most valuable sort of openings would be the most frequently played openings from OTB games and correspondence games from the last five years. There will be drawish lines and firecracker lines. But the collection will be useful because of the pragmatic value and because the engines may show us a weak point in a commonly played line.

These lines will be the most drawish. They will also compress ratings considerably. They will also give you the most realistic Elos and yield the best opening statistics to help with practical OTB analysis, IMHO. In summary, I agree with you.

[pohl4711]

I like the idea behind balanced lines, and I think they're actually better for weak engines, unfortunately we are getting to the point where tons and tons of games are draws not just at the highest level, but at the levels below it, where engines such as Stormphrax, Willow, etc. are.

Thats exactly the point. It is not about liking/disliking when choosing UHO-openings or balanced openings. It is about what is working and what is not working.
So, of course, for weak engines, balanced openings are better. Because they work fine here. But for strong engines, balanced openings just do not work anymore. And the number of engines, which are "too strong for balanced openings" is getting bigger and bigger over time. And so it is only a question of time, when all ratinglists, using balanced openings, will "statistically implode".
I find it really strange, that this simple logic is still a point of discussion... Because I can hardly imagine, what anybody can not understand here.

[pohl4711]

the Pohl openings:
https://www.sp-cc.de/uho_2024.htm

these UHO openings look good to me, whererby the drawkiller
openings may be a bit too artificial (to resemble 'real' chess)

That was the main reason to build UHO. Because it is not artificial like my other opening-concepts. There are more than Drawkiller:
https://www.sp-cc.de/anti-draw-openings.htm

In UHO, only lines out of human games from the Megabase are used. Nothing is changed, edited.
And because, I made not only one UHO-opening set, but many, with increasing advantage for white, the draw-ratio in any engine-tournament or test or ratinglist can be adjusted easily to a vlaid range of 45%-60%, by choosing the UHO-eval-range, which works best on your settings:
Eval [+0.85;+0.94] 6mvs: 6696 lines 8mvs: 19303 lines 8mvs_big: 37622 lines
Eval [+0.90;+0.99] 6mvs: 6292 lines 8mvs: 17145 lines 8mvs_big: 33559 lines
Eval [+0.95;+1.04] 6mvs: 5238 lines 8mvs: 14256 lines 8mvs_big: 30372 lines
Eval [+1.00;+1.09] 6mvs: 4740 lines 8mvs: 13227 lines 8mvs_big: 25916 lines
Eval [+1.05;+1.14] 6mvs: 4125 lines 8mvs: 11660 lines 8mvs_big: 22828 lines
Eval [+1.10;+1.19] 6mvs: 3460 lines 8mvs: 9601 lines 8mvs_big: 20600 lines
Eval [+1.15;+1.24] 6mvs: 3161 lines 8mvs: 8940 lines 8mvs_big: 17470 lines
Eval [+1.20;+1.29] 6mvs: 2717 lines 8mvs: 7869 lines 8mvs_big: 15464 lines

https://www.sp-cc.de/uho_2024.htm

And look here, to see, how much better the statisitcs are, when UHO is used, compared to 2 balanced opening-sets:
https://github.com/official-stockfish/S ... -911982470
I think it is fair to say, that UHO blew the balanced openings out of the water...
(3 testruns in Fishtest. 60000 games each, so nobody can doubt these results)

Thats because Vondele (Stockfish maintainer) wrote this later in the linked thread:
"We have quite successfully adopted UHO as the training book. We did find a recent case where there is quite a difference between the Elo measured on the 8moves and the UHO book (#3937) but overall the experience is good. Several major tournaments have adopted UHO-style books as well."

To make this clear: Because Fishtest uses UHO-openings, OpenBench uses UHO-openings, chesscom-tournament use UHO-openings:
At least 99% (more likely 99.9%) of all engine-games, played on this planet, are played, using UHO-openings, right now. (Fishtest alone plays 1-2 billions of games a year using an UHO-book (they build their own huge book from Lichess games, but the concept is still my UHO-concept, of course)). So, today, balanced openings are already dead, considering their amount of using in the worldwide computerchess, compared to UHO...