Stockfish 070324 a Disaster

[Eduard]

Then they should do this, and create their own club of SF programmers. I posted there because it's a dev version of the engine that I'm not happy with. It's not about official versions like SF 16.1. For me it was therefore logical that I post there. Where else should I post my criticism if it's a dev version? Just because I can't contribute a formula for improvement, I'm not allowed to post there? Its OK. Bye SF developers.

[Eduard]

Here is what the small net was made for:

The L1-128 smallnet is:
- epoch 399 of a single-stage training from scratch
- trained only on positions from filtered data with high material
difference
- defined by abs(simple_eval) > 1000

And what's wrong with that? Basically nothing if it works correctly. But it does not! It's only good for the bullet stats. But Bullet is not the chess world's greatest asset! I don't know how SF could be improved so that, in addition to Bullet, SF also remains good for trustworthy analyses. But do I have to know this to be able to criticize? Is it that you have to keep your mouth shut just because you don't know the solution yourself? This is apparently the case for the SF developers. Good to know.

[Ciekce]

Viewpoints are never wrong. A viewpoint is what you or I or someone else values as important.

I view the sky as green and water as dry

[Eduard]

And for me the topic SF Developers is now settled. I left the SF channel on Discord. I will continue to voice my criticism of the engine when I feel it is necessary. The developers are only one side, the other side is the users. Many of them are also dissatisfied with the development of Stockfish. Most of us don't care about bullet crap. We want to have a nice analysis engine. We long for it. Unfortunately, Stockfish developers no longer offers us this. I now know many correspondence chess players who play at the highest level, and I know that these players want a better engine for their analysis than the current Stockfish dev. That's why there will still be good SF clones that are better for analysis!

[Graham Banks]

A lot of time and effort from many enthusiasts goes into developing Stockfish, so I'd rather thank them than criticise them. Users are very fortunate to be able to enjoy the fruit of their endeavours at no cost.

[Eduard]

I'm not ungrateful. But if I have legitimate criticism, I should still be allowed to say so. Who are the SF developers developing the Stockfish engine for? Actually open source for the whole world, right? Because the project is open source. Who will thank me if I point out weaknesses in the engine so that it can improve? By the way, without users everything is useless. If there are no SF users, the project could be closed. It's always a give and take. Or has that changed now and the developers are now the gods to whom you can only say thank you, always thank you?

[Graham Banks]

I'm not ungrateful. But if I have legitimate criticism, I should still be allowed to say so. Who are the SF developers developing the Stockfish engine for? Actually open source for the whole world, right? Because the project is open source. Who will thank me if I point out weaknesses in the engine so that it can improve? By the way, without users everything is useless. If there are no SF users, the project could be closed. It's always a give and take. Or has that changed now and the developers are now the gods to whom you can only say thank you, always thank you?

Constructive criticism is fine if it's respectful and posted in the right place. It would usually be welcomed I would think, as it gives possible avenues for further improvement.

Yes, both developers and users have a part to play, but without the developers there wouldn't be an engine for us to use.
A bit like no chicken, no egg.

[gordonr]

I'm not ungrateful. But if I have legitimate criticism, I should still be allowed to say so. Who are the SF developers developing the Stockfish engine for? Actually open source for the whole world, right? Because the project is open source. Who will thank me if I point out weaknesses in the engine so that it can improve? By the way, without users everything is useless. If there are no SF users, the project could be closed. It's always a give and take. Or has that changed now and the developers are now the gods to whom you can only say thank you, always thank you?

Legitimate criticism? You talk about wanting an engine for correspondence analysis but yet post an initial test position and complain that it required 48s. Is that really too long for correspondence? Not only that, you fail to highlight that Stockfish will immediately show a winning move for that position (e.g. Bxa7) but you're not happy because it's not the winning move that you want to see. Again, for the purposes of correspondence play, Stockfish is displaying a winning PV from the start. It is not losing a won position.

[Eduard]

There is still one point that I cannot understand. Slow Mover has been removed as an option. The argument was that the user could make mistakes and the engine would play weaker.

If you don't change the parameter, you can't make a mistake.
Anyone who changes the parameter without knowledge is stupid.
If you change the parameter to experiment (which makes sense and can be fun), you won't make a mistake.

There is no good reason to remove this option. It's pure arrogance on the part of the developers and the desire to boss the user.

However, when SF 16.1 was introduced, an option for the small network was created. Before this option did not exist. The danger for an uninformed user is that they install the wrong network in the desire to implement a better network. Wrong because of dimensions 128 and 2560, which can be confused. And if other dimensions of the networks were to come, the chaos would be even greater.

Those who cannot cope with slow mover are less able to cope with networks.

The slow mover option has been removed, but the small network option has been implemented. Bravo!

[Eduard]

I'm not ungrateful. But if I have legitimate criticism, I should still be allowed to say so. Who are the SF developers developing the Stockfish engine for? Actually open source for the whole world, right? Because the project is open source. Who will thank me if I point out weaknesses in the engine so that it can improve? By the way, without users everything is useless. If there are no SF users, the project could be closed. It's always a give and take. Or has that changed now and the developers are now the gods to whom you can only say thank you, always thank you?

Legitimate criticism? You talk about wanting an engine for correspondence analysis but yet post an initial test position and complain that it required 48s. Is that really too long for correspondence? Not only that, you fail to highlight that Stockfish will immediately show a winning move for that position (e.g. Bxa7) but you're not happy because it's not the winning move that you want to see. Again, for the purposes of correspondence play, Stockfish is displaying a winning PV from the start. It is not losing a won position.

There are just two examples from my test set that are known to some users, they are well-known positions. There are more positions and I'm not talking about correspondence chess only. But also about human tournament chess. Both.

I'm telling you seriously: I don't want to use this engine to analyze my own tournament games! Never, be sure. If this were the only Stockfish in the world, I would definitely change provider!