@Onyx_Chess said in #58:
> This is an *extremely* complex issue that rises FAR above, "I hate Magnus for making accusations without a smoking gun," and "I hate Hans for cheating!"
Yes - this is right, I think. I've tried to make the same arguments, but it seems most in the thread are more interested in just banging on about their pet points on one side or the other, without seeing that there's a whole range of shades of grey here.
This really is a very complex issue, and how it unpacks will shape the future of this wonderful game. I think that's why I feel strongly about it, even though in the big picture, how I feel about it has basically no impact on the actual outcome :) which is well beyond my pay grade.
@Eyon-chess said in #1:
> When I saw how he was able to analyze and explain this great OTB game against GM Antipov (2615) in 2022 Spring Chess Classic in Saint Louis Chess Club, it's clear for me now that he certainly has a chess-debth inside that match his OTB-performance. That's big evidence for honest OTB play on GM-level.
>
> So yeah, I changed my mind now. Here is the link:
I'm nowhere near qualified enough to understand the difference between GM and SuperGM analyses.
I'd need to defer to what the SuperGMs have to say about the matter.
I feel like this is the best 'go-to' thread for all things chess related.
Just saying.
How's everyone doing today?
@LegendaryQueen
I was referring to current top engine ubitquitously used "on" large online human chess player population as sparring partner or own chess out-of game analysis tools (but to stay in thread, well, also imposed by engine assisted players, even if only 1, "on" their human opponent).
I am not saying engines will always be liked that. So thanks for the outlook from the links. I will keep them for when i get more curious about that.. (now into human chess more, but not forgetting what i learned already).
Quantum computing if it is what i understand of it, is going to help the neural net functions cost impediment, even with current immutablel "pre"-historic engineXengine ELO defining competitions formats stay the same.
But i would hope some imagination other than mine would also revise the competitives set of goals for computer things used by humans outside of engineXengine tournaments... The old slow clunky things may have needed the current primary goal, of code optimisation for speed, given the human spec only concerned with wins, in clocked symetric settings, but nothing prevent imagination to actually compete on the quality of the leaf evaluation functions in as shallow depths as needed .
like we humans do on any position, not only the quiescent ones with non-small imbalance. and most of the ugly current top engine behavior is about it being blind early, and always resorting to depth.... and more depth.. there are plenty of positions where an experienced human can see very shallow positiional clues on just the current psition or nearby in skeleton shallow caluculations, that the current top engine seems to be blind to, and keeps exploring like a Knight path doing the travelling salesman problem on the whole board: zillions of steps. (some of which ("rare" compositions?) positions don't even need humans to consult EGTB).
That aspect is what top engine fail us learning humans, whether as opponent or as single position evaluation machine (via its "best" move variation evaluations from somewhere beyond 16 plies, like 22 here on lichess or more if we knew the extension status in output).
But they have been like that and nobody noticed until another species of algorithms uncovered shallow blindness in top engine top ELO constant winner until then.. NNue is a leaf evaluation, helper to classical leaf evaluation, on "quiescent only" positions of the legal position maximal set. The admissible (quiescent) positions out of legal set that are being evaluated as leaf evaluation during play by NNue, are those under a certain (unknown to me, blog says small imbalance in the SF meaning, not Sillman, btw).
Talking about A0, and for the span betwee SF8 and SF12, of LC0. That other species. A dwarf versions of that, NNue came to the rescue for the small imbalance leaf positions that non-NNue part of the leaf evaluation fonction, could not see at input depth yet. But even NNue, is trained by non-NNue SF moderate depth single tree seach value score. So, the blindness is still there but input-depth of first SF call (on the human visible position), plus moderate input depth, classical visible reward...
That it may have improve slowly before that patch, via making classes of near mate (or other terminal outcomes), positions characteristics, was not enough compared to the SF11 blindness on small static imbablances (by definition). what is small might be a tweakable parameter now... but NNue is more costly in sequential computation currency (in ELO=speed).
I would welcome any statement that would make the above a incorrect high level model. please without linking out, unless well linked to soure code or its documentation offspring anywhere (official blog included).
but not talking about all engines of the future.... (so thanks for the links, and bringing that aspect. all is not hopeless self-blinding).
@Onyx_Chess said in #58:
>From the bullying to Magnus' reactions...all of it was set in motion by cheaters.
>*Cheating is the problem...so let's all just decide to not do that any more.*
So what you are saying is that you think that because you claim to be a fan of chess, you should ignore the facts on whether a teenager actually cheated at the event, and blindly accuse them if they cheated years ago in a different format as a kid, propagate false rumours with sexual connotations when you can't find evidence, disparage the worlds leading expert when he candidly explains the facts are not in favour of your accusations, encourage others join your vitriolic crusade even if there is video evidence of themselves cheating, and even if you also have been on camera cheating. And now that the instigators display their complete lack of corroboration to your agenda, that you pretend the hatred was not about this teenager you have vilified, but was rather focussed on a moral endeavour, like you were always the rational moderator and cheating was the enemy you were actually targeting?
See? We know all that is not what you think. I am showing you what bullying is (what the people you are covering for are doing). I don't agree with what I have said above, it is an illustration, and I hope you can understand the matter from a broader perspective, and by me putting you as the individual accuser, it may help you understand where their logic is flawed. And I completely accept that I am painting things in a light which doesn't take the full facts into account, but it is only to show how those same bullies continue to cherry pick factoids which protect their egos from accountability for what turns out to be actually a large amount of lies.
So who do you really think should be banned from chess.con? The middle aged adults caught cheating on video, or the teenager who earnestly apologised for cheating 2 years in the past when he was a kid, and the facts now corroborate his apology?
@Nomoreusernames said in #65:
> So who do you really think should be banned from chess.con? The middle aged adults caught cheating on video, or the teenager who earnestly apologised for cheating 2 years in the past when he was a kid, and the facts now corroborate his apology?
Earnest apology would be admitting your wrongdoing and facing the music - that's not exactly what happened, is it? I agree that the way this has unfolded is messy but your insistence on a completely one-sided narrative doesn't actually help your case. Making up your own facts is an approach best left to politicians, not chess players, eh?
Your straw man about cheating would be funny if I didn't think you might actually believe it. There's no comparison between firing up an engine in a tournament online and playing some games with a group of people behind you yelling out comments. One is malice, the other is, perhaps at worst, ill-judged. Neither is ideal, but there's a world of a difference between having an intent to unfairly win sequentially and some streamed stupidity.
I simply don't understand why it all has to be one way here. It simply isn't 1 or 0... lots of grey.
@SomewhatUnsound said in #66:
> Making up your own facts is an approach best left to politicians, not chess players, eh?
" I don't agree with what I have said above, it is an illustration", I explained it and the motivation. Not politician, holding up a mirror.
>Your straw man about cheating would be funny if I didn't think you might actually believe it.
It is not shrouded in statistical misunderstanding, it is video evidence. You may call it not cheating, but the sites rules disagree, and that is what cheating is, not playing by the rules.
>I simply don't understand why it all has to be one way here. It simply isn't 1 or 0... lots of grey.
It isn't one way at all, it is to help people see the bullying. Why are you trying to hide that?
A couple of things @Nomoreusernames. First, you didn't really respond to my points, which is a pity. I asked if you can really compare using an engine with what you continue to complain about w. Magnus.
Re: bullying. I agree Hans is under a lot of pressure, but a lot of it he has brought on himself, too. Is it bullying or is it consequences of his actions and his way of dealing with people? I do not like this is playing out in the court of public opinion, but it's like you're just hitting one note on a piano, over and over.
Real bullying is nasty, and I hope this doesn't resonate with you for the reason of personal experience. If you are being bullied, maybe at school, I would talk to a teacher or a trusted adult. If you're well out of school, you should be able to see the situation is much more complex than you're making it out to be.
>@SomewhatUnsound said in #68:
>A couple of things @Nomoreusernames. First, you didn't really respond to my points, which is a pity. I asked if you can really >compare using an engine with what you continue to complain about w. Magnus.
I have said that Magnus went against the rules of the site, which is cheating. If you meant about specifics, then I would agree that engine cheating has a higher likelihood of being pathological, but not necessarily if you are a kid, and you accept and apologise.
>Re: bullying. I agree Hans is under a lot of pressure, but a lot of it he has brought on himself, too.
He is not under a lot of pressure, don't downplay what they are doing to him, that's denial.
>Is it bullying or is it consequences of his actions and his way of dealing with people? I do not like this is playing out in the court >of public opinion, but it's like you're just hitting one note on a piano, over and over.
How many points do you want? I could say you don't want to look at the truth because it doesn't suit your ego.
>Real bullying is nasty, and I hope this doesn't resonate with you for the reason of personal experience. If you are being bullied, >maybe at school, I would talk to a teacher or a trusted adult. If you're well out of school, you should be able to see the situation >is much more complex than you're making it out to be.
It is complex, so you have to deal with it, don't just let it go over your head! The only way is to stop the bullies. If you want to turn your head and pretend they are acting morally or ethically, then I can't stop you, but I can make sure that they and you are opposed. The bullying is predatory but what makes it worse is when otherwise sensible people turn the other way. You should try be objective, just because Hikaru is fashionable at the moment, Magnus is powerful, Rensch is back by plenty money, does not warrant their flouting of acceptable behaviour within a community, it doesn't mean they are allowed to bully which ever teenager they choose.
@tryan82 said in #3:
> They may not be common, but ANY cheater is a problem that must be dealt with. Chess.com has MILLIONS of members, so 14 in 10,000 is still a LOT of people. And up until this fiasco, it seems like chess.com was only giving them slaps on the wrist. Hopefully that will change now. But, the reality is that there are forms of online cheating that are impossible to catch. If someone has an engine open on their phone while playing on their computer, or an opening book, there's no way you'll ever catch that, and it's an unfortunate reality.
To who every person who believes chess.com does not like cheaters think again.I was a paid member of chess.com once they offered to "refund" points to me when an opponent of mine "won" they gave me the game score.
Because I played less than 50 games it was easy for me to see who.
The person? was not flagged or banned.
I left
It is an unfortunate reality so is lying about what you'll do.
This topic has been archived and can no longer be replied to.