mmt Vs. Ovyron (G4 D5 BG2)

[jp]

But Stockfish? Pfft! I can believe it'd do it if a skilled centaur provided for it a good book (leaving it into a position where it can win by itself) but otherwise you overestimate it. Unassisted Stockfish 13 will destroy Stockfish 11 with some 100 elo advantage, and corr chess controls allow one to find S13's moves already, just like I was finding Stockfish 11's moves on Stockfish 9's times, because the moves are there in the open, the moves can't be hidden from us.

This is the time in discussions when you ramp up your claim into an even more grandiose claim instead of just giving evidence for the original claim. You're now making unsupported claims about an engine version that doesn't yet exist.

[Ovyron]

They'll release Stockfish 12 once they hit the 50 elo improvement mark, and Stockfish 13 once they hit the 50 elo improvement mark (over S12.) The only way I'm being wrong is if improvement halts and we don't get to see those versions, but I guess this is something that would take years to prove.

Save this to your bookmarks.

[Ovyron]

Qf5

1. g4 d5 2. Bg2 Bxg4 3. c4 c6 4. Qb3 e6 5. Qxb7 Nd7 6. Nc3 Ne7 7. cxd5 exd5 8. d4 Rb8 9. Qa6 Rb6 10. Qd3 Ng6 11. h3 Be6 12. Nf3 Bd6 13. h4 h5 14. b3 Nf6 15. Bg5 O-O 16. e3 Re8 17. Kf1 Bg4 18. Ne1 Bb4 19. Na4 Rb8 20. Nc2 Be7 21. f3 Be6 22. Nc5 Bc8 23. Kf2 Nd7 24. Ne6 Qa5 25. Bxe7 Rxe7 26. b4 Qb6 27. Ng5 Ba6 28. Qa3 Rbe8 29. Bf1 Bxf1 30. Raxf1 Qc7 31. Qd3 a5 32. a3 axb4 33. axb4 Qd6 34. Rfg1 Nb6 35. Qf5 Nf8

[d]4rnk1/4rpp1/1npq4/3p1QNp/1P1P3P/4PP2/2N2K2/6RR w - -

[jp]

They'll release Stockfish 12 once they hit the 50 elo improvement mark, and Stockfish 13 once they hit the 50 elo improvement mark (over S12.) The only way I'm being wrong is if improvement halts

But that is not your grandiose claim! Your grandiose claim is that you can predict all of SF13's moves now, even though SF13 doesn't exist yet! (Does depth not matter at all in your claim?) Actually, we should do the test now. Pick a random position or two and get you to predict now SF13's moves. Are you up for it?

[Ovyron]

How is that grandiose? Analyze a position, the move you'll have will have some elo performance, it may be the best possible move on the position (if it is you're done because your move is a Stockfish 13 move or better). If it's not, you continue analyzing, until eventually you'll find a better move (is this the best one? If so, you're done, as it's Stockfish 13's move or better.)

Continue analyzing still until you find a better move...

See how this goes? As the elo of your move increases you'll eventually catch up to Stockfish 13's move (say, Stockfish 13 at 12 hours/move in Uri's machine) and surpass it (if Stockfish 13's move isn't best.)

This all follows from logic.

Heck, you don't even need a skilled centaur for this, if Stockfish 13 is 100 elo stronger than Stockfish 11, all you need to do to find its move, assuming you get 50 elo per doubling of time control, is giving Stockfish 11 96 hours per move, and it'd find Stockfish 13's move, or one better. As Stockfish 11 would beat Stockfish 13's moves under these conditions, its moves are the same or better. All you're doing by upgrading to Stockfish 13 is saving time.

We don't need Stockfish 13 for this, we can assume Stockfish 9 is just released, and I'm talking about Stockfish 11, so the question is if with Stockfish 9 one can find Stockfish 11's move given enough time. Of course because Stockfish 9 at 96 hours/move is going to beat Stockfish 11 at 12 hours/move, so it's playing the same moves, or better.

[Uri Blass]

How is that grandiose? Analyze a position, the move you'll have will have some elo performance, it may be the best possible move on the position (if it is you're done because your move is a Stockfish 13 move or better). If it's not, you continue analyzing, until eventually you'll find a better move (is this the best one? If so, you're done, as it's Stockfish 13's move or better.)

Continue analyzing still until you find a better move...

See how this goes? As the elo of your move increases you'll eventually catch up to Stockfish 13's move (say, Stockfish 13 at 12 hours/move in Uri's machine) and surpass it (if Stockfish 13's move isn't best.)

This all follows from logic.

Heck, you don't even need a skilled centaur for this, if Stockfish 13 is 100 elo stronger than Stockfish 11, all you need to do to find its move, assuming you get 50 elo per doubling of time control, is giving Stockfish 11 96 hours per move, and it'd find Stockfish 13's move, or one better. As Stockfish 11 would beat Stockfish 13's moves under these conditions, its moves are the same or better. All you're doing by upgrading to Stockfish 13 is saving time.

We don't need Stockfish 13 for this, we can assume Stockfish 9 is just released, and I'm talking about Stockfish 11, so the question is if with Stockfish 9 one can find Stockfish 11's move given enough time. Of course because Stockfish 9 at 96 hours/move is going to beat Stockfish 11 at 12 hours/move, so it's playing the same moves, or better.

1)We do not know that stockfish9 at 96 hours/move is going to beat stockfish11 at 12 hours/move
all the elo that we have is not based on this slow time control.

2)Even if stockfish9 at 96 hours/move can beat stockfish11 at 12 hours/move it does not mean that stockfish9 always play better moves.
It is possible that in part of the cases stockfish9 is better and in part of the cases stockfish11 is better.

[jp]

How is that grandiose? Analyze a position, the move you'll have will have some elo performance, it may be the best possible move on the position (if it is you're done...). If it's not, you continue analyzing, until eventually you'll find a better move (is this the best one? If so, you're done, as it's Stockfish 13's move or better.)

Continue analyzing still until you find a better move...

See how this goes?

Again these are wild claims, just like your claim you can magically tag all drawn positions without TBs.

You just assume you can tell if it's "the best possible move", so that "if it's not you continue analyzing", etc.
How would you know if you've found the best possible move?

Even if you were certain it's suboptimal, how do you know you'd be able to find the move or moves that are better.

You make a whole bunch of assumptions. It's not at all logical.

[Ovyron]

How would you know if you've found the best possible move?

Part of being a skilled centaur is knowing when spending more time in the position wouldn't be worth finding a better move (if it existed.) If I didn't know mmt would still be waiting for my reply to his Bg2, yet this point can be achieved in much shorter time, sometimes as fast as 20 seconds (like on Zenmastur's challenge where the best possible move is found already if you don't care about mating faster) sometimes as fast as 10 minutes (the time it was taking me to predict Harvey's moves) and sometimes as slow as 6 days (which seems the minimal time required to achieve perfect chess.)

jp, do you play centaur chess? Most of the answers to these questions would be evident to you if you tried it. Once it turns out you can easily do something you thought was incredible for someone else to do, it's not grandiose anymore. The first time you tag a position as drawn because you know a side can't make progress no matter what moves it plays, regardless of the scores shown by engines, you'll realize "so this is magic tagging."

I know the kind of output Stockfish at 12 hour/move displays, because I've done it, and matched it, and surpassed it, that's why I can talk about what I can do or not. If you went and tried it perhaps you'd find that those moves are nothing special, and that not only my claims are true, but that it's not something worth calling home about. Unassisted engines are dumb.

You make a whole bunch of assumptions. It's not at all logical.

Assumptions based on my experience on the subject. I don't see from where your assumptions that I'm wrong come from.

[jp]

jp, do you play centaur chess? Most of the answers to these questions would be evident to you if you tried it.

Many people on this forum play or have played correspondence chess, but only you claim: i) to have a magic ability to tag all drawn positions, ii) the ability to predict all the moves of engine versions that don't currently exist, iii) that correspondence games "prove" things, etc.

Many of those CC players are openly sceptical of your grandiose claims. e.g. we're waiting for your replies to Uri's questions. So far, your reply to one of his questions was just that you'd (incorrectly) tag the won position as a draw, but it doesn't matter because you would magically have found a simpler win before you reached the position he gave as an example. Again, that is just a wild assumption that there even was a simpler win in a hypothetical game leading to that position. That position could have been part of the only winning line for White.

The first time you tag a position as drawn because you know a side can't make progress no matter what moves it plays, regardless of the scores shown by engines, you'll realize "so this is magic tagging."

You don't know with certainty whether a side can make progress or not. That's why you admitted you'd have tagged Uri's example as a draw. This confession of yours contradicts another claim you made, that if Zenmastur's test position had a single winning line, you'd have found it quickly.

[Ovyron]

I claim that you, jp, could do it too. I'm willing to teach you. But are you willing to learn or are you going to raise your hands in the air and just claim that it's impossible? Where would we start? What's the thing that you think you can't do that makes you think nobody else can and therefore I shouldn't be able to do?

That's be a good start, because if you learned to do something I claim I can do then it'd be easier to believe it can be done.

I've never claimed to be able to do anything extraordinary, there's probably people around that were doing these things already back when I was just starting, and people that do them much faster and efficiently than me. If someone doubts these things can be done they have unhealthy levels of skepticism (healthy skepticism is doubting other people's claims, it's unhealthy when you start from the basis that those people are lying, wrong, or that they can't know those things.)