Traxler: Don't even think of playing Ke2!!!

My computer only reaches depth 31 and gives Nxf7 above Bxf7 but does this even matter when we are playing between humans? 

No, my point is only that it's the sort of position that comps seem to misassess - I had quite a detailed  look at the recent-ish  Robert Ris Traxler DVD by Chessbase to see how he tried to make it playable (TLDR - he can't) and it was notable how often the comp would say some line is +3, only to follow its mainline for a few moves and it would show 0.00 and so on.

Good point  it's the horizon effect at it's finest

True, i was just thinking there might be discovered tactics on that f file

To be honest I don't trust engines. They can't even solve Mate in 13s! Just because a chess puzzle starts with a weird (but perfectly legal) starting position doesn't mean it should affect the computer's analysis. Stockfish is the best one, but they should improve the subtle tactics and endgame tablebases.

Actually engines treat such kind of positions very well.

So, don't blame your engine if you don't know how to make good use of it...

Such an interesting opening.  Kids at the local clubs love it.

Lol I have nothing to blame for. When I did a bad use of my engine? I still think that Nxf7 is better, are you gonna tell me that you can play 42 perfect moves in a row to prove me wrong?

You can believe whatever you like, but the truth is that 5.Nxf7? is a bad move both theoretically, AND practically. More than that, your "proof" about the opposite, which you have posted in previous posts, clearly shows that you have no idea about how to use an engine (or whatever you used to post that crap).

Yeah you keep thinking that Bxf7 is best becouse the computer likes it at 42 depth 

You say I don't know how to use an engine (wich it pretty simple) but you using your engine 'well' haven't found a direct refutation of my line 

Reading issues?

I claimed that 5.d4! is best, although 2.Bxf7+ is a close second- both give white the advantage.

Using an engine is not simple at all, no matter what you think.

Your first line (where you initialy claimed that white is winning, and then that it's a draw) is hopelessly lost, and analysis was posted.

The second one is also crap, and refuted by established praxis (mostly correspondence games, since only idiots are willing to play this OTB as white).

Yes it is. Although it's a nightmare to prep against, I had a guy from my club that kept playing it against me in each tournament. It made every game a battle from the very start. It's quite nice because it usually brings a decisive result (we don't play perfectly so somebody blunders it away). Way more fun the having to win an equal pawn ending.

I recall a long exchange I had with Yigor a few years back trying to convince him a certain line we were looking at after Nxf7 was a draw, because it's well known that Nxf7 should be a draw with best play. The engine swore up and down White was winning, of course, because it couldn't see over the horizon. I steered the game into a draw and eventually we came to a consensus.... however I discovered later we were wrong. This was one of the lines where White had already screwed up and Black had 0-0-0 with advantage, which I wasn't looking for because I was looking for a draw, and the engine wasn't looking for because the engine still thought White's material advantage must be winning if Black didn't have a draw or mate, and now the engine could even see the draw, just not the mate which might come many moves past the horizon with best play. These are things that happen when you try to look at this kind of position with an engine. It's much better to begin with games as Pfren is showing. You can't simply let the engine analyse on its own, or even intervene with more sensical moves at important junctures as I was doing to get to the draw in that case without the possibility of missing important things.

There is software which takes care of some of the engine's limitations- e.g. IDEA, part of the Aquarium software. A lot of correspondence players and analysts use IDEA, and I did the same for a while, but I was not satisfied: The procedure was very slow, and a few times important lines were pruned. So currently, I am doing branching and variation picking by hand, and so far I can't complain from my results in correspondence chess.

In any case, using an engine properly is not at all simple, but some clueless guys think that they have mastered the universe by acquiring their new Core i9...

Thanks; I didn't know about that software.

In my initial line white is wining but poucin showed me a move that the engine was missing and then it's a draw, the second line is better for white. I don't think you realize that if a GM lost using some line it doesn't mean it's losing. You are talking about perfect play so showing a match means nothing.

Blindness? Or drunkedness? Maybe both?

Your first line (9.c3?) loses by force (#49), while the second one (9.d6 Qxd6 10.c3?) gives Black a large, probably decisive advantage (#75).

I'd like to see you making mate in 49  you are proving nothing with this but ok

It's funny because if everyone had really digested post #16 on page 1 from May this thread could have been shorter. Or knowing this forum it probably wouldn't have been...