Timeline for How to find 'Thermopylaes' with engines?
Current License: CC BY-SA 4.0
27 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 17, 2022 at 20:23 | comment | added | Mobeus Zoom | If an algorithm doesn't satisfy you and you need code, then ask on StackOverflow. | |
| Jun 16, 2022 at 22:35 | comment | added | Peter Fischer | No it doesn't, and it's a mere comment. If "trivial to write a search...", and " just compare", then please provide; maybe I shouldn't have removed the expectation that I ask to be provided with a step-by-step tutorial of such a search. Also, the dice rolling part is a mistery. Sorry, no kudos. Can you @MobeusZoom provide for a threshold of 2nd best move being more than say 0.75 worse than the thermophylae? | |
| Jun 16, 2022 at 22:05 | comment | added | Mobeus Zoom | @PeterFischer Doesn't koedem's answer satisfy you? Just compare the top move eval to the 2nd best move's eval and set a threshold for what constitutes a 'blunder'. | |
| Jun 16, 2022 at 8:27 | answer | added | David | timeline score: 1 | |
| Jun 15, 2022 at 16:03 | comment | added | Peter Fischer | Thanks! The box helped to stumble upon NAG, where apparently NAG 8 stands for "singular move (no reasonable alternatives)". On that base could find a piece of stockfish code that I'll have to test. Ty @Andrew | |
| Jun 15, 2022 at 15:30 | comment | added | Andrew♦ | If it helps, the standard chess jargon for what you're calling a Thermopylae is "only move". The pgn representation for an only move is a hollow box □ | |
| Jun 15, 2022 at 15:11 | comment | added | Peter Fischer | Removed obv. misleading opening tag. Starting from tabiae, we are not talking about opening lines. - Opponent bringing us to thermopylae state is fine, since we are preparing lines found this way. - No, we are not necessarily talking about trading material for initiative; the topic is broader, see ex. 2, which is even far away from its tabia. See also my last reply. Anyway, I'm looking for a way to have a software spit out lines that feature narrow paths for one side when we steer into them. | |
| Jun 15, 2022 at 14:58 | history | edited | Peter Fischer |
edited tags
|
|
| Jun 15, 2022 at 11:24 | comment | added | ferdy | We are talking about opening lines, that means this line is equal or close to equal, it is difficult to create critical positions from near to equal positions in fact your opponent can also try to bring you to thermopylaes state. The only way to do this is to give up or sacrifice material or position and in return you will get an initiative to bring/test your opponent into critical positions. | |
| Jun 14, 2022 at 17:00 | comment | added | Peter Fischer | You missed the point of there need being equal-ish positions to start from and end up in; see definition. Consider this a preparation issue. No mentioning of giving up material, but may be part of it as long as it remains equalish; no payment, sorry. Apart from that, you got it though, and thanks for the term critical. Will throw on my googling and possibly rephrase. | |
| Jun 14, 2022 at 12:53 | comment | added | ferdy |
On go through tabiae (the ends of opening lines) and extract lines that lead to such thermopylaes so you have an opening line of say 10 plies, you want to find a continuation line such that your opponent will be brought to a series of positions where there is only one playable move (critical position). And in doing so you are willing to give up a pawn or two worth of evaluation or even a whole piece just to bring your opponent into a series of thermopylae states. If the opponent happens to find all the best moves, you will lose the game - the payment.
|
|
| Jun 13, 2022 at 20:07 | history | edited | Peter Fischer | CC BY-SA 4.0 |
edited title
|
| Jun 13, 2022 at 15:11 | comment | added | Federico Poloni | Doesn't basically any tactics quiz provide an answer to this question? | |
| Jun 12, 2022 at 22:40 | comment | added | koedem | You can of course simply solve this with a specific search that looks for this, e.g. it would be trivial to write a search that compares the top move in each PV position with the second best move, then you'd only need some way to put that into objective values. However, I do want to point out a fun different approach: randomness in the evaluations. If you add a random term even just to the final leaf evaluations of a search tree, you inherently incentivize each player to minimize the number of good options for their opponent because that reduces their number of chances for a good dice roll. | |
| Jun 12, 2022 at 7:32 | comment | added | Peter Fischer | Made an edit to the question, added another, much less trivial example, and cut the expectation part to open the question to a wider public. | |
| Jun 12, 2022 at 7:29 | history | edited | Peter Fischer | CC BY-SA 4.0 |
deleted 101 characters in body
|
| Jun 12, 2022 at 7:23 | history | edited | Peter Fischer | CC BY-SA 4.0 |
added 299 characters in body; edited title; added 1 character in body
|
| Jun 12, 2022 at 7:01 | comment | added | Peter Fischer | If this makes things clearer, think about it as another number next to a move, similar to the current eval or the time spent on the move, s.a. "Qd1, 0.00 eval, 0:45 move time, 0 alternatives that aren't blunders". I am looking for strings where one side (lets call him Tal...) can lead the game into and down a line that features many such '0 alternative moves that don't blunder away the game'. | |
| Jun 12, 2022 at 6:52 | comment | added | Peter Fischer | Also, DdogBosses comment seems pretty newbish; there is much more flesh to this bone than simply: shows detoriating position. That's plain wrong, and the part 'if no good move for opponent, then only bad moves' is trivial to the point I didn't even consider a reply, figuring people would be smarter than that. - I admit the example can be improved, and fine, going to reword if reall necessary, but for now claim that people with a more than 'elementary' understanding of chess will come to my rescue. | |
| Jun 12, 2022 at 6:44 | comment | added | Peter Fischer | I am directing this question to people with an outspokenly good understanding of programming, not chess itself. The Tal quote is accurate, think about it a tad more! 'Wide' for a multitude of ways how to handle a position and 'narrow' for only-moves may be unfamiliar terms, yet fully on point; am sorry if this doesn't come across. - Question remains: how to give counts for the number of equalish continuations and so how to extract those tabiae that feature many, even consecutive 'narrow' paths. | |
| Jun 11, 2022 at 23:53 | comment | added | Philip Roe | I downvote also because your example, in comparison with what Tal was talking about, is trivial and makes it quite clear, I regret to say, that you have little idea how chess is played at a level higher than elementary. You seem to believe that you have discovered a new strategic concept but you have not, and your attempt to dress it in intellectual flummery is distasteful. i am being plainspoken only because you did ask for clarification.. | |
| Jun 11, 2022 at 20:40 | comment | added | DdogBoss | @RewanDemontay Oh, I was pointing out a quote by someone else. Maybe it was Nimzowitsch or Lasker. | |
| Jun 11, 2022 at 20:23 | history | edited | Rewan Demontay | CC BY-SA 4.0 |
General improvements
|
| Jun 11, 2022 at 19:42 | comment | added | DdogBoss | What's the quote? A decreasing amount of moves is a sign of a deteriorating position. If your opponent doesn't have good moves left, then they only have bad moves. In the opening, this doesn't happen because both players have a variety of options. | |
| Jun 11, 2022 at 19:08 | comment | added | Peter Fischer | I'd appreciate an explanation to why the down-vote - it could help me improve later questions in general, and this one in particular. | |
| Jun 11, 2022 at 18:37 | history | edited | Peter Fischer | CC BY-SA 4.0 |
added 68 characters in body; edited title; deleted 2 characters in body
|
| Jun 11, 2022 at 18:31 | history | asked | Peter Fischer | CC BY-SA 4.0 |