25 Nov 08
Originally posted by Jie😀
Conditional moves? I could have sworn User 124624 uses conditional trolling to troll after my posts. These pesky trolls. 😛
25 Nov 08
Originally posted by no1marauderA human player would be inclined to use conditionals primarily in forcing situations: automatic recaptures, forced checking sequences, "book" openings, etc. In complex, non-forcing situations, most players would prefer to double-check calculations each move. Routine use of conditionals in such positions would only arouse suspicion.
This will greatly help engine cheats; all they'll have to do is program in a bunch of possible moves by their opponent and give Fritz/Rybka/Hiarcs, etc. etc. a couple minutes and VOILA!; they'll be able to move even faster than they do now.
Further, machines suffer from horizon effect. Employing conditionals [without letting them recheck their calculations each ply down the road] just makes the effect more pronounced, and the engine's play weaker overall.
It may even be possible for site owners or game mods to compare conditional trees to engine analysis, providing yet another way to match up moves with an engine. So - there's no need to wait until they have played 500 games - just ban them once they have enough games with conditional analysis that matches an engine's.
In summary, conditional moves help the cheat detectors far more than they do the engine users.
25 Nov 08
Originally posted by SwissGambitYou're seriously deluded if you think the "horizon effect" of present engines is any serious problem with setting up conditional moves in advance.
A human player would be inclined to use conditionals primarily in forcing situations: automatic recaptures, forced checking sequences, "book" openings, etc. In complex, non-forcing situations, most players would prefer to double-check calculations each move. Routine use of conditionals in such positions would only arouse suspicion.
Further, machines s ...[text shortened]... summary, conditional moves help the cheat detectors far more than they do the engine users.
The idea that using conditional moves might be indicator of engine use, if actually applied by Game Mods, will seriously chill their use. I personally see no necessity for them if everybody rechecks their calculations every move (which I seriously doubt) and regard your post has self-contradictory and misinformed.
25 Nov 08
Originally posted by sh76I plan on using it for the openings, where I'll program in the best line I've found in a certain opening.
I don't know about anyone here, but I'm not touching that feature. Even with my (relatively) low game load, I like to think and re-think and blunder check every move 10 times before I make it. Chess is complicated enough one move at a time.
D
25 Nov 08
2 edits
Glad they are here, it will speed up obvious moves.
Be Very Very Careful
Remember your opponent can deviate any time he wants from a
conditional sequence.
TD Harding's book on postal miniatures has a section called the
Black Museum and warns strongly against con/moves.
That said I'm sure we going to see some howlers and disputes
in the coming months.
Cheating:
If a cheat sends 10 Fritz 1st choice moves and his opponent accepts them,
he too will have a series of Fritz 1st choice moves thus given him a high match up.
I hope there is some way Mods can detect that a player has accepted
the moves given to him by his opponent. Else some poor guy may
get banned for simply accepting a conditional line.
26 Nov 08
program in an opening trap and watch the points rack up 😉
I do recall some game or other of correspondence chess where the first person sent 1. e4
The second person sent 1...b6, any 2... Bb7
And the first person sent back 1... b6 2. Ba6 Bb7 3. Bxb7
The second person resigned.
26 Nov 08
1 edit
Originally posted by no1marauderI've seen horizon effect too many times in post-mortems with Fritz to discount it. It undoubtedly would make an engine play weaker if it is forced to commit to a move just because it liked it 4 ply or more ago. How much of a problem that is depends on the cheat and what he is trying to gain. If he wants to beat lots of strong players, it is a more serious problem, for him at least.
You're seriously deluded if you think the "horizon effect" of present engines is any serious problem with setting up conditional moves in advance.
The idea that using conditional moves might be indicator of engine use, if actually applied by Game Mods, will seriously chill their use. I personally see no necessity for them if everybody re ...[text shortened]... y move (which I seriously doubt) and regard your post has self-contradictory and misinformed.
The idea that using conditional moves might be indicator of engine use, if actually applied by Game Mods, will seriously chill their use.
I doubt it. In a complex middlegame position, the odds that a human's analysis tree will exactly match an engine analysis tree are very low. A legit human player has nothing to fear.
I personally see no necessity for them if everybody rechecks their calculations every move (which I seriously doubt) and regard your post has self-contradictory and misinformed.
My post was a bit more nuanced than that. I didn't say people check their calculations every move; I said they were likely to check them every move in complex, non-forcing situations.
26 Nov 08
Originally posted by greenpawn34I don't think the opponent is given a choice about conditional moves. It sounds like they are triggered automatically by the server.
Glad they are here, it will speed up obvious moves.
[b] Be Very Very Careful
Remember your opponent can deviate any time he wants from a
conditional sequence.
TD Harding's book on postal miniatures has a section called the
Black Museum and warns strongly against con/moves.
That said I'm sure we going to see some howlers and disputes
in ...[text shortened]... m by his opponent. Else some poor guy may
get banned for simply accepting a conditional line.[/b]