On using chess engines in CC

vzografos

Only Chess

12 Sep '11 22:34

LittleDonkey
Little Donkey
Joined
31 Mar '07
Moves
130114
14 Sep '11 20:53
Scenario Two: Author One is going to write a book. He has a dictionary and a notebook and access to a whole stack of information.
Author Two is going to write a book. He too has a dictionary and a notebook and access to a whole stack of information.

Probably neither will write a best seller but they have the same tools at their disposal.
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:00
Originally posted by LittleDonkey
Scenario Two: Author One is going to write a book. He has a dictionary and a notebook and access to a whole stack of information.
Author Two is going to write a book. He too has a dictionary and a notebook and access to a whole stack of information.

Probably neither will write a best seller but they have the same tools at their disposal.
One has to be a bit careful with analogies because they tend to oversimplify the problem at hand leading to logical fallacies.

Due to the fact that you are equating writing a book (a creative process) with openings (an "if-then" rule following process) I think a more correct analogy of the second scenario (which by the way I believe gives an unfair advantage to the opening engine player) is this:

Both authors have to do a literal translation of a chinese text
Author one has a 100 page dictionary
Author two has a 1000 page dictionary.
If the process requires negligible thought process, who will do a better translation?
tharkesh
Joined
04 Sep '10
Moves
5716
14 Sep '11 21:001 edit
since you try to do an academic exercise here, don't mess up things diverting from the obvious.

To put it into words you seem to understand:

Offline:
Check the score with the help of an engine

Online:
Commit a move

That's obviously cheating
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:021 edit
Did you just paste around what I said above?

You are not an accountant or something are you?

Why is it not possible for a second to think outside of the RHP box you seem to be stuck in?
thaughbaer
Duckfinder General
223b Baker Street
Joined
25 Apr '06
Moves
33101
14 Sep '11 21:051 edit
Originally posted by vzografos
hmmm....I am not really convienced that it is so easy to draw conclusions on how an engine score vs an RNG would affect the player.

For example, a false score indicating the existence of a good move might cause the player to examine the position more thoroughly, pay more attention and generally have a better game. The only thing that the player (thinks wish I could actually test this and see how an RNG actually affects a player. Food for thought.
You're only looking at one possible outcome and applying it to the whole game. The RNG is also telling the player play the obvious and move on when more thought would be better. And it's all random. Quick/slow/quick/slow/quick/quick/slow.. the player can't ignore the RNG. You'll produce an effect but it won't be better chess.
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:071 edit
You're only looking at one possible outcome and applying it to the whole game. The RNG is also telling the player play the obvious and move on when more thought would be better. And it's all random. Quick/slow/quick/slow/quick/quick/slow.. the player can't ignore the RNG. You'll produce an effect but it won't be better chess.

Can you expand a little bit your thought? I didn't quite understand your point.
LittleDonkey
Little Donkey
Joined
31 Mar '07
Moves
130114
14 Sep '11 21:08
If you want to make a golddigging analogy think of the following:

Both players have a spade
One of them has a device that beeps if there is gold within a 10km radius.[/b]

A 10km radius is an arbitrary distance. Why 10km and not 1 metre? You are setting a large area to re-inforce your point. By setting such a large area you are saying there is no advantage when clearly there is.
JS357
Joined
29 Dec '08
Moves
6788
14 Sep '11 21:101 edit
Originally posted by thaughbaer
You're only looking at one possible outcome and applying it to the whole game. The RNG is also telling the player play the obvious and move on when more thought would be better. And it's all random. Quick/slow/quick/slow.. the player can't ignore the RNG.
In one of your 5 active games I put you at a -3.

Seriously.*

Reaction?

🙂

look down below...

*kidding!
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:11
Originally posted by LittleDonkey
True. I chose an arbitrary large distance to illustrate that a score from the engine (as a 10km beep from the device) does not give you a direction where to look at. More precicsely, it does not give you some idea of the next few moves.

And just like gold digging you need to search for it. And searching can be the same as searching over 3-4plys or more (which for humans can be quite hard).

So yes. You are right. The distance is arbitrary and actually irrelevant. It could be 10 or 100 km. As vast as the possible moves in chess.
tharkesh
Joined
04 Sep '10
Moves
5716
14 Sep '11 21:15
well then, think outside the rhp box. i told you earlier, that on other sites this is ok and really allowed.

here it is cheating. very simple.

(but certainly you are clever enough to understand the tos yourself, you are just making 'academic thought experiments' here, after all)
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:21
Originally posted by JS357
In one of your 5 active games I put you at a -3.

Seriously.*

Reaction?

🙂

look down below...

*kidding!
hehehe.

Wouldn't be surprised actually. I suck. I am really very good at getting any position from +10 to -5. 🙂

But interesting though. How would I react. Ok lets put it into perspective.
Assuming that my position evaluation skills are relatively ok then:

-If I evaluated a position to be +X say, and the engine gave me the same, then I would continue playing as I do (no extra effort to find stronger moves). Perhaps I would become a bit lazy and complacent.

-If I evaluated the position to be 0 or -X and the engine gave me +X I would attempt to find the strong move. With, of course, no guarantee that I would actually find it. Still I would try

-If I evaluated the position to be 0 or +X and the engine gave me -X first I would panic 😀, well I would be concerned because there is little you can do unless your opponent blunders or misses the strong move.

I mean from the above (ok very strict rules but for arguments sake), I dont see any clear advantage from knowing the score. Perhaps it can lead you to pay more attention at a certain position but that's really not the same as saying that you will always find a good, strong move.

Of course, the other argument is that using an engine like that may be considered (somehow) as a non-specific aid in positional analysis, and then perhaps is an unfair advantage.

I am just not convienced that the player can always convert this aid into an advantage. Perhaps it also depends on the level of the player.
thaughbaer
Duckfinder General
223b Baker Street
Joined
25 Apr '06
Moves
33101
14 Sep '11 21:22
Originally posted by vzografos
Can you expand a little bit your thought? I didn't quite understand your point.
You're saying the RNG is feeding the player some anaylsis of the "opportunity" in the position at random. The information is only of be benefit if it's accurate. If the "opportunity" throughout an entire game were truly random you'd break even. You can't just pick the cases where it helped. I don't think it's truly random though.
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:23
Originally posted by tharkesh
well then, think outside the rhp box. i told you earlier, that on other sites this is ok and really allowed.

here it is cheating. very simple.

(but certainly you are clever enough to understand the tos yourself, you are just making 'academic thought experiments' here, after all)
Well anyway. It was not oriented towards RHP or any site whatsoever.

It was more to start a discussion about these two scenarios that have been on my mind. Namely if and how one can take advantage of an engine positional analysis (in terms of a single number) and a dense opening book.

Trust me, If I wanted to cheat I wouldn't be bothering will all this.
greenpawn34
e4
Joined
06 May '08
Moves
42492
14 Sep '11 21:25
I think I have it.

A player sees his postion evaluated by a box and plays a move based
on that evaluation.

I'm thinking depending on the playing strength of the player they will see
that they are winning/losing and play accordingly.

The random number aspect will have the player distrusting the number
and play the board.

Sorry to bring cheating back into it:

You cannot leave an OTB game and ask someone if they think you are better
or worse. Even without any analyse seeking an answer would be cheating.

Getting told you are better or worse (play for a win, play for the draw) is
also cheating. It's what the yoghurt dispute was all about in the 1978 Karpov
- Korchnoi match.

Just wanted to clear that bit up.

But I think I see where you are going.

Is it two players play OTB with a dial next to their clock showing what a
mainfame thinks of each players position.

Both players can see each other's dial so one player knows for sure he is better
and the other knows he is worse. (though any player +1600 should realise this too.)

Interesting. Actually it close to what goes on in the minds of good players.

What you need are test dummies to play a game and write '+' '=' or '-'
after each move depending on how they see their postion.

Then compare the results with a box.

Using different classes of player it may produce something we already know.
Good players evaluate chess positions better than weaker players.
vzografos
Joined
02 Aug '11
Moves
2648
14 Sep '11 21:34
Originally posted by thaughbaer
You're saying the RNG is feeding the player some anaylsis of the "opportunity" in the position at random. The information is only of be benefit if it's accurate. If the "opportunity" throughout an entire game were truly random you'd break even. You can't just pick the cases where it helped. I don't think it's truly random though.
I think I understand what you are saying. Or at least I hope so 🙂

But then you assume that the player can always convert any opportunity to an advantage. It would be like solving a chess excercise at every move without knowing if you got ir right or wrong.

I need to think about this for a second.
Namely, if the "opportunity" signaling is wrong or random or biased (and of course the user does not have too many samples to determine its an RNG) would that cause you to miss all oportunities (hence breaking even), try harder to find oportunities and exploit them, or have no effect whatsoever.

Ok, some opinions I have read claim that an engine that gives the score will always give some advantage.
My RNG example was trying to illustrate the opposite. If the engine gives you and advantage then a RNG should give you a disadvantage. Right?
Does the opposite also hold?
If the RNG has no effect whatsoever what does that say for the engine?

Only Chess Forum

On using chess engines in CC