The case against using computer chess engines during games on stream.

Sort:
ChessianHorse
I think commentators should not see the eval bar constantly. I remember some SCC matches that went something like this (paraphrasing):

White makes a move, initial big swing in favor of black.
Hess: wow I really like this for black now!
Rensch: Yeah, that was a terrible move.
Eval bar swings back to equal (without a move being made)
Hess: On second thought, this is not so bad for white actually.
Rensch: Yeah I agree.

Commentators can get really biased by the engine. Of course you could see that as a good thing, but I don‘t.

The viewers should have the option to turn on engine evaluation for themselves.
InfiniteFlash
Toire wrote:
InfiniteFlash wrote:

Hi there,

I've been spectating a bunch of chess streams on Chess.com's ChessTV platform, the St.Louis Chess Club's stream and others. One noticeable characteristic I've seen on streams can be described in the image below. 

 

 

Sincerely, 

InfiniteFlash

Good post, some well presented points, the majority with which I agree.

 

Thanks for reading. 

InfiniteFlash
jonathanpiano13 wrote:
I think commentators should not see the eval bar constantly. I remember some SCC matches that went something like this (paraphrasing):

White makes a move, initial big swing in favor of black.
Hess: wow I really like this for black now!
Rensch: Yeah, that was a terrible move.
Eval bar swings back to equal (without a move being made)
Hess: On second thought, this is not so bad for white actually.
Rensch: Yeah I agree.

Commentators can get really biased by the engine. Of course you could see that as a good thing, but I don‘t.

The viewers should have the option to turn on engine evaluation for themselves.

 

Yet another reason why commentators should be careful using the engine to evaluate positions.

InfiniteFlash
madratter7 wrote:

It adds significant tension and interest because then the question is whether the evaluation is correct, or whether the commentators are correct. And it can be either one. For example, say you are in a position (as occurred in the World Championship) where the Commentators realize that it leads to an ending with Bishop in the wrong corner. That can often lead to draws. So are they right that it is a draw, or is the engine right that there is an advantage. That leads to tension and interest, just as when the commentators themselves disagree.

As someone who has written my own chess engine in the past, I think I have a fairly good idea of why they can be quite messed up in there evaluation at times. I will take the opinion of a grandmaster over that of an engine in numerous positions. That said, the engines are clearly more accurate than the grandmaster in most positions, and can beat them handily. In otherwords, I simply disagree with your opinion that for all intents and purposes they are God when playing chess when compared to us. And I think I have an informed opinion in that area. In many positions it is true they will wipe the floor with us. But there are plenty of situations where the grandmaster will realize the engine is evaluating a position incorrectly.

As for argument #4, it doesn't at least for me. If you watched the World Championship there were times where the engine was swinging wildly in evaluation between moves. It would kill suspense if the players were always accurate and the engine was always accurate. Neither is true. So the outcome of the game is in doubt regardless of what the engine says.

I do agree it would be nice if the user could turn the engine off. Unfortunately, given the limitations of twitch as a broadcast platform that isn't possible unless they stream it twice, one with and one without the engine.

 

I understand your point about your first paragraph. It seems that the middleground for using the engine is not never to use nor to always have it on display. It's, as I mentioned in my solution, to use it after substantial analysis prior. That seems to add the appropriate amount of tension.

I do agree with you that the engine simply doesn't understand some chess positions that well.

 

 

Carlsen is on record for saying that engines don't understand the Berlin endgame (above) much, if at all. 

 

 

I'm aware of it evaluating closed positions such as the Classical Variation of the King's Indian (above) where it overrates White's advantage due to White's space advantage. 

However, I think these openings are outliers as you have implicitly said with the statement " That said, the engines are clearly more accurate than the grandmaster in most positions, and can beat them handily. "

It's an overwhelming most. It's not just 60% of positions, its at least 90% of them.


 

"As for argument #4, it doesn't at least for me. If you watched the World Championship there were times where the engine was swinging wildly in evaluation between moves. It would kill suspense if the players were always accurate and the engine was always accurate. Neither is true. So the outcome of the game is in doubt regardless of what the engine says."

I think what I was arguing for in argument #4 was more along the lines that, I like not knowing what direction the game's result is heading towards, which the engine is reliably figuring out for us.

I'm also thinking, now in hindsight, that my argument #1 is intertwined with argument #4. The devaluing of commentary works in tandem with the tension in a position. Somehow, I'm thinking during streams that if the stream's quality goes down due to the engine, the suspense goes down as it takes away from the passion the commentators and I had from the position. I suddenly lose interest in looking at the position if the engine spits out the answer for me. Please let me know if I'm making any sense here or where I'm confusing things. 

 

"I do agree it would be nice if the user could turn the engine off. Unfortunately, given the limitations of twitch as a broadcast platform that isn't possible unless they stream it twice, one with and one without the engine."

 

I think this is common ground we all agree on. At least give us the option of including it or not! cry.png

Nullvhwk

I don't actually want to send this but still I'm sending... I think the engine moves along with the moves they ckme up should be discussed in the stream and the evaluation bar won't be distracting because all the moves will be explained and if people have any other moves to tell then they can post in Live chat. In that way the evaluation bar can also be used for good purposes and people won't rely on the evaluation so much as they do nowadays but there may be people who are not able to think of other moves other than the moves given in the stream so the people need to help them to let their understanding grow and to think by themselves and not explaining all of the moves is the best option. It's my opinion so PLEASE DONT SCOLD ME OR BEAT ME.

 

THE SECKND LAST LINE MAY BE IGNORANT TO MANY PEOPLE MAYBE SO PEOPLE CAN IGNORE ITS their choice (calming down at the last few words)

((Forgot to add that there needs to be an option to turn off or on the evaluation bar during live stream and its true that eval bar is not necessary for the viewers during a long game but for bullet matches its ok and I agree will most points what InfiniteFlash says))

 

This is my first post in my whole life.