chess online

chess online
GameKnot related: Computer Analysis Suggestion
« Back to forum
FromMessage
markb56
05-Feb-11, 15:26

Computer Analysis Suggestion
I'd like for there to be a way to either HIDE or DELETE the computer analysis (CA) on a given game. For example, how about providing a profile setting to disable seeing any CA if one doesn't want to be influenced. Hiding any CA would be particularly important when viewing games in the GK games database (and this should be done automatically!).

I'd also like to be able to delete the CA for any game where I was the one who originally requested it. Some reasons for this might be wanting to hide my stupidity. Or if it's a game in which I won, having CA seems like rubbing salt into my opponent's wound.

bhidragon
05-Feb-11, 15:43

Hide My CA Games
I'm not crazy about everyone on GK being able to see my CA games. In fact, I've NOT CA'd a game on GK because I didn't want the world to see the analysis of lines I always play.
markb56
06-Feb-11, 12:51

And unfortunately for you bhidragon, there's nothing you can do to stop other players from CA-ing your games and leaving them that way for all to see.

I don't mind other players looking at my CA games (and if anyone wants to CA one of my games, then knock yourself out). But if I later decide I don't want CA on one of my games then there should be some way for me to delete the CA.
tactical_abyss
06-Feb-11, 15:29

Best way to avoid alot of this CA stuff is simply not to do it at all on GK....ever.You can do your own post analysis of the game using your own private analysis program after the game is over.Infact,its my understanding from past forum posts that the GK CA is not nearly the strongest or best out there anyway,so for a few dollars,you can purchase your own analysis program on any of the online chess stores,and even get a more in depth analysis with graphs and more.This of course,does not prevent someone else doing a CA on your game,but it will minimize the amount of CA's overall,statistically.

If I had my way,i'd have the option to make any or all my games never available for anyone to view!That way,I could play the same killer out of book moves indefinitely with a new player,then place it on lock down again,until one day it is figured out!
This will never be,of course,but I can dream about it,can't I?
baronderkilt
09-Feb-11, 19:55

The objection ...
to other players seeing the CA of past games is a very valid one I think. Bad enought that everyone can simply see the Game itself. To play a fantastic improvement in one game may end its usefulness after that.

(Back in my day of Postal, we really had to DIG to even find some few sample games of all but the most famous or prolific opponents. And to find a half dozen examples was a major plus to me. Seeing their style if naught else. Now it ia given on a platter and one can even gain expectations and error lines of the opponent. Back then one could use an improvement in at least all ongoing games, and start another series of games before it was published. Or often avoid its appearing for years, if ones opponent was not hot to show their loss to it, lol. "Hey world, look at the great new move I lost to~!" yeah, right haha)

But the solution of not showing the CA you order, is not a solution (as TA suggests) since they can simply order a CA of any GK game of interest to them. In that regard, i think it would be nice if GK DENIED a CA of anyone that a player is currently playing at the time. So they must order it before or after ...
ionadowman
10-Feb-11, 03:43

Well...
... Either you have a database, or you don't: that's what it comes down to. If there's going to be a Gk database of games played on GK, then it won't do to be selective. Selecting, by whatever criteria, what goes in it and what does not, defeats any purpose to which a database may be put - and that goes for good as well as ill. So that's the decision that would have to be made. Yes or no to a database altogether.

Actually, I believe there is a selection criterion used: some minimum rating level of the games. It's a fairly arbitrary cutoff point, and no doubt people would like to see it adjusted up or down... But what about a database of master games? Ought there to be one of them? You might find the majority of users would like it to remain. If we have a Masterplay database, then why not a GK database? Again I think most would prefer to retain it than otherwise.

I also think it well to access others' games.

I refer to the dtabases and others' games fairly often, but rarely, if ever, to research what people play in order to gin some advantage thereby. My own attitude is that such advantages as might be obtained are nigh negligible. Others will no doubt take a different view.

How about analysed games? I like to show off something I'm fairly proud of, or maybe an interesting idea or play. Something interesting in someone else's game, that might lead to annotating the whole thing (and in any case I have annotated a few games 'on request').

You know: it is my firm belief that the games of intermediate level players - say 1300-1700 - are sadly neglected in chess literature. Of course such games will contain several inaccuracies and mistakes, but they will contain some very good moves as well. What they will show are ideas more accessible to beginning and intermediate players than master games will be. Occasionally such games are greatly entertaining, full of incident, withal containing a lot of food for thought.

Keep the databases and games histories (except maybe private, unrated games if you like). As for the argument presented by markb56 in respect of his CA games, what is the objection? That his opponent might obtain some advantage thereby - having seen how you play a certain line? But that can be a double-edged weapon at that. Suppose in a given game (against aweaker opponent, say, you select what you consider to be a second-best move, and go on to win. Joe Schmuuck, investigating your past record sees this line, finds a good continuation against it, and gets ambushed by your preferred choice... There ain't all that much advantage in looking up people's history.

tactical_abyss
10-Feb-11, 07:13

ion,
I will agree with you mostly,but not completely.In regards to not gaining much advantage in looking up people's history of games,I feel I have won many games by doing just that!One of the things my opponents overlook many,many times is their OWN weakness at certain middlegame stages of their past games and FAIL to even bother researching their own past losses!And this is even with 2200+players!Take a 2200+player that has played 600 GK games.
I will look way way down his list from perhaps years ago and pick one game that his positional weakness on move # 33 lead to his loss on move #50.Now,that player could easily "correct"his positional weakness if confronted in the same scenario again,but never checks and does not care!So he losses again under similar or exacting conditions.
Now,yes,it can be double edged as you say,BUT I personally keep my game load down to a minimum and DO check all my "small"amt of losses 100% of the time(now at only 6 losses in 137 games),unlike some of the heavier bulk game players that have 600-1000 past games.So the odds of me lossing a 2nd time in the same midgame scenario is next to 0% unless i'm drinking too many margaritas!
sccadams
10-Feb-11, 12:39

Too many margaritas
Is a very big positional and tactical blunder indeed.
ionadowman
10-Feb-11, 13:55

A material disadvantage...
... being several margaritas ahead....
easy19
11-Feb-11, 12:00

Observation..
It has nothing to do with the suggestion.

but i was looking at this game and the computer analyses
game

And as you all can see white is loosing the game. but the computer scores tell us that white is winning..
Do i have it all wrong or is the GK Engine Confused ?
black_cat_hamlet
11-Feb-11, 12:43

Huh?
The engine does say that White is losing for all the game... maybe you were looking at the board
as Black?
easy19
11-Feb-11, 13:10

That should be it wen i jump to the last move the numbers are mirrored i din,t know it was doing that.