chess tiger on iphone

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

ozziejoe
Posts: 811
Joined: Wed Mar 08, 2006 10:07 pm

Re: chess tiger on iphone

Post by ozziejoe »

yes, the amateur definitely blunders a few times. I think this app is pretty close to perfect. i wonder what happens if you have training level on and you change to aggressive, very aggressive , and suicidal...I imagine you would see about a 50 elo drop for each variation on the main personality? It is good to practice against the very aggressive player...see them all the time in blitz
User avatar
Steve Maughan
Posts: 1221
Joined: Wed Mar 08, 2006 8:28 pm
Location: Florida, USA

Re: chess tiger on iphone

Post by Steve Maughan »

hi Christophe,
tiger wrote: I think it involved Bxg2 Kxg2 Qb7+ and then Bxc3. Tiger thought you could not recapture on c3.

The problem here is that the line is quite forced, and creating an horizon effect in a forced line results in a too obvious mistake.
How about using a SEE like function to spot these obvious forced lines i.e. Only limit the horizon when the Bxc3 SEE value is > 0

Cheers,

Steve
mhalstern
Posts: 484
Joined: Wed Nov 18, 2009 1:09 am

Re: chess tiger on iphone

Post by mhalstern »

Chris,

This works on the iPad as well, as I have not had any problems with iPhone apps on the iPad.

I have 2 comments.

1 - I just played a 1 minute game and flagged. The program did not call my flag and claim a win. My time showed 0, but I was allowed to continue playing.

2 - If you patch this, or create a native IPad version, please program it so the screen orientation can change to landscape when the user moves the iPad to landscape position.

Thanks
User avatar
tiger
Posts: 819
Joined: Sat Mar 11, 2006 3:15 am
Location: Guadeloupe (french caribbean island)

Re: chess tiger on iphone

Post by tiger »

ozziejoe wrote:yes, the amateur definitely blunders a few times. I think this app is pretty close to perfect. i wonder what happens if you have training level on and you change to aggressive, very aggressive , and suicidal...I imagine you would see about a 50 elo drop for each variation on the main personality? It is good to practice against the very aggressive player...see them all the time in blitz

The elo drop should not be as dramatic as 50 elo points. Maybe 20-30.

In order to illustrate how much 20-30 elo points represents, it is interesting to note that:
- in order to be able to tell apart with 95% confidence two players separated by a 30 elo points difference, you need them to play 375 games.
- if the elo difference is only 20 elo points, 845 games are needed.

So I think that changing the playing style does not affect the strength in a significant way, practically speaking.


// Christophe
User avatar
tiger
Posts: 819
Joined: Sat Mar 11, 2006 3:15 am
Location: Guadeloupe (french caribbean island)

Re: chess tiger on iphone

Post by tiger »

Steve Maughan wrote:hi Christophe,
tiger wrote: I think it involved Bxg2 Kxg2 Qb7+ and then Bxc3. Tiger thought you could not recapture on c3.

The problem here is that the line is quite forced, and creating an horizon effect in a forced line results in a too obvious mistake.
How about using a SEE like function to spot these obvious forced lines i.e. Only limit the horizon when the Bxc3 SEE value is > 0

Cheers,

Steve

Then the program immediately becomes too strong because it will only make a mistake when there is an advantageous capture on another square, or if a pinned piece cannot recapture, and so on. These are uncommon cases. A SEE is actually almost as good as a full quiescence search most of the time. But a QSearch is both faster and more accurate, so I stopped used a SEE at the horizon a long time ago.

It's the main problem. Our good old techniques are so good that we cannot use them as-is to reduce strength.

And our way of thinking is so biased toward competitivity that we already want and know to improve things when we see a mistake. And we already have a whole directory of efficient techniques for each of them.

When you are trained to think competitively, it's hard for you to find lousy solutions!


// Christophe
User avatar
tiger
Posts: 819
Joined: Sat Mar 11, 2006 3:15 am
Location: Guadeloupe (french caribbean island)

Re: chess tiger on iphone

Post by tiger »

mhalstern wrote:Chris,

This works on the iPad as well, as I have not had any problems with iPhone apps on the iPad.

I have 2 comments.

1 - I just played a 1 minute game and flagged. The program did not call my flag and claim a win. My time showed 0, but I was allowed to continue playing.

2 - If you patch this, or create a native IPad version, please program it so the screen orientation can change to landscape when the user moves the iPad to landscape position.

Thanks

My philosophy is that I do not want to create a to stressing playing environment. I do not want to prevent the player from finishing the game.

However it is true that I could at least play a sound to warn the user.

I'm getting a lot of requests for improvements on iPad. It's true that after all an iPad is like an ideal virtual chessboard: it has the right size and you move the pieces with your hands.

The iPad is much better suited to chess than an iPhone or an iPod touch.

So it will get a special treatment in time. Please be patient.


// Christophe