<H> Good thinking brah- that extra six millimeters could save your life in the field.

How is your connection to cg.com in general? I'm having lag problems here in Korea.

Example right now I timed out twice before getting to this page.

<LAG PROBLEM cg.com timing out very difficult to get on and navigate the site:

< 499 registered members online, plus 1156 anonymous users. >

The foregoing post is from page 23 of the thread of comments to Karjakin vs Caruana, 2016. I did not want to clutter that thread with a discussion on a non-germane topic, but it seemed OK to address it here.

In passing and for complete context, I will note that the first part of the above post refers to an earlier comment relating to the fact that Gates and Buffett have not put any portion of their respective large fortunes towards sponsorship of chess.

With regard to the current state of the art for computer engines that play bridge, I have a few thoughts, although I am knowledgeable about neither computer programming nor bridge strategy beyond an extremely rudimentary level.

My first thought is that programming a computer to play bridge is fundamentally different from writing a program for chess analysis due to the fact that a given player lacks perfect knowledge of two other hands (both opponents’ hands in the case of the declarer, and the hands of the declarer and one’s partner in the case of the two defenders).

Although each player of a hand of bridge lacks perfect knowledge of how cards are distributed, I would have thought that with knowledge of the bidding conventions being utilized by the respective partnerships (and it is my understanding – possibly incorrect - that partnerships must disclose the bidding conventions they are using), the bidding history for a hand and the cards already played as the play of a deal proceeds, it would be possible to write a program that would be able to implement a statistically perfect play of the hand. By statistically perfect, I mean a strategy that has more likelihood of success than any alternative strategy taking into accounts the statistical probability of the various possible distributions of unknown cards. If this is correct, I would say that such a program would have "solved" (or "busted") bridge.

<juan31: chessgames.com: The colors and the function "delete" , returned, the sizes continue giants.>

(Had to refresh the page repeatedly.)

Is there some (or several) fundamental flaw(s) in the foregoing?>

That's totally off-topic and fascinating. I do not fully understand the rules of contract bridge so I am not qualified to opine on that specific game, but I have read about research done with computers and poker, another card game that relies on imperfect knowledge. In fact, computer pioneer John von Neumann tried to tackle the problem of poker and failed, although he came up for a mathematical solution to a simplified poker derivative, in which the betting takes place simultaneously instead of in order. Obviously a ton of progress has been made since then but the last I checked we still haven't "solved" poker, which seems so simple compared to bridge.

Hmmm... that's the problem with these changes. Possibly it's broken on some tablets and works well on others. But for all I know you two are looking at virtually the exact same thing.>

http://www.miamiherald.com/news/nat...

I'm not computer literate but how long does it take to get a proper server up and running? AFAIK you have been working on this for almost a year now?

Not working properly yet, so please don't give up.