BBO Discussion Forums: Carding convention when the ROBOT defends. - BBO Discussion Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

Carding convention when the ROBOT defends. Carding

#1 User is offline   rhm 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 3,092
  • Joined: 2005-June-27

Posted 2011-April-26, 03:28

Does the ROBOT react to carding from a partner in defense and if yes what Carding conventions does it use?

Rainer Herrmann
0

#2 User is offline   barmar 

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 21,585
  • Joined: 2004-August-21
  • Gender:Male

Posted 2011-April-27, 17:27

I think frequently GIB gives count, I'm not sure if it actually reacts to its partner's signals.

I've also seen it give its partner a ruff even when the partner didn't echo.

#3 User is offline   xxhong 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 328
  • Joined: 2010-November-11

Posted 2011-April-28, 13:49

It doesn't give any signals (no matter counting or attitude). Often it plays the highest card that doesn't cost a trick in its limited sample size(which often costs you a trick in a single dummy analysis). Other times, it simply gives the lowest card no matter it encourages or not. It pays no attention to your signal. It does have some opening lead conventions (K from AK in suits, A from AK in NT, 4th best, top from sequences), although it frequently overrides such leading conventions if the simulation says otherwise.

View Postrhm, on 2011-April-26, 03:28, said:

Does the ROBOT react to carding from a partner in defense and if yes what Carding conventions does it use?

Rainer Herrmann

0

#4 User is offline   barmar 

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 21,585
  • Joined: 2004-August-21
  • Gender:Male

Posted 2011-April-30, 00:52

Maybe it's selective memory, but when I'm declaring I think I can usually tell if a suit will break 3-3 or 4-2 by the count signals the bots give.

#5 User is offline   Bbradley62 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,542
  • Joined: 2010-February-01
  • Gender:Male
  • Location:Brooklyn, NY, USA

Posted 2011-April-30, 20:11

View Postbarmar, on 2011-April-30, 00:52, said:

Maybe it's selective memory, but when I'm declaring I think I can usually tell if a suit will break 3-3 or 4-2 by the count signals the bots give.
Agreed.
0

#6 User is offline   Rain 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,592
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Singapore

Posted 2011-May-02, 11:33

GIB definitely gives count. It just doesn't do it religiously. And it seldom lies when giving count too. Relying upon it as declarer is a sound and often profitable strategy.
"More and more these days I find myself pondering how to reconcile my net income with my gross habits."

John Nelson.
0

#7 User is offline   xxhong 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 328
  • Joined: 2010-November-11

Posted 2011-May-02, 12:25

View PostRain, on 2011-May-02, 11:33, said:

GIB definitely gives count. It just doesn't do it religiously. And it seldom lies when giving count too. Relying upon it as declarer is a sound and often profitable strategy.

Rain, I did a lot of testings. Seriously, Gib plays no counting signals. In many situations, when counting signal is very very important to decide how many rounds defenders need to duck, gib plays very randomly. You simply mixed the situations when gib plays high-low with those situations gib prefers to play the highest card it thinks that doesn't cost a trick. About 50% of times, that happens when they have even number of cards, which really doesn't mean they have even number of cards.
Gib does have some regular patterns in defending, including always inserting the lowest of its honor sequence at 3rd seat, always tending to cover an honor at early stage of the play, but it doesn't play any counting or altitude signals, unless bbo has modifies the code and added the counting signals recently, which I don't think is the case.
0

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users