Confusing Total Points and HCP

[Bbradley62]

Until this version {30, still the current version}, Gib was somewhat weird about bidding NT slams. In defiance of the conventional thinking ( need fewer points w/a fit ), it chose to need MORE points to bid a suit slam. As of this version, things are more mainstream - it will hunt for 33 HCP to bid a slam , and a little less to bid a suit slam.

How about for bidding games?

[hv=nn=Robot&n=SQ652HDAQ543CQ853&d=e&v=o&b=2&a=P1H(Major%20suit%20opening%20--%205+%20%21H%3B%2011-21%20HCP%3B%2012-22%20total%20points)P1S(One%20over%20one%20--%204+%20%21S%3B%206+%20total%20points)P1N(2-3%20%21C%3B%202-3%20%21D%3B%205%20%21H%3B%202-3%20%21S%3B%2012-14%20HCP)P3N(4+%20%21S%3B%2017-%20HCP%3B%2013+%20total%20points)]360|270[/hv]

[mike777]

This seems yet another example of so called "simulation of game bids"

 

 

An example where with the same cards and auction the robot will choose different bids.

 

 

I like your title of total pts and hcp but the robots seem to disagree with you, and that seems to be the issue.

 

 

Often in these threads humans seem to disagree with the machines or at the very least the overseer of the machines. If computing power/speed doubles every 12-`18 months all of this should be overcome by the machines or Skynet soon, very soon.

[johnu]

You were lucky GIB didn't count 5 points for the void, a point for the 5th diamond, and make a slam try.

[iandayre]

This seems yet another example of so called "simulation of game bids"

 

 

An example where with the same cards and auction the robot will choose different bids.

 

 

I like your title of total pts and hcp but the robots seem to disagree with you, and that seems to be the issue.

 

 

Often in these threads humans seem to disagree with the machines or at the very least the overseer of the machines. If computing power/speed doubles every 12-`18 months all of this should be overcome by the machines or Skynet soon, very soon.

 

Yes, we disagree strongly. That is the main point of this Board, to point out areas where GIB needs improvements. Unfortunately I think your optimism is ill-founded. GIB has been around for years, and the computing power exists to improve it significantly, but BBO will not invest the people and resources necessary to do the job.

[johnu]

Often in these threads humans seem to disagree with the machines or at the very least the overseer of the machines. If computing power/speed doubles every 12-`18 months all of this should be overcome by the machines or Skynet soon, very soon.

 

Computing power is not really the problem. If the algorithms are bad, you just get bad bids and plays sooner rather than later. At this point, the entire bidding library probably needs to be rewritten in a comprehensive way and eliminate much of the adhoc patching that has been going on for years. As a software developer, I have seen too much code that has become incomprehensible over the years where what should be simple patches become major undertakings.