[virgosrock]

---

Amazing comedy of errors. Note myGIBBO's pass over 2C - cant bid 2S, cant support C with Ax, cant bid 2NT so passes.
Note the final asinine double by GIBBOpp.
Rule should be "Ignore 50% of hcp in suit where you have 10 hcp and they have a fit"
This is not difficult. Did they not test GIB exhaustively before they sold it? We do regression testing all the time.

vrock

[steve2005]

Gib doubles on hcp not tricks which is the mistake.
You don't say type of game so don't know if Gib ran a simulation. If a simulation said to double need to program new simulation program,

[smerriman]

I think it would be best to keep this forum concentrated on regularly occurring, clear-cut bidding sequence bugs, not a multitude of highly competitive auctions like this. I suspect it's more likely to put off the developers from reading through all the threads to find more fixable bugs than achieve anything else.

Once again, a human (West) has made a bid that is completely inaccurate based on GIB's system. Can you think of many hands where West does have 1- heart, 10+ total points, passed initially, and 4 spades makes?

[steve2005]

smerriman, on 2017-December-01, 17:38, said:

Once again, a human (West) has made a bid that is completely inaccurate based on GIB's system. Can you think of many hands where West does have 1- heart, 10+ total points, passed initially, and 4 spades makes?

Gibs bidding system is ridiculous. West isn't allowed to balance unless they are pretty sure they can make contract. What a ridiculous constraint.
And yes I it is not unexpected that 4♠ makes even with west having 4♠ which Gib doubled without knowing anyways.
I love people who say that had to double they had 20 hcp. Well this is what Gib does.

[smerriman]

No, GIB is not doubling on HCP, it's doubling on a simulation. It doesn't matter whether you think 3♦ should mean something else - you have full knowledge of what GIB will take it to mean when you made it; if you get punished for misleading your partner that's your own fault.

No idea what you mean about West having 4 spades; that's impossible since the opposition have shown 4-4. It's almost impossible to find a West hand matching the bidding that doesn't set 4♠, so the code that ran the simulation was correct. It doesn't need a 'new simulation program'.

[steve2005]

smerriman, on 2017-December-01, 19:15, said:

No idea what you mean about West having 4 spades; that's impossible since the opposition have shown 4-4.

No opps have shown at least 4-4

[smerriman]

steve2005, on 2017-December-01, 19:18, said:

No opps have shown at least 4-4

Exactly.. so West can't have 4 spades like you said, or there would be at least 15 spades in the deck.

[cloa513]

smerriman, on 2017-December-01, 19:15, said:

No, GIB is not doubling on HCP, it's doubling on a simulation. It doesn't matter whether you think 3♦ should mean something else - you have full knowledge of what GIB will take it to mean when you made it; if you get punished for misleading your partner that's your own fault.

No idea what you mean about West having 4 spades; that's impossible since the opposition have shown 4-4. It's almost impossible to find a West hand matching the bidding that doesn't set 4♠, so the code that ran the simulation was correct. It doesn't need a 'new simulation program'.

No way is it doubling a simulation it is dumb rule double.

[smerriman]

Maybe I'm missing something. With partner holding a known singleton or less in hearts, aren't you going to get a heart, a diamond, a heart ruff, and a diamond ruff at the very minimum? Unless there's a 5-1 heart split, and even then partner still has to have several more high cards to make up the 10 points..

[virgosrock]

smerriman, on 2017-December-01, 17:38, said:

I think it would be best to keep this forum concentrated on regularly occurring, clear-cut bidding sequence bugs, not a multitude of highly competitive auctions like this. I suspect it's more likely to put off the developers from reading through all the threads to find more fixable bugs than achieve anything else.

Bad doubles are clear cut regularly occurring bidding sequence bugs. Why didn't partner bid 2D over 2C? therefore partner does not have 10+. There is ZERO Room in GIB system for negative inference. BUG! Is GIB hand defensive or offensive? ZERO knowledge of this BUG! This occurs frequently.

Once again, a human (West) has made a bid that is completely inaccurate based on GIB's system. Can you think of many hands where West does have 1- heart, 10+ total points, passed initially, and 4 spades makes?

Then the Human West should have bid 2D and so on. Pls read above.

No offense intended.

vrock

[smerriman]

So you disagree with the description of 3♦. If that is changed, the double won't happen. The logic that led to the double is not the issue.

[johnu]

If there were generic N/S/E/W names on the bidding diagram, I'm not sure whether I could identify which seats were occupied by GIB.

[virgosrock]

smerriman, on 2017-December-01, 21:51, said:

So you disagree with the description of 3♦. If that is changed, the double won't happen. The logic that led to the double is not the issue.

Yes, maybe that is the issue. Well filtered to get to the nub of the problem. maybe the concept of balancing does not exist in GIBBO.

thanks
vrock

[virgosrock]

johnu, on 2017-December-02, 03:40, said:

If there were generic N/S/E/W names on the bidding diagram, I'm not sure whether I could identify which seats were occupied by GIB.

Looks like sarcasm. If so, Over 1d or 1H opening I stretch to bid 2C because it is preemptive, possibly forces a negative or support double. Yes, you could go for a number. I have, rarely. This time the club suit is good and spades are QJxx. Hopefully that addresses sarcasm, if intended, as it pertains to virgosrock's bidding.

thanks
vrock

[steve2005]

smerriman, on 2017-December-01, 21:51, said:

So you disagree with the description of 3♦. If that is changed, the double won't happen. The logic that led to the double is not the issue.

3♦ is bad meaning. As are most meanings in balancing seat. Who waits that long with good ♦ and 10 pts to bid. But still 4S is in danger of making take all the black tricks and finess black Q's, V original assessment is right doubling with AKQJxx ♥ is just pure gamble

[zhasbeen]

smerriman, on 2017-December-01, 17:38, said:

I think it would be best to keep this forum concentrated on regularly occurring, clear-cut bidding sequence bugs, not a multitude of highly competitive auctions like this. I suspect it's more likely to put off the developers from reading through all the threads to find more fixable bugs than achieve anything else.

Once again, a human (West) has made a bid that is completely inaccurate based on GIB's system. Can you think of many hands where West does have 1- heart, 10+ total points, passed initially, and 4 spades makes?

"I think it would be best to keep this forum concentrated on regularly occurring, clear-cut bidding sequence bugs"

I couldn't agree more, and with a strong emphasis on "regularly occurring". First things first, and then maybe we can worry about some of the others.

I get as frustrated as anyone, but if even even half of major issues could be fixed it would make a huge difference.

[steve2005]

zhasbeen, on 2017-December-02, 09:35, said:

I couldn't agree more, and with a strong emphasis on "regularly occurring". First things first, and then maybe we can worry about some of the others.

Very little is getting fixed from the forum.
I see no reason people can't vent on whatever topic they want.
Besides you don't know before you post if something is fixable.

[virgosrock]

steve2005, on 2017-December-02, 18:52, said:

Very little is getting fixed from the forum.
I see no reason people can't vent on whatever topic they want.
Besides you don't know before you post if something is fixable.

Plus it would be REALLY nice to let this forum know when a bug is fixed and what was the new Rule that came out of the bug.

vrock