-
Posts
1,317 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Everything posted by georgi
-
Thanks for reporting. Fixed in v30. ( It won't deny Aces if they are not skipped ).
-
To use book bid 5NT GIB needs 25+TP ( used to be 28+TP) based on new ranges for 6,7 levels in suit/NT in next v30. NT ranges are increased a bit, suit levels are decreased a bit. However, with having 22TP GIB cannot use it directly. 6M looks sufficient and it picks the call. After simulation it bids 5NT in 99% of cases. Probably it was basic GIB, but anyway it will be addressed for improvement.
-
Thanks for reporting. It's bug due 5th major. Initially GIB used to open only with 4 cards and now since it was allowed to open with 5th balanced major it bid 3NT instead. Fixed in next version. 4M is having priority , then comes 3NT.
-
These sequences were "unlocked" in v30. Now shows just 10+TP and it's really LR+ instead of LR as now. That change captures UNNT & Michaels overcall where opponent shows two or at least one known suit. Similar to 1♥ - 1♠ - 2♠ -> Cue: limit raise or better -- 3+ ♥; 11+ total points That will help more loaded hands to go through LR+ instead jump to game with unknown strength of 12-16HCP and prevent partner from investigating ( including responder as well because opener makes attitude call also to LR+ range).
-
Thanks for reporting. Old bug. It's fixed in next v30. Problem was that in some sequences like here, GIB was passing due issue with Redouble pass, not "realizing" it's not our suit. Including sequence upper, another sequence on lower level as example: 1♦ - 1♥ - 1♠ - P 2♥ - X - P - P XX - P - P?! now Gib will look at the card and will return/bid best suit we showed fit in. Will pass redouble, but only if that's our suit. Well, that's mainly bug for Basic GIB as its not able to simulate over a problematic call. Advanced GIB is trying to find a way to escape when get tough.
-
Currently for GIB Double by opener is always SuppX once responder could rebid his suit on level 2. That includes also 1♣ - p - 1♦ - 2♣ X ( case where it's some cuebid like 1♥ - p - 1♠ - 2♥ is not SuppX as 1♥ shows 100% 5+ cards ) Of course, it's specific case when 4th seat overcall matches opening (rather minor overcall). It will be discussed but also should be taken in account when you have: 4M333 shape while opening 1♣ where 4th is not partner's suit. So what would you like to do here? Opening 1♦ is kind of easier as you are no longer 4M333, but could have 4♦333 or 4♦4♣ 3rd in partner's major and 2nd in OM. And once you pass such overcall with some 4333 or 4432 trying to avoid misinterpretation by partner you will break the sequence and won't be easy to describe hand. If you find partner with 5332 and 7-8HCP, could he reopen with double or to reopen with 2M? However, it seems that by opening 1minor it's more likely to have support 3 cards than 5+ and opponent overcalling in same suit and overriding SupX meaning to be profitable.
-
It seems to be v21(windows version 5.2.21 rented GIB at tables) as it looks now: 4♥ -> Cue bid -- 1+ ♦; 5+ ♠; 21- HCP; 7- 8421 HCP in ♦; Ace in ♥; 18-22 total points
-
Secondary support after no negative double
georgi replied to Bbradley62's topic in GIB Robot Discussion
Supposedly shouldn't be done with 3+, but what else North could do instead? South's hand still could be 4423. Would be addressed for review about what 2♠ should show. Then it might help North's bid to be more accurate. -
Thanks for reporting. It will be investigated.
-
Don't understand Gib's rebid sequence with nmf
georgi replied to lycier's topic in GIB Robot Discussion
It has been fixed previously for 1m - 1M 1N - 2om Fixed in next version and for 1♥ - 1♠. So - with INV values and 5+♠ GIB would use NMF - with GF values and 4+♠ GIB would use NMF - with INV but bal 4th♠, GIB would invite directly ( 11-12HCP ) otherwise ( like here in example ) GIB would pass. -
Why does GIB consistently not have what it promises?
georgi replied to broze's topic in GIB Robot Discussion
Thanks for reporting. It's fixed in next version. Classic(book) rebid is pass, based on 3NT needs 5+HCP. But after simulation all it could find is 3NT pretending it has 1HCP more. But it won't go over 3NT developing 4+HCP. -
Thanks for reporting. Fixed in v30. It proceeded after simulation due bug created in v29. It will not proceed with basic robots. Bug applies only for this sequence though. ( 5♥ - 5♠ signoff with spades ).
-
Thanks. Fixed in next version. Not to use NMF with not sufficient amount.
-
direct free bid, new suit, is 11+TP. ( and forcing ) Here GIB has 10TP, so it goes through double.
-
GIB explanations and robot tourney card distributions
georgi replied to Rita50's topic in GIB Robot Discussion
And by the way, this tourney was played in 2012-09-23. Long long time ago and lots of versions of GIB were deployed since then. -
GIB explanations and robot tourney card distributions
georgi replied to Rita50's topic in GIB Robot Discussion
The explanation is very likely badly fetched ( probably due network issue at the moment). It doesn't represent what GIB "thinks" it shows. Explanation you see is usually for client. GIB pokes in explanation DB so it gets the normal explanation. All calls from North have the meaning: 1♥ - Major suit opening -- 5+ ♥; 11-21 HCP; 12-22 total points 2♥ - Opener rebids suit -- 5+ ♥; 11-21 HCP; 12-22 total points; forcing to 3N 3♥ - 11-21 HCP; twice rebiddable ♥; 12-22 total points; forcing to 3N 4♦ - 2+ ♦; 11-21 HCP; twice rebiddable ♥; 12-22 total points 5♦ - -- 3+ ♦; 11+ HCP; twice rebiddable ♥; 12-16 total points About finesses, there is no conspiracy in neither tourney as hands are randomly dealt. As more you play as more stats you will get. In one tourney with 8 boards could get 6 contracts in spades, in another could get 3 slams in row. GIBs don't pass cards under the table when declarer plays a card :) -
Yes. 2♠ over all calls by South. 1NT shows 12-14 and 2♥/X shows just 4/3 cards fit. So South doesn't show strength at this moment. Over pass it might be reasonable to bid pass, but it's very specific case, and 4 trumps are hard to express later especially if opps reach level 3 when next round to give a call comes.
-
Yes, fix covers this kind of raises as well. West would bid just 2♠ now. Raising to 3 would need also more unbalanced hand than 4333.
-
2NT -> Invitational to 3NT game -- 3+ ♣; 18-20 HCP; 22- total points; likely stop in ♠ 3NT is currently strange saying exactly 21HCP ( would be reviewed ). But anyway 2♣ is 6-10TP, which could be 3-4HCP with 6th♣. Any rebid by South except pass shows 15+TP as 2♣ is NF, so it's rather invite to some game. So 2NT looks that if partner has 5+HCP to accept.
-
This is not a bug, but deliberate action by North. There is a scenario when GIB makes fake cue with K, pretending it's Ace if it has a void and values for slam. Something like GIB's 'last train'. Aceless, but with sufficient points + void. GIB doesn't show voids as cue except using Splinter or 4NT answer 5NT/5X showing even/odd + void. With other values it would bid Splinter, but too strong now as splinter is limited 13-16TP. It will be addressed for improvement. But currently it's as per design. South could discover this using 4NT instead 4♦. If GIB had a void in diamonds or hearts ( for sure void is not in clubs as it would pretend it has other ace ), as the advantage here is trump is the superior spade, it will be shown. GIB would reply 6♥ and all would be clear. 4♦ doesn't change probable further action as if North replies 4♥ South would bid 4NT anyway. And 5♠ denies ♥AK as 5♥ is below 5♠ call and it was skipped by South.
-
Thanks for reporting. 2♣ followons are main point for next version, but meanwhile this is fixed. 5NT crossed cases where it was used as GF as something like-> Grand slam force (D trump) -- 2+ D; 31+ total points So here as fit is not established 5NT would be treated as just balanced huge HCPs hand ( same explanation as described above ), but GIB will not bid automatically 6♦. Maybe also 5NT needs limit as practically it's 31-32. With 33 6NT would be bid directly, although very rare hand but still.
-
Thanks for reporting. Both cases fixed. First fix includes also explanations adjustment depend on Pass,2NT,3♥,3♠( which is rather invite showing not minimum opening ), 4♥/4♠ and limited lengths in unmentioned major. 2NT is no longer invite just sign-off contract. 3NT shows just in case 2-3 cards in both majors, but GIB will not continue after 3NT by partner. With fit (one should be 4th if not max) in both majors GIB might be not max, but not minimum as well. So likely 16-17. For second just GIB will not overestimate values once partner denies max and GIB has no reserves for game.
-
Actually for both minors openings it was 4+4+ ( only after 1NT direct overcall ) and it's corrected in next version. If overcall 1NT is in balancing seat, lengths are fine. So if it's reverse would show more points + length 5+4+. if it's lower by rank would show only length 5+4+.
-
It wouldn't as it shows 9+ minor cards and unlikely fit elsewhere to exist and while partner is in 12-22 range it needs 9+TP to proceed. And this time 2♠ would be 3- ♦, so West wouldn't be able to deny fit and poke in other suit.
-
If the pair has agreement to make cuebids over showing good major opening in GF sequence you probably shouldn't. GIB further calls are natural, unrelated with spades. like: 1♠ - 2♦ 3♠ - 4♣/4♦/4♥ You will probably get 5♥ answer, but is that enough ? Probably is as if partner doesn't have King of trumps he has 2A and hopefully aside king/queen to discard losing heart. If has the King + Ace then you need ♣KQ or ♥K. or ♦QJxx. Lots of combinations, but you need cue in hearts unless 4NT answer is 5♣.
