Hi all
Occasionally it would be nice and a bit less brutal if there were a bit of additional error variance added to GiB's play. Memory lapses etc
P
Page 1 of 1
GiB needs an error circuit
#2
Posted 2022-January-19, 18:30
Are you talking about the Advanced bots? My wife and I were practicing with a pair of basic robots and saw one of them toss a protected stopper for no good reason (not squeezed, not done to allow his partner an entry). I sent a robot report on the "weird" behavior.
#3
Posted 2022-January-19, 19:27
armantt2k, on 2022-January-19, 18:30, said:
Are you talking about the Advanced bots? My wife and I were practicing with a pair of basic robots and saw one of them toss a protected stopper for no good reason (not squeezed, not done to allow his partner an entry). I sent a robot report on the "weird" behavior.
While there are times it does it for no reason, most of the time it's because it assumes other players at the table are playing the same bidding system, interprets the bids accordingly, and won't believe for a moment that you don't have the points / shape you promised (thus assuming you must have cards that you don't, and then thinking what it throws is therefore irrelevant).
#4
Posted 2022-January-24, 10:48
armantt2k, on 2022-January-19, 18:30, said:
Are you talking about the Advanced bots? My wife and I were practicing with a pair of basic robots and saw one of them toss a protected stopper for no good reason (not squeezed, not done to allow his partner an entry). I sent a robot report on the "weird" behavior.
I think the OP was asking for GIB to be programmed to make mistakes sometimes, so it will be more like a human player. Like when a human doesn't realize that a spot card has become good, or miscounts a suit, or doesn't notice an important discard.
But GIB makes enough mistakes of other kinds that I think it makes up for it.
#5
Posted 2022-January-24, 15:03
barmar, on 2022-January-24, 10:48, said:
I think the OP was asking for GIB to be programmed to make mistakes sometimes
Smerriman presented a pretty good case that this is true in one algorithm (a >= error that nobody bothered to fix).
And if it was open source, who knows, but then again probably this is academic.
I have one partner who is allergic to post mortems and argues that all corpses stink

Page 1 of 1