BBO Discussion Forums: Deep Finesse Bug - BBO Discussion Forums

Jump to content

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

Deep Finesse Bug

#1 User is offline   lamford 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,417
  • Joined: 2007-October-15

Posted 2015-July-16, 06:01


Matchpoints.

In the pub after the game, we often do not believe Deep Finesse, but it is always right and stated EW could only make eight tricks in hearts here. We quickly found the only winning defence, which is an initial diamond lead and North wins and continues diamonds. South ruffs the second spade and can cross in clubs for a second spade ruff. However, when one loads the hand into Deep Finesse (and I just downloaded it again in case my version was corrupted), it does not analyse a heart contract at all, perhaps in disgust at our opponent's auction. It displays the message:

"New cache cell: Existing cache Cell: Unexpected Error: INTERNAL ERROR: inconsistent prc won results",
under the banner of Deep Finesse Alert.

Does anyone else get this error, as it is the first time I have ever seen it? And it handles other hands fine still!
I prefer to give the lawmakers credit for stating things for a reason - barmar
0

#2 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,068
  • Joined: 2004-April-22
  • Gender:Female
  • Location:UK

Posted 2015-July-16, 06:18

When I press "GIB" on your diagram it produces the correct solution. I thought GIB uses the same DD solver as Deep Finesse. Or is that no longer true?

Not that I don't believe you of course, just a suggestion that it might be a problem in the surounding code rather than with the DF algorithm itself.
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#3 User is offline   lamford 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,417
  • Joined: 2007-October-15

Posted 2015-July-16, 06:25

View Posthelene_t, on 2015-July-16, 06:18, said:

When I press "GIB" on your diagram it produces the correct solution. I thought GIB uses the same DD solver as Deep Finesse. Or is that no longer true?

Not that I don't believe you of course, just a suggestion that it might be a problem in the surounding code rather than with the DF algorithm itself.

Thanks, helene. If anyone has Deep Finesse separately, they can confirm your theory.
I prefer to give the lawmakers credit for stating things for a reason - barmar
0

#4 User is offline   bgm 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 129
  • Joined: 2012-December-07

Posted 2015-July-16, 09:20

I receive the same error after inputting the deal. The version is 2014.2
0

#5 User is offline   lamford 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,417
  • Joined: 2007-October-15

Posted 2015-July-16, 10:15

View Postbgm, on 2015-July-16, 09:20, said:

I receive the same error after inputting the deal. The version is 2014.2

Thanks. My version is the same, released on May 24 2014. If you swap the eight and ten of spades, it can analyse the hand then! But if you swap the ace and jack of diamonds, so that most leads beat 3H, it cannot do so.
I prefer to give the lawmakers credit for stating things for a reason - barmar
0

#6 User is offline   broze 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,001
  • Joined: 2011-March-08
  • Gender:Male
  • Location:UK

Posted 2015-July-16, 14:16

I have had actually had a lot of bugs with Deep Finesse. Sometimes it will take you through to trick 8 and suddenly decide to say that DD is a different number of tricks. I have switched over to Bridge Solver which has given me no trouble at all and has lots of great capabilities.
'In an infinite universe, the one thing sentient life cannot afford to have is a sense of proportion.' - Douglas Adams
0

#7 User is offline   joemanjo 

  • PipPip
  • Group: Members
  • Posts: 20
  • Joined: 2007-January-25
  • Gender:Male
  • Location:Bangalore India

Posted 2015-July-17, 09:30

I have an old version of DF, circa 2000. It does not have any bells and whistles, only Win and Loss marked against each card in the hand on lead, does not give number of tricks plus/minus like GIB. The solver had no issues in identifying the lead which beats it and subsequent play as well.
0

#8 User is offline   Shavit1 

  • PipPip
  • Group: Members
  • Posts: 11
  • Joined: 2015-May-23
  • Gender:Male

Posted 2015-July-18, 05:27

I have the 2000 and 2013 versions, both work perfectly. Have removed version 2014 permanently after some bad experiences. All its extra features were useless for me anyway.
Shavit
0

#9 User is offline   slothy 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 690
  • Joined: 2003-October-14

Posted 2015-July-18, 10:47

It uses the same original code base.....

As far as i am aware, DF tried to refactor code for later update to make it faster (which it is :) ) ...this may be the source of the bug in the later versions and slipped through the testing suite....

Alex
gaudium est miseris socios habuisse penarum - Misery loves company.
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