Re: [RFC] Over half the "Golf" kpat games are unwinnable

El dimecres, 7 d’agost de 2019, a les 23:29:31 CEST, Shlomi Fish va escriure:
> shlomif[kpat]:$base/b$ head kpat-golf.stats.txt
> 1 lost (57 ms)
> 2 won (20 ms)
> 3 lost (160 ms)
> 4 won (66 ms)
> 5 won (8 ms)
> 6 lost (182 ms)
> 7 lost (32 ms)
> 8 lost (65 ms)
> 9 lost (263 ms)
> 10 won (411 ms)
> shlomif[kpat]:$base/b$ tail -2 kpat-golf.stats.txt
> 15354 lost (55 ms)
> 15355 lost (49 ms)
> shlomif[kpat]:$base/b$ cat kpat-golf.stats.txt | perl -lapE '$_=$F[1]' | sort |
> uniq -c | sort -n
>    6905 won
>    8450 lost
> shlomif[kpat]:$base/b$
> 
> ==============
> 
> This may make it frustrating to play. What do we do?

Isn't this a known thing with solitaire games? Sometimes even if you're amazing playing the game may be unwinnable.

But i agree that one of the reasons for using a computer vs actual cards is that you can ask it to do "magic" for you, so maybe we could have an option in the settings menu saying "only generate winnable games"? 

though maybe that should be per game? Do we have solvers [that are fast enough] for all games?

Cheers,
  Albert

> 
>