[Kde-games-devel] Review Request 109539: KNetWalk: New score calculation method

[Ashwin Rajeev]

> On March 19, 2013, 3:05 a.m., Parker Coates wrote:
> > If we want to switch to a new scoring algorithm, there's a lot more to it than just changing the code calculating the score.
> > 
> > First we'd need a method of dealing with the old highscores. Do we just drop them? Do we try to convert them to the new system? Is that even possible. Leaving them in place is not an option as it's a completely apples-to-oranges ranking.
> > 
> > Second, we should try to future proof the highscore logic so that it doesn't need to be changed again later. The big one I can think of is sub-second accuracy. Whole seconds are a pretty big unit to work with when comparing times, especially on Easy difficulty.
> > 
> > Third, any changes need to be reflected in the documentation.
> > 
> > I'm not trying to be pessimistic or imply that it can't be done. It just isn't a change to be taken too lightly.
> 
> Ashwin Rajeev wrote:
>     I know the problem is complicated that is why i asked other's opinion in mailing list. But it seems there are only few active members.
>     
>     Is it possible to delete/change highscore by checking if it created with old method?
> 
> Albert Astals Cid wrote:
>     Deleting the highscore is the least of your problems, you only need a kconfig entry to know if you've cleared them or not and to clear them just wipe out the underlying KConfig entries.
>     
>     My concerns are various:
>      a) Is the new highscore algorithm really better
>      b) What if the user does *not* care about if the new highscore algorithm is better and he doesn't want to lose his highscores?
> 
> Ashwin Rajeev wrote:
>     a) I believe this this algorithm is better than current one. But both algorithm doesn't consider the minimum moves required to complete the game. So i think i need to consideration this factor while calculating the highscore or restrict it within a certain range.
>     b) I don't have a solution to this issue.
> 
> Albert Astals Cid wrote:
>     a) So you think its better but can still be better. Let's aim for that "still better" one, don't want to delete everyone highscores again next month no?
>     b) That's easy, ask them and keep the old code around (after all it's 2 lines)

a) I made minimum moves constant with new shuffling algorithm. So i think now this highscore algorithm is ready to use.
b) With new shuffling algorithm users cant beat their old highscore because of constant minimum moves to win. So i think it will be better to remove old highscores. But i dont know how to do that?
Is it possible to get sub-second accuracy with time calculation.


- Ashwin


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/109539/#review29484
-----------------------------------------------------------


On March 23, 2013, 1:24 p.m., Ashwin Rajeev wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/109539/
> -----------------------------------------------------------
> 
> (Updated March 23, 2013, 1:24 p.m.)
> 
> 
> Review request for KDE Games.
> 
> 
> Description
> -------
> 
> Let the score be time taken to complete in seconds, that also means one with less score gets on top.
> Each penalty move will add 10 to score.
> 
> 
> This addresses bug 172089.
>     http://bugs.kde.org/show_bug.cgi?id=172089
> 
> 
> Diffs
> -----
> 
>   src/mainwindow.cpp dc437fe 
> 
> Diff: http://git.reviewboard.kde.org/r/109539/diff/
> 
> 
> Testing
> -------
> 
> It may make problem with previously saved high scores.
> 
> 
> Thanks,
> 
> Ashwin Rajeev
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-games-devel/attachments/20130409/b0a4729d/attachment.html>