|
From: | Joe Wells |
Subject: | [glob2-devel] hidden forbidden areas, successful game counter for releases (was: Rewrite and disappearance syndrom) |
Date: | Wed, 10 Oct 2007 16:57:53 +0100 |
User-agent: | Gnus/5.11 (Gnus v5.11) Emacs/22.1 (gnu/linux) |
"Leo Wandersleb" <address@hidden> writes: > * globs get trapped by invisible forbidden areas (place forbidden to > force recalculation of forbidden gradient) Do you think this is the problem, that the forbidden gradient is not being recalculated properly when the hidden forbidden area are removed? My suspicion up to now has been that the problem is that hidden forbidden areas are sometimes not being removed at all. I knew already that placing ordinary forbidden areas could sometimes solve problems caused by the hidden forbidden areas, but I have not been sure why this can solve the problem. I still recommend making the hidden forbidden areas not be hidden. > Only release what really works and test everything else for at least > 100 games. with only 3 games played per week that takes long but > still i would vote to have an official game-report counter to the > next release with strict feature freezes. and 100 games is not a > joke. if we want 100 games to be played per day some day this is > only a few games. I like this idea of requiring a certain number of games to be played before releasing. Only successful (no crashing or serious bugs) games should qualify. I like even more the idea of an automated counter of some sort. For non-network games (like I play) the game could offer to fill out a web form to increment the counter. I think it is good to count successful network games because the testing that I have done only finds bugs in non-network games because I don't play in network mode (my laptop computer is far too slow to even consider this unless someone finds a way to double the speed of glob2). -- Joe
[Prev in Thread] | Current Thread | [Next in Thread] |