[aquaria] Scenario colision detection issues in achurch.org hg/aquaria tip

Henrique de Moraes Holschuh hmh+aquaria at hmh.eng.br
Thu Dec 30 22:45:32 EST 2010

On Thu, 30 Dec 2010, Ryan C. Gordon wrote:
> >Trying to track this is driving me mad :p  I will recompile with -O2
> >just to make sure it is not gcc doing something nasty.
> Recompile with -O0, so there are no optimizations at all.

Same problem.  Tried sending a screenshot earlier, but it was too large for
the list.  I will have to reduce the resolution, and send one screenshot per

Right now, I have Naija sitting in a "invisible wall".  Energy form blasts
hit these invisible walls.  They are nearby two pillars.  It is difficult to
tell with precision, but there are two pillars in the screen, and it DOES
look like there are TWO invisible walls that the blasts can hit (and Naija
can grab).  But they are not in an exact offset from the pillars.

Other pair of pillars that are in a similar configuration nearby (but in
slighly different rotations show the same behaviour, only their invisible
shadows are positioned a bit differenly.  Rocks are not working very well

Looks like the colision mask for the pillars are not exactly where the
drawing objects are.  I have no clue what is happening to the rocks,
sometimes they block the passage (e.g. vertically), sometimes not.

Since the problem happens only with doors, foreground pillars, and rocks,
does this tell you something?  Is there some debug mode I could enable to
see the boundaries of the objects?

  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh
-------------- next part --------------
A non-text attachment was scrubbed...
Name: problem-2.jpg
Type: image/jpeg
Size: 13276 bytes
Desc: not available
URL: <http://icculus.org/pipermail/aquaria/attachments/20101231/a453453a/attachment-0001.jpg>

More information about the aquaria mailing list