Negative width breaks collisions?

Bug?

http://dl.dropbox.com/u/13338856/negati ... %20bug.cap

Feature

Use a dummy

I really didn't want to use a dummy, but If I that's the only solution I'll use it. Any other workarounds???

Not sure if it will work but you could try overlap at offset.

Might have to include a condition where width is less than original width, or a variable with the starting width.

How is exactly is it a feature newt? I'd say it's a bug. It would be so much easier to do some things if negative widths didn't break collisions. You can rotate the sprite 180* and turn on auto mirror if you want to be able to flip it with out breaking collisions. You can also make another animation, but that's a waste of vram.

A feature is any unexpected behavior that can be exploited. Being able to have negative values should defiantly qualify for that definition, as you would expect an error.

Although broken collisions are somewhat less exploitable, it should still qualify. Especially from the standpoint of the alternatives.

They are:

Stop the ability to have negative values.

Rewrite it to allow for collision detection in negative values.

Leave it as is, and possibly fix it in some future release.

Given the situation we have for the upcoming 1.0, and the probability that the devs might consider the first alternative easier than the second, I say lets go with the third.

I agree with davioware. I don't think this was intentional. Also negative height might be bugged as well. I haven't tried that yet.