The Crusher "Bug"

[Sun]

Normally, this would fall under a suggestion, but since it has been mention many times in the past, I felt that there should be a topic of discussion on this much similar to that Super form thread.

This has been discussed at various points in the past, but I can't help but feel as if this occurs in SRB2 far more than it did in the classic games. Recently in a Coop game in DSZ2, I saw the crusher bug occur three times to players(and one or two of the times it happened, they were Tails, who has a lower height). For those who don't know, the crusher bug is when you jump into the gap between a floor and a rising platform,, and get "crushed" by the rising platform once you land. This apparently happens as a result of your character standing up, and since your character's height is great than the current gap, you get crushed.

Now in the classics, I think I had it happen to me once with Hill Top Zone's rising platform sequences, but I'm pretty sure it didn't occur elsewhere(an example being the rising pillar in Sandopolis Zone; I've also tried to get the bug to occur in Metallic Madness, and it hasn't happened so far).

Basically, I don't think this should be present in SRB2, as it is really doing more harm than "good."

 

[Liliam]

http://dl.dropbox.com/u/26940009/files/srb2/crushtest.zip

Here is a test wad for the crusher bug. In the center there is a sector with a ceiling too low to stand up, but big enough to fit while spinning and jumping. It should prove to be impossible to die here, because when you land from a jump, Sonic tries to stand up, but is immediately thrown back into a spinning position because he can't fit.

Off to the sides are two moving sectors. One moves down, the other up. They are separate simply for the convenience of being able to spindash directly into either one.

When the sectors are moving down, everything is fine. The problem presents itself when the sector starts accelerating back up.

There must be some sort of weird race condition happening here that's causing Sonic to die before he has a chance to try and tuck into a roll. The narrower the space Sonic has to fit inside, the quicker he mysteriously dies when the sectors start moving up.

 

[SSNTails]

Well, the floor and ceiling get moved independently... so maybe the floor moves up first, then it's too tiny of a space, and he dies before the ceiling gets a chance to move up as well?
Maybe the thinker for these things needs to decide when it's appropriate to move the floor first and when it's appropriate to move the ceiling first to prevent this.

 

[Kaysakado]

But why would the crush check be in between the floor moving and the ceiling moving? Or are you saying they're in completely different tics?

 

[SSNTails]

No, the check happens every time a floor or ceiling moves. I don't feel like explaining it right now, but this is the proper way for it to work.

If a dual floor-ceiling mover were written to be a little bit smarter as to whether to move the floor or ceiling first, I think this problem might be fixable.

 

[Sun]

The main crusher bug I'm talking about though is for example when you jump and press against the crushers in THZ2, and die when you land just after the crusher starts moving up. You can also do this by pressing against the opening door in ERZ2 after the two blue polyobjects(here, the floor actually goes down). No rising floor is involved with those.