PiBoSo Official Forum

GP Bikes => Bug Reports => Topic started by: BerScott on April 24, 2014, 03:39:10 PM

Title: CTD after bike crash
Post by: BerScott on April 24, 2014, 03:39:10 PM
It happens specially in the track I'm actually working in. Sometimes, not always, when I crash and the bike falls, GP Bikes freezes and an error window is displayed. I press "OK" and then another error window appears. Here are two pics of the two error messages:

First error message
(http://i.imgur.com/Swv3LCu.png)

Second error message
(http://i.imgur.com/Kp3H2Aj.png)

And the Windows 7 app crash message
Problem signature:
  Problem Event Name: APPCRASH
  Application Name: core.exe
  Application Version: 0.0.0.0
  Application Timestamp: 52b0b0ff
  Fault Module Name: core.exe
  Fault Module Version: 0.0.0.0
  Fault Module Timestamp: 52b0b0ff
  Exception Code: 40000015
  Exception Offset: 0011a978
  OS Version: 6.1.7601.2.1.0.256.48
  Locate ID: 3082
  Additional Information 1: 20ea
  Additional Information 2: 20ea213a5d13c436ee7599369789f165
  Additional Information 3: 74b7
  Additional Information 4: 74b710e7070e580f84e5feba14467a6b
Title: Re: CTD after bike crash
Post by: C21 on April 24, 2014, 06:13:43 PM
One specific bike or several bikes? Modded or default?
Title: Re: CTD after bike crash
Post by: BerScott on April 24, 2014, 06:19:29 PM
Default and modded bikes.
Title: Re: CTD after bike crash
Post by: RBp on April 24, 2014, 06:24:02 PM
ode error is has something to do with the model I think, hard to explain.......   if a fence is only a single plan and the bikes falls dead on top it gives this error. it why it ask for Armco/walls to be model in 3d I think.

Also happen with holes in floor and you hit the edge the wrong way

Hope that make sense,  there could be other cause to which I don't know about yet.
Title: Re: CTD after bike crash
Post by: HornetMaX on April 24, 2014, 07:53:18 PM
That error is produced by an assertion failing in the ODE solver used to integrate the vehicle dynamics (which, btw, is this one: http://www.ode.org/ (http://www.ode.org/)).

This kind of error typically happens when a vector that must be normalized turns out to have zero length. Now why that happens is hard to say, could be due to plenty of things: weird cases in collision detection, singularities in kinematics (e.g. quaternions), plenty of stuff.

I've seen this error in GPB a bunch of times in beta3 but probably only once or twice in beta4 since it's out.
If you have it regularly, it's likely to be a track problem.

MaX.
Title: Re: CTD after bike crash
Post by: BerScott on April 24, 2014, 08:08:00 PM
Max, I think you're right, it seems to be a track problem. This track has lot of terrain irregularities, so it must be what you've guys said.