View Issue Details

IDProjectCategoryView StatusLast Update
0001135FSSCPphysicspublic2007-07-24 17:15
ReporterArcanum Assigned To 
PrioritynormalSeveritycrashReproducibilityalways
Status resolvedResolutionunable to reproduce 
Product Version3.6.9 
Summary0001135: ai doesn't like you killing it's target too fast? (FSP sm2-04a "Where Eagles Dare")
DescriptionAssert: vm_vec_mag(&rot_axis) > 0
File: C:\temp\fs2_open_3_6_9.t\code\Math\VecMat.cpp
Line: 2294
[This filename points to the location of a file on the computer that built this executable]

Call stack:
------------------------------------------------------------------
    ai_formation() ai_frame() ai_process() ship_process_post() obj_move_all_post() obj_move_all() game_simulation_frame() game_frame() game_do_frame() game_do_state() gameseq_process_events() game_main() WinMain() WinMainCRTStartup() kernel32.dll 7c816fd7()
------------------------------------------------------------------

FSO: RC7 FSP: 3.0.3
Additional InformationI have no clue what category this falls under, so I'll put it under physics, just 'cuz it looks like it involves vector math somehow.

I discovered this when investigating another, apparently unrelated crash bug in the same mission: 0000233:0001134

The assert occurs if one or more ships in Gemini wing are destroyed very quickly at the start of the mission (via cheat ~+k)

Judging by the call stack, it appears that if Beta wing (which has initial orders to attack Gemini wing) loses its targets too quickly, too soon, it's attack vector or formation gets fubar'd.
TagsNo tags attached.

Activities

taylor

2006-11-06 11:17

administrator   ~0007088

Please try to replicate this with the (recently updated) rc7dot9 build: http://icculus.org/~taylor/fso/testing/rc7dot9.rar

I can't get this to happen so I'm not sure if it has recently been fixed in another way, or if I'm just not able to replicate it as easily as you can.

Arcanum

2006-11-09 16:07

reporter   ~0007099

Appears to be fixord in dot9

taylor

2006-11-09 18:02

administrator   ~0007101

Hmm, ok. I'll leave this open until this weekend, in case you or anyone else can replicate it again. I'm not quite sure which fix might have dealt with this problem, and I'll feel a lot better about resolving this if I know that much at least. :)

Goober5000

2007-07-24 10:11

administrator   ~0008277

Can we close this?

taylor

2007-07-24 16:48

administrator   ~0008279

Yeah. Still don't know what the problem/fix was though. It might be better to simply mark it a not-a-bug, or anything else other than "fixed", though.

Goober5000

2007-07-24 17:15

administrator   ~0008281

K. Marking as "unable to reproduce".

Issue History

Date Modified Username Field Change
2006-11-03 02:02 Arcanum New Issue
2006-11-06 11:17 taylor Note Added: 0007088
2006-11-09 16:07 Arcanum Note Added: 0007099
2006-11-09 18:02 taylor Note Added: 0007101
2007-07-24 10:11 Goober5000 Note Added: 0008277
2007-07-24 16:48 taylor Note Added: 0008279
2007-07-24 17:15 Goober5000 Note Added: 0008281
2007-07-24 17:15 Goober5000 Status new => resolved
2007-07-24 17:15 Goober5000 Resolution open => unable to reproduce