Fixed - Simulator closes app unexpectedly

When clicking around and playing with the virtual joystick the simulator closes the app and returns to the home screen. I can't find the exact sequence causing this, but it seems to be a combination of both clicks and joystick.

One thing that might help for debugging is in Applications/Utilities/Console - you might be able to find a stacktrace from when the crash occurred to pinpoint what object it was caused by.

Yes an output reading from the console would help a lot :)

How exactly do get get the stacktrace from the console? (I'm not a Mac user, so I'm not familiar with where things are)

The Applications/Utilities/Console just shows me an empty console prompt, what do you want me to do with that? Spawn the simulator from there? If so, how do I do this?

We are talking about the XCode console, not the Mac terminal/console :)

You can find the right one when you have the XCode window in front and go to "Run -> Console"

The console will always be there containing information from all the previous runs. When the game crashes it should throw a stacktrace in that window.

Sorry, I got the wrong console - the one I'm thinking of (and I suspect has the same result as opening it from Xcode like Andos mentioned) is in Developer/Applications/Console... I think.

Here is the console output:

[Session started at 2010-09-08 07:27:46 -0600.]
2010-09-08 07:28:05.671 iosTest[302:207] Got opengl error: 1282
2010-09-08 07:28:43.556 iosTest[302:207] *** Terminating app due to uncaught exception 'NSRangeException', reason: '*** -[NSConcreteData subdataWithRange:]: range {212945, 4132204093} exceeds data length 21101513'
*** Call stack at first throw:
(
0 CoreFoundation 0x02826919 __exceptionPreprocess + 185
1 libobjc.A.dylib 0x029745de objc_exception_throw + 47
2 CoreFoundation 0x027df078 +[NSException raise:format:arguments:] + 136
3 CoreFoundation 0x027defea +[NSException raise:format:] + 58
4 Foundation 0x0026ee4a -[NSData(NSData) subdataWithRange:] + 381
5 iosTest 0x00004c2c -[CFile getSubData:] + 95
6 iosTest 0x0007eea1 -[CSound load:] + 296
7 iosTest 0x0001054b -[CSoundBank load] + 655
8 iosTest 0x000150f8 -[CRunFrame loadFullFrame:] + 4149
9 iosTest 0x00008964 -[CRunApp startTheFrame] + 269
10 iosTest 0x00007e9c -[CRunApp playApplication:] + 276
11 iosTest 0x00003b36 -[CRunView timerEntry] + 457
12 Foundation 0x0024ac99 __NSFireTimer + 125
13 CoreFoundation 0x02807d43 __CFRUNLOOP_IS_CALLING_OUT_TO_A_TIMER_CALLBACK_FUN CTION__ + 19
14 CoreFoundation 0x02809384 __CFRunLoopDoTimer + 1364
15 CoreFoundation 0x02765d09 __CFRunLoopRun + 1817
16 CoreFoundation 0x02765280 CFRunLoopRunSpecific + 208
17 CoreFoundation 0x027651a1 CFRunLoopRunInMode + 97
18 GraphicsServices 0x034162c8 GSEventRunModal + 217
19 GraphicsServices 0x0341638d GSEventRun + 115
20 UIKit 0x004c2b58 UIApplicationMain + 1160
21 iosTest 0x00002912 main + 104
22 iosTest 0x000028a1 start + 53
)
terminate called after throwing an instance of 'NSException'
Stack dump:
0. Running pass 'Linear Scan Register Allocator' on function '@gldLLVMFPTransform27'

I had a CSound-related stacktrace as well - even if it doesn't seem to be in a place where sound is called at all, try removing all sound from the application before building.

This seems to happen when loading the frame itself, not when playing the sound.
Can you give me any information on which sounds files you use in your game? Format, length and bitrates would be good. :)

Here are the details of the track that's probably involved.

Although I'm not sure why it doesn't occur on start up. I suspect it's because normal buttons disappear but may still be click-able.

Format: .wav file
Length: 1:40
Size: 17 MB (in raw format)
Bit Rate: 1411 kbps

This bug is still occurring in beta 2.

Can you send me your MFA?

Sure

I'm still getting the "data out of range"-type exception as well - here's another MFA that demonstrates the issue.

http://www.clickteam.info/davidn/games/neverlinkhere/gravqx-iphone.mfa

One more crash log file, after removing some of the sound from the application - it seems to always fail to start up if the application tries to play any sound at all. Notably, this one "causes integer overflow" rather than indicates that it's gone out of the expected range.

The bug in the sounds is corrected.