For some reason the iPhone simulator takes several seconds (~12-13s) to start my app when I run it in iOS 6.0 or 6.1 mode (that's before the app's default screen appears, so it doesn't seem to be caused by the app's code). I think that's actually slower than when I start the app on the device. If I switch to 5.0 or 5.1 it takes just 2-3 seconds.
Has anyone else noticed this? Any solutions? (I've cleared the whole ~/Application Support/iPhone Simulator directory, but it didn't help.)
You can select iOS Simulator from top menu. After it you can Reset Contents and Settings (third from top). it will reset you simulator. Remove all application from your simulator. Please try after it.
So I made an app, And it runs on all possible platforms except the ipad device. I would understand if it ran on the ipad device and not the iphone device since the processor is better on the ipad, but this has me stumped. It also worked fine on the ipad simulator. Can anyone think of a reason why this could be?
There are too few details to give an explanation. A possibility is: when you run the app in two different devices, iPhone and iPad, some nib files could be different. Even the app delegate is different if you use XCode templates for universal apps. So this means that it is possible that the code executed is different in the two devices.
So imagine that in the iPad app delegate / view controller you're allocating - at launch - a lot of memory (scroll view with many large images for example). Then the app will not crash in the simulator (it uses your Mac memory). The app will not crash in the iPhone (different code). So at the end it will crash in the iPad only.
Other possibilities: you're going short in memory very soon. But the iPad 1 has less memory than iPhone 4G.
Other possibility: the watchdog timer kills your app at launch. It happens on the iPad 1 only because you're loading a few large images and it takes a lot of time, just enough to be killed by the watchdog.
Other possibility: iPad 1 is still with iOS 3.2, while iPhone and simulator are linked to iOS 4.0. In this case it is easy to have the app crashing in iPad 1 by calling one of the many new APIs introduced with iOS 4.
As you can see there are many possible explanations...
add target for device or convert it to universal application
Yes, it's easy to target both platforms. If you login to ADC and read the Programming Guide, it'll provide the basics for specific development. I can't legally speak of the "how."
Um. If you follow their directions, build two targets, share your source within the same project... then you'll be set. It's the same SDK, just different Nibs & main(), from what I can tell.
You say different xibs... you might be missing an Outlet connection somewhere.
Ok gents, sorry but I did not provide enough information for any of you to correctly answer this, but! here was my problem.
I had a login screen with 2 textfields, and a submit button. The submit button would release the current view and add a new view. I would usually press submit while the keyboard is being shown. Aparantly this makes everything blow up. You must manually hide the keyboard first. Hopefully this saves somebody some time.
The GUI of my application stops responding after the iPod has been in standby mode. The application I have made is based on the "Utility application" template from the wizard in xcode - but the original application still works after the iPod returns from standby.
My application plays music (AudioUnit) and when the iPod enters standby the music stops but when the iPod leaves standby mode the music starts playing again but as mentioned the GUI does not respond to touches anymore. All I can do is to shut it down.
I have read about this problem here:
What happens to an iPhone app when iPhone goes into stand-by mode?
but that did not help me enough to solve the problem.
Sounds like your Audio Unit has been stopped.
If your GUI relies on the AU's progress then that would explain the problem.
Check your AU's state in applicationDidBecomeActive.
I didn't think standby mode stopped AUs, I thought only audio interruptions
did that. Anyway, you should probably look at AudioSessions as well, as you'll
find that phone calls, alarms, SMSs and ipod music will also stop your Audio Unit.
Good luck.
Actually, I've been running into this problem on the iPod application itself, in just general daily use. That is, if you are using the 3.0 firmware OS, then this is most likely a bug in the iPhone. I guess you can relax, and know that it's probably not your own code that is causing the problem. I'd file a radar ticket with Apple if you haven't already so that they can address it in the next bug fix release.
It all happens just because I have an UISlider in an UIToolbar. If I remove the UISlider from my GUI it still responds after waking up from standby/sleep mode.
Can someone explain me why that causes this problem (is it a bug in Apple's code)?
The solution is here:
http://www.iphonedevsdk.com/forum/iphone-sdk-development/4558-app-crashing-upon-quit.html
Simply don't put a slider in a toolbar. Make a work around so it looks like the slider is in the toolbar instead. It is a bug in Apple's code.
So I'm developing and iPhone game right now and everything has been working just fine when I test it on my first gen iPod touch. Last night, I added in some NSUserDefaults stuff to save a few variables for it. I ran it on the simulator a few times and it worked perfectly, but when I built the app and put it on my iPod, all I got when I launched it was a black screen. Anyone have any ideas? They would be much appreciated, thank you.
I recommend deleting the application from your iPod Touch first then trying to install it again. When are you trying to read from NSUserDefaults? In your applicationWillFinishLaunching call? If so, try to comment that out to ensure that it really is the source of your problem.
You need to run it on the iPod touch with the debugger switched on - you should be able to track this down.
Do you know about setting a breakpoint for exceptions - if you tell the debugger to break on objc_exception_throw it is extreemly useful in these cases.
Do you have a custom Default.png file? If no, then you will see a black screen while your application loads. Depending what type of (possibly unending) operations you are doing on load this might be the reason you are seeing a black screen. Remember the simulator runs on top of the processor in your mac which is significantly faster than the processor in your iPod Touch.
I have heard of apps not working properly on the simulator but working properly on the actual iPhone device. Has anyone experienced an app that runs perfectly in the simulator but not on the actual iPhone device?
Filenames are case-sensitive on the iPhone, but not in the simulator.
So, for example, if you try to load an image with UIImage *iconImage = [UIImage imageNamed:"MyIcon.png"], but your resource is actually named "myicon.png", then it will work on the simulator, but not on the device.
If your app is graphics intensive, like say a game, the performance of the simulator DOES NOT resemble at all that of the hardware. Your application will probably be smooth and work great on the simulator, but on hardware it'll likely render at a crawl unless you know what your doing. You can easily go from 60fps to 3fps between Simulator and hardware.
The order in which function/constructor parameters are evaluated is different:
int i = 0;
int f() { return ++i; }
int a, b;
int test(int p1, int p2) {
a = p1;
b = p2;
}
test( f(), f() );
//simulator: a = 2, b = 1
//device: a = 1, b = 2
Trigonometry functions may return different results:
float a = cosf( 0.108271248639 );
printf("%.12f", a);
//simulator: 0.994144439697
//device: 0.994144380093
I know there are some differences in the OpenGL ES implementation between the device and the simulator. From what I understand this is mainly because of the graphics chip on the iPhone (PowerVR MBX) having vastly different capabilities than other mac machines. Many of the hardware limits are not enforced in the simulator, so it is entirely possible to get something running in the simulator that will totally crash on the device.
There are also some OpenGL ES extensions that are supported by the iPhone hardware that are not supported in the simulator. I believe the major one is PowerVR texture compression (PVRTC).
Another problem area can be memory footprint. Anecdotally, I have not seen the simulator automatically enforce the memory limitations of the device. Therefore, it is possible to have something that runs in the simulator fine, happily consuming copious amounts of RAM and never bothering to free any of it only to be swiftly terminated after a short continuance of such behaviour when running on a device.
There are certain bits of code that won't work on the simulator (using the iPhone Keychain, for example), but for almost all applications, the simulator will work exactly like the iPhone.
That said, there's absolutely no replacement for testing on a real device.
I had a problem running a relatively simple 1/30 sec timer to do updates for a game. It runs fine in the simulator, and freezes out input on the device.
Also note that you will be compiling against the OS X frameworks (where applicable) when building for the simulator so you could be using methods and classes that aren't available on the iPhone versions of the frameworks.
One example I can think of off the top of my head is NSPredicate. I was able to compile and run an app using NSPredicate in the simulator, but it wouldn't compile for the device since that class isn't available.
Fingertips are larger than the 1 pixel endpoint of a mouse cursor. To do proper, even minimal, usability testing you should deploy your App to a device.
If you enable GCC_ENABLE_FLOATING_POINT_LIBRARY_CALLS, your app will crash all over the place in the simulator but work on the iPhone.
Quartz graphics calls in the iPhone simulator are faster than Java2D calls on the same computer--wicked fast.
I've had issues in memory-hungry applications where the Simulator would work just fine (because it would assume the iPhone/iPod Touch's memory was all yours to play with), while the device would crash (because other apps had leaked and Apple background services had eaten up some memory) and I hadn't implemented proper memory management or a response to the didReceiveMemoryWarning selector.
One big thing that took me a while to spot was that the simulator does not support device tokens, so any code that involves those will not run on the simulator.
I had a bug where the app would work fine on the simulator, but would crash when I ran it on a device because there was a bug in the device token code. I couldn't figure it out for the longest time!
There are many trivial examples. For example you can allocate far more memory on the simulator than on a real phone. You cannot receive push notification on the simulator if you don't have a retina Mac, the display dot pitch is different.
At a more fundamental level, the simulator is just that, it simulates the iPhone OS X using Mac OS X. This is evidenced by the fact that the filesystem on the simulator may not be case sensitive but on the phone it will be. More subtly, it does not emulate hardware so things like location will not work the same and 3D is going to be very different - especially if you are using Metal.
You should always test on real hardware.
Without considering the performance differences between the two, there used to be some things that the simulator didn't do correctly - i.e. it would mess up audio in some cases (see this question). However since the 2.2 SDK this issue has been resolved and the sound seems to be fine in the simulator. That's not to say that there is some other incompatibilities lurking down there! (Just none I've run into)
Regarding sounds, I was having the same problem. The issue was that the sound encodings that the Simulator supports is a different set of sounds than the device. I hope that helps.
I had many problems with libraries and frameworks when moving from the simulator to the device. Not least that they seem to have different architectures!
I have seen the positioning of objects, like toolbars be different on simulator than on the phone. Very annoying.
Yeah....
Apps compiled for 2.x will work fine on 3.0 device, but it will crash on 3.0Simulator
Note: 1. If you compile for 3.0, app will work fine on 3.0 simulator also...
2. a)Compile for 2.x and launch the app in simulator.
b)Now change the iPhone Simulator Hardware to "3.0".
c)Then launch the app we installed earlier in step a).
CRASH !!!!!!!!
movie file (m4v type) as my exprience is first time playing properly
but at second time it flickers screen of simulator...
whereas in iPhone device it works fime...
I had some sound effects that played fine in the simulator, but not on the device. I had to change the format to something that the device would handle.
If status bar of application is hidden,In case of simulator it still consumes touch event. But in the device it behaves perfectly.
Yes - it happened to me the other day. I'm new to the iphone, and so had deleted MainWindow.xib thinking it was unused. The app worked perfectly on the simulator - but crashed when installing on the phone.
Another issue we ran into was our three20 dependencies, which were set to ios 3.2 instead of 4.1. Worked perfectly in the simulator, but bombed on the device (since the files were compiled for the wrong arch).
iphone video library is not accessible
in simulator but code work fine on
actual device
Resource loading in the simulator is MUCH faster than in the device. For example, loading and displaying a sequence of full-screen UIImages (like a rudimentary video) can look very smooth in the simulator, and choppy on a device.
In fact, remember that there is a huge speed difference between different devices. The original iPhone and the iPhone 3G are slower than the iPod touch 2nd Gen, which is also much slower than the iPhone 3GS, and so on.
When trying to access the UIDevice.currentDevice(), it returns iOS Simulator instead of the actual device you're testing. This sucks, since you can't do certain things on the simulator.