iPhone 14 Pro Camera Shaking and Vibrating in Apps Like Snapchat, TikTok, and Instagram for Some Users

[beeman]

Summary

Just days after its initial launch the iPhone 14 Pro and Pro Max rear cameras already giving users issues, apparently in apps like Snapchat, TikTok, and Instagram the rear camera's main lens vibrates uncontrollably while in use. Apple has yet to respond to requests for comment.

 

Quotes

Quote

The issue has been reported by users across Twitter, Reddit, TikTok, and the MacRumors Forums since the iPhone 14 Pro models launched on Friday, but it is unclear how widespread the problem is or what the underlying cause is. The main lens on iPhone 14 Pro models features "second-generation" sensor-shift optical image stabilization, and it is possible that the stabilizer is acting up due to a software bug, but this is merely speculation.

 

My thoughts

The broader issue here is the lack of support that Apple gives developers prior to releasing new new devices. It's not like developers can develop good software without access to big changes on image sensors, etc. that people rely on in their everyday use of the phone. It might be interesting to see what Kris Collins has to say about the iPhone 14 Pro, especially since she was given one by Apple themselves for content creation, where the a good amount of her work is on TikTok and IG. Pretty disappointing to see this kind of issue happening so early on such a "new" and "best ever device."

 

Sources

https://www.macrumors.com/2022/09/18/iphone-14-pro-camera-shaking-vibrating-issue/

[suicidalfranco]

"It's a feature"

"You're using it wrong"

"If you're going to complain about it why did you get an iPhone? Just buy an android"

^ //thread

 

[jaslion]

Sounds like the optical image stabilization is going crazy.

[Sauron]

Likely just a firmware bug, it's weird it wasn't tested before launch with some of these apps though... you have to expect a lot of people will use instagram and tiktok, right? Unless someone at Apple pushed a last minute change to the camera API right before launch which accidentally caused this side effect...

[DrMacintosh]

The problem is that if this is just a "software bug" that bug can cause real damage to the camera assembly as demonstrated by Luke in the OP. Apple must be willing to replace affected units once a fix is implemented. 

[BobVonBob]

I've got nothing but a hunch to go on, but I think the image stabilization is backwards. Instead of compensating left, it compensates right, which makes it worse, which makes it want to compensate even more, and the cycle repeats. If you've ever seen a someone attach power steering backwards it's that but with image stabilization.

 

Why this only happens in certain apps I've got no idea. Maybe those apps use their own image stabilization algorithm and Apple reversed the direction of the stabilizing motors this generation?

 

Spoiler

 

 

[wanderingfool2]

37 minutes ago, Sauron said:

Likely just a firmware bug, it's weird it wasn't tested before launch with some of these apps though... you have to expect a lot of people will use instagram and tiktok, right? Unless someone at Apple pushed a last minute change to the camera API right before launch which accidentally caused this side effect...

It depends on the probability of it happening really and what is the initial trigger of it.  There could potentially be an edge case that is a series of events that weren't quite tested.  At the moment, I am not really going to blame Apple until it becomes known just how widespread this thing is (or a fix is released so people can analyze what went wrong).

 

 

At least initially looking on Twitter, it might be an IOS16 bug...as there was at least one person who chimed in saying their older iPhone started experiencing the same thing

[tim0901]

35 minutes ago, BobVonBob said:

Why this only happens in certain apps I've got no idea. Maybe those apps use their own image stabilization algorithm and Apple reversed the direction of the stabilizing motors this generation?

Very much doubt that much control is exposed in the API. I wouldn't expect any control over the built-in OIS to be available to app devs - that's not the Apple "it just works" way. OIS shouldn't be something an app dev has to worry about and should be completely invisible to the developer.

[Rauten]

Looks like quite a QA snafu, specially since those apps are only used by, you know, tens of millions of users.

 

Someone's going to lose their job(s)...

[BobVonBob]

2 minutes ago, tim0901 said:

Very much doubt that much control is exposed in the API. I wouldn't expect any control over the built-in OIS to be available to app devs - that's not the Apple "it just works" way. OIS shouldn't be something an app dev has to worry about and should be completely invisible to the developer.

Maybe Apple has its own fancy OIS with special sauce and sprinkles that they use in their own camera app, and a currently broken OIS that other apps get when they want to access the camera. Now that would be the Apple way.

[Somerandomtechyboi]

"Iphone 14 pro max with incremental upgrades from the previous gen with very goofy ahh vibrating camera as a feature, buy now at 1200$ to make very goofy ahh videos"

[Dracarris]

2 hours ago, beeman said:

The broader issue here is the lack of support that Apple gives developers prior to releasing new new devices. It's not like developers can develop good software without access to big changes on image sensors, etc. that people rely on in their everyday use of the phone.

I doubt that the camera API apps use was changed in any way related to this issue such that early updates to apps would've prevented this. 

2 hours ago, beeman said:

Pretty disappointing to see this kind of issue happening so early on such a "new" and "best ever device."

It's an early firmware bug. Oh, it's from Apple you say? Lets create some headlines from it!

[Fasterthannothing]

1 hour ago, DrMacintosh said:

Apple must be willing to replace affected units once a fix is implemented. 

You must be new around here Apple doesn't admit they released a bad product much less do a recall 

[beeman]

24 minutes ago, Fasterthannothing said:

You must be new around here Apple doesn't admit they released a bad product much less do a recall 

They don't use the word recall, they use the word "service program," and there are a lot of them it would seem

 

https://support.apple.com/en-ca/service-programs

[wanderingfool2]

Apple has acknowledged the issue and is going to send out a fix next week

https://www.cnet.com/tech/mobile/apple-confirms-iphone-14-pro-camera-shake-issue-says-fix-coming-next-week/

 

 

One thing that I think will be interesting is whether or not they extend the warranty on the camera itself now.  There has already been the case of people reporting that it has messed up the focus on their phone and Apple replaced it...but I think it gets a bit tricky when you think that if your phone still works, it still could have been damaged, just not to the point where it doesn't function.  That might mean in a year or two when the warranty is gone you might start experiencing problem (as general wear and tear have pushed the components beyond the breaking point)

[Ravendarat]

2 hours ago, wanderingfool2 said:

Apple has acknowledged the issue and is going to send out a fix next week

https://www.cnet.com/tech/mobile/apple-confirms-iphone-14-pro-camera-shake-issue-says-fix-coming-next-week/

 

 

One thing that I think will be interesting is whether or not they extend the warranty on the camera itself now.  There has already been the case of people reporting that it has messed up the focus on their phone and Apple replaced it...but I think it gets a bit tricky when you think that if your phone still works, it still could have been damaged, just not to the point where it doesn't function.  That might mean in a year or two when the warranty is gone you might start experiencing problem (as general wear and tear have pushed the components beyond the breaking point)

Its possible but I would say unlikely unless its a wide spread problem in a few years. If its not a wide spread problem then the burden of proof would likely be on the owner to prove that this problem, which it sounds like should be fixed inside 3 weeks of launch, is what caused the problem. That would be an almost impressible thing to prove in my mind

[hishnash]

On 9/20/2022 at 3:07 AM, Sauron said:

you have to expect a lot of people will use instagram and tiktok, right? Unless someone at Apple pushed a last minute change to the camera API right before launch which accidentally caused this side effect...

I don't think you can use these apps while offline so they are likly not valid test targets as people with pre-release phones cant go testing apps that will sniff every little bit of info about the device they are running on and dump it to the server.  

 

They will download apps from the App Store and before opening them fully block all network traffic then open them and test... 

[Sauron]

13 hours ago, hishnash said:

I don't think you can use these apps while offline so they are likly not valid test targets as people with pre-release phones cant go testing apps that will sniff every little bit of info about the device they are running on and dump it to the server.

That's possible but I find it hard to believe Apple can't get Instagram to give them a test build with an NDA... it's in their mutual interest to have this work as smoothly as possible on launch. Typically large partners (i.e. the developers of some of the most popular apps on the store) would receive insider information on the new hardware and APIs before launch anyway.

[suicidalfranco]

2 minutes ago, Sauron said:

That's possible but I find it hard to believe Apple can't get Instagram to give them a test build with an NDA... it's in their mutual interest to have this work as smoothly as possible on launch. Typically large partners (i.e. the developers of some of the most popular apps on the store) would receive insider information on the new hardware and APIs before launch anyway.

I find it hard to believe that Apple, in testing, can't just spoof the HW info sent\sniffed to apps

[Sauron]

40 minutes ago, suicidalfranco said:

I find it hard to believe that Apple, in testing, can't just spoof the HW info sent\sniffed to apps

That too, but again, I'm sure they work directly with some of these developers

[BuckGup]

This has happened on every iPhone since the X came out I believe. It doesn't seem to be widespread but this has happened before 

[wanderingfool2]

3 hours ago, Sauron said:

That too, but again, I'm sure they work directly with some of these developers

I agree, there is no way that they hadn't tested this against the most popular apps out there.  If they didn't test it, they couldn't be sure that there wasn't a major bug waiting for them affecting all phones.  They wouldn't want that...plus they probably don't have to worry too much about major developers leaking stuff from the meta-data collected because that would mean they have to actively track things such as features.

 

New features would be under new API's which the app itself wouldn't have, so the most that devs would probably get is info regarding some of the internals.  The other thing to note, some of the developers do get early access to the API layers in order to have apps that use some of the new features at launch.  So yea, I agree it's very unlikely that they didn't do testing on this.  What is a bit concerning though is the apparent amount of people who had an issue, as that to me seems like they are missing something critical in the way they are testing.

[WhitetailAni]

1 hour ago, BuckGup said:

This has happened on every iPhone since the X came out I believe. It doesn't seem to be widespread but this has happened before 

That's due to faulty OIS, which usually is rare and easily fixed under warranty. For this to affect all devices implies there's a fault in the design.

 

Apple:
It's not something you can fix in software. You couldn't fix antennagate in software, you can't fix cameragate in software.

[Sauron]

2 hours ago, wanderingfool2 said:

What is a bit concerning though is the apparent amount of people who had an issue, as that to me seems like they are missing something critical in the way they are testing.

If they did test software that opens up the possibility of something much worse than a bug: a manufacturing defect. If it's the latter the recall will be painful.

[wanderingfool2]

1 hour ago, Sauron said:

If they did test software that opens up the possibility of something much worse than a bug: a manufacturing defect. If it's the latter the recall will be painful.

My thought would be that there is an edge case that they didn't consider in testing...e.g. if the API is being used in a specific way and you have a specific setting or something like that.  It would explain how it's wide-spread but not necessarily spotted in testing (mixed in with maybe a slight hardware fault).

 

Who knows though, hopefully the hackers pick apart the update to figure out what Apple changed.