iOS 14.5 Beta 7 ― Bug Fixes, Changes, and Improvements

[Shirasaki]

When you install the app in MacOS Big Sur it shows you how to give it the permission it needs.

I mean iOS. The one run on iPhone iPad etc. macos at least can “hack” somewhat, but no such luck for iOS devices unless I do jailbreak.

 

[heiskanenej]

If this ‘feature’ is so important to you, I’d recommend you switch to a plethora of android phones that I’m sure you know have said feature and if you want to stick to an iPhone, as I said, there will be a ton of third party apps on the App Store.

I have never seen that feature on Android phones. Or maybe I have not tried because that’s not what normal user needs. At least my OnePlus 8T with LineageOS doesn’t have that feature.

 

[ManuCH]

Someone is getting nit-pick and I fail to understand your motivation.
Some of mine I use a tripod, others I do not. Mostly I do not. The issue exists for each. I have posted both.
It originally came to light a few beta’s ago in 14.5. First observed doing low light photos. Suspected cause is a change to the post pic processing in stock iOS.

It has nothing to do with pic post processing in the stock iOS app. Simply put, sometimes the stock app doesn't actually use the 2.5x lens when you switch to 2.5x zoom. The logic behind it is that if the camera app believes the lighting is too low for the 2.5x lens to take a reliable picture (without motion blur etc), it uses the 1x lens. For the average user it usually produces a better result.

If you want to force the iPhone to use the 2.5x lens, you need to use a third party app. It has always been like that. It could be that iOS 14.5 tweaked the logic a bit and uses the 2.5x lens more conservatively as there may have been user complaints about blurred pictures in the dark with the 2.5x lens.

So no, nothing to do with pic post processing. It's just the logic deciding which lens to use that may have changed a bit.

 

[gwhizkids]

Not a tripod. Both hand helded but very stable.
Not portrait mode. Just normal 2,5x
The focus was on the tree in both

Thanks. So I would say in reply:

-You took these in low light and handheld. I don’t know what the exposure time was. But I’d say there’s a good likelihood that there was some movement, however imperceptible, that affected the Apple photo.
-ok
-if the focus was on the tree and not the car, it’s quite possible the car license tag was out of focus. When you’re shooting at low light levels, you are shooting at very low f stops, meaning the aperture is wide open. When you do that, the depth of field is very shallow, meaning you might not have had the tag in focus.


If I had to guess, I’d say that imperceptible motion in the one photo is your most likely culprit. The detail on the license tag appears to be motion blur.

If you want to really convince me on this, you need to do this comparison under controlled circumstances, I.e., on a tripod, with the exact same settings for exposure, aperture and focus. And then include the EXIF data in your post. Otherwise, it is impossible to say if the differences are user derived and not attributable to the software.

In any event the differences are likely imperceptible to anyone who does not magnify the photo to a large extent.

 

[gwhizkids]

Someone is getting nit-pick and I fail to understand your motivation.
Some of mine I use a tripod, others I do not. Mostly I do not. The issue exists for each. I have posted both.
It originally came to light a few beta’s ago in 14.5. First observed doing low light photos. Suspected cause is a change to the post pic processing in stock iOS.

I’m getting nit picky because (1) you and the other guy are using extreme language to describe a nearby imperceptible difference in the photos (“watercolors”) and (2) because you are doing so (at least he or she is) without controlled conditions. Shooting handheld in low light is virtually guaranteed to generate motion blur. But sometimes you get lucky. Using a tripod eliminates that source of possible image “contamination”.

Bottom line: I do not like blanket assertions made without adequate experimental evidence.

 

[ManuCH]

I’m getting nit picky because (1) you and the other guy are using extreme language to describe a nearby imperceptible difference in the photos (“watercolors”) and (2) because you are doing so (at least he or she is) without controlled conditions. Shooting handheld in low light is virtually guaranteed to generate motion blur. But sometimes you get lucky. Using a tripod eliminates that source of possible image “contamination”.

Bottom line: I do not like blanket assertions made without adequate experimental evidence.

I'm pretty sure it's not motion blur. It was the iOS camera app using the 1x lens instead of the 2.5x lens and applying digital zoom. Which can be easily verified in the EXIF data.

 

[gwhizkids]

I'm pretty sure it's not motion blur. It was the iOS camera app using the 1x lens instead of the 2.5x lens and applying digital zoom. Which can be easily verified in the EXIF data.

That makes sense too. Interpolation (what digital “zoom” is - basically guessing) will cause that effect also.

All of this goes to show that the Apple apps are good for basics but if you want better results, go with a product like Halide.

 

[Populus]

That makes sense too. Interpolation (what digital “zoom” is - basically guessing) will cause that effect also.

All of this goes to show that the Apple apps are good for basics but if you want better results, go with a product like Halide.

I wanted to share a thought, now that you mention that digital zoom is basically an interpolation. Based on that idea, couldn’t Apple improve this digital zoom using AI thanks to the 16 cores of the A14 Neural Engine? Couldn’t we see better digital zoom in future iOS releases, especially on iPhones with more Neural Engine cores, like those with A12-A13 SoC and even more on A14 devices?

 

[ManuCH]

I wanted to share a thought, now that you mention that digital zoom is basically an interpolation. Based on that idea, couldn’t Apple improve this digital zoom using AI thanks to the 16 cores of the A14 Neural Engine? Couldn’t we see better digital zoom in future iOS releases, especially on iPhones with more Neural Engine cores, like those with A12-A13 SoC and even more on A14 devices?

They do indeed improve digital zoom with AI. A pure interpolation without AI would look much worse.

 

[dk001]

It has nothing to do with pic post processing in the stock iOS app. Simply put, sometimes the stock app doesn't actually use the 2.5x lens when you switch to 2.5x zoom. The logic behind it is that if the camera app believes the lighting is too low for the 2.5x lens to take a reliable picture (without motion blur etc), it uses the 1x lens. For the average user it usually produces a better result.

If you want to force the iPhone to use the 2.5x lens, you need to use a third party app. It has always been like that. It could be that iOS 14.5 tweaked the logic a bit and uses the 2.5x lens more conservatively as there may have been user complaints about blurred pictures in the dark with the 2.5x lens.

So no, nothing to do with pic post processing. It's just the logic deciding which lens to use that may have changed a bit.

So when 14.5 b1 came out this issue did not exist for normal to bright light. Somewhere around 4 it became an issue. I would have to go back to see when I caught it. You can force the use of the 2.5 on Camera and the the issue still exists. The default to the primary lens vs the 2.5x lens was supposed to be for low light photos however the last few betas this is now occuring for even well lit pics.
Whether I use the primary or 2.5 lens, I still see the same issue which is why I feel it is more than just lens selection.

 

[dk001]

I’m getting nit picky because (1) you and the other guy are using extreme language to describe a nearby imperceptible difference in the photos (“watercolors”) and (2) because you are doing so (at least he or she is) without controlled conditions. Shooting handheld in low light is virtually guaranteed to generate motion blur. But sometimes you get lucky. Using a tripod eliminates that source of possible image “contamination”.

Bottom line: I do not like blanket assertions made without adequate experimental evidence.

I do a lot of shots and this really shows in post processing. The original 1.6/2.5 lens selection was only low light and ProRAW really helped there. This issue has become worse over the last few betas.

I have done shots with and without a tripod, same item, same lighting to showcase the difference.

End of the day I had hoped to get more use from Camera but am content to stick with Halide to get the quiality I am looking for. I just hope that Apple fixes the current problems and allows Camera to properly utilize all the lenses.

That’s why I file the feedback.

 

[ManuCH]

So when 14.5 b1 came out this issue did not exist for normal to bright light. Somewhere around 4 it became an issue. I would have to go back to see when I caught it. You can force the use of the 2.5 on Camera and the the issue still exists. The default to the primary lens vs the 2.5x lens was supposed to be for low light photos however the last few betas this is now occuring for even well lit pics.
Whether I use the primary or 2.5 lens, I still see the same issue which is why I feel it is more than just lens selection.

If you see it happening even for well lit photos, you should look at the EXIF data. There it says what lens it has been using. If it still used the 1x lens, then we know it's a bug in lens selection (or it's intentional to be more conservative to avoid motion blur). So yes, it was for sure good that you submitted the feedback! And in the meantime, use a third party app if you want to be sure that the iPhone is really using the lens you want to use.

 

[buklauu]

So when 14.5 b1 came out this issue did not exist for normal to bright light. Somewhere around 4 it became an issue. I would have to go back to see when I caught it. You can force the use of the 2.5 on Camera and the the issue still exists. The default to the primary lens vs the 2.5x lens was supposed to be for low light photos however the last few betas this is now occuring for even well lit pics.
Whether I use the primary or 2.5 lens, I still see the same issue which is why I feel it is more than just lens selection.

ManuCH is right here, the iPhone has a very different opinion of what 'well lit' means, and it uses the main camera and digitally zooms in because it believes that'll be a better picture. It's not a bug per se, just Apple being aggressive in their usual philosophy of 'knowing what's best' for the consumer. The main camera apertures for all the iPhone 12s got widened to f/1.6, which should allow for 27% more light, and the software seems to be biased towards it because of this. So third party apps are your best bet to force what you want.

Even with the right camera, iPhones' computational photography tends to smudge out finer details in an attempt to reduce noise. This is noticeable when you zoom in, but likely looks 'cleaner' if you're just scrolling through social media (which is the primary audience Apple is adjusting for).

 

[dk001]

ManuCH is right here, the iPhone has a very different opinion of what 'well lit' means, and it uses the main camera and digitally zooms in because it believes that'll be a better picture. It's not a bug per se, just Apple being aggressive in their usual philosophy of 'knowing what's best' for the consumer. The main camera apertures for all the iPhone 12s got widened to f/1.6, which should allow for 27% more light, and the software seems to be biased towards it because of this. So third party apps are your best bet to force what you want.

Even with the right camera, iPhones' computational photography tends to smudge out finer details in an attempt to reduce noise. This is noticeable when you zoom in, but likely looks 'cleaner' if you're just scrolling through social media (which is the primary audience Apple is adjusting for).

After doing a bit of playing around I finally convinced it to use the 1, 2.5, then Halide 2.5 in order.
You can see the difference between the 1 and 2.5. This was using AF on a tripod from a range of 6ft.
Point of focus was the trellis. Standard op for the Camera was to digitally zoom. After covering the 1.6 lens, killing the app then relaunching, after a couple of attempts it used the 2.5 zoom lens. All of the attached photos are as is - nothing was done post snap.


Even in bright light it was very hit / miss for the 2.5 to engage instead of digital zoom.
Something is not right with the Camera app lens selection. In ProRAW I could not get the Camera app to select the 2.5. Pita.

btw - I don’t see how these are designed with SM as the primary audience. You could go much poorer and lose nothing in that venue. Then again …. for now I’ll stick with Halide and report changes to Camera app quality as they occur.

 

[TheIntruder]

If you're shooting in the dark it's dark if the flashlight is not on. You can see in the dark? If I can't use the iPhone flashlight to get around in the dark I have to hold 2 things, a flashlight and the iPhone.

Why are people trying to complicate such a simple feature. The light is already on the phone. I would like the option to use it like a flashlight while taking a picture.

Save yourself the agita and use a different app. There are lots of good choices available, and they bring other advantages as well. Once I adopted a third-party camera app, I haven't looked back.

I just encountered your use case just yesterday, while trying to take a manual focus macro shot of the tiny, poorly contrasted text on an item. Lighting the subject first, instead of relying on a flash pulse, allowed a properly focused shot to be taken without having to take multiples with the flash and then pick the best. This was in a room lit with sunlight; wasn't even in the dark.

That's something most users won't encounter, so it's not likely to be a capability that the stock app will include.

 

[gank41]

Save yourself the agita and use a different app. There are lots of good choices available, and they bring other advantages as well. Once I adopted a third-party camera app, I haven't looked back.

I just encountered your use case just yesterday, while trying to take a manual focus macro shot of the tiny, poorly contrasted text on an item. Lighting the subject first, instead of relying on a flash pulse, allowed a properly focused shot to be taken without having to take multiples with the flash and then pick the best. This was in a room lit with sunlight; wasn't even in the dark.

That's something most users won't encounter, so it's not likely to be a capability that the stock app will include.

My problem has been a good camera app that does the SAME as the stick camera app: Photo, panoramic, square, and then also video with the same Slow-mo and time-lapse options. Is there a good app that does all that, paid or not?

 

[gank41]

Also, my S6 Apple Watch is at 50% right now after putting it on at around 9AM CST. So that’s awesome. My iPhones at 30%, which is still fine considering how much I’ve been staring at this thing all day. Must have been more indexing and stuff, more than normal. I still plan on wiping both devices when 14.5 is officially released.

::Edit:: I was able to go from 9AM CST to 11:30PM CST and put it on the charger for bed at 36%.

 

[s2mike]

Don’t want to jinx it, but so far all of my iMessage links have showed up properly…

Alright, at this point I’m gonna pronounce that this bug has been fixed.

Haven’t had a single link show up incorrectly on Beta 7.

 

[dk001]

12 Pro Max
Phone
Since beta 7 the Setting "Call Blocking & Identification" is being disabled every couple of days or so. This morning was now the fourth instance of this feature being disabled. Not turned off. It literally is not in settings. has been working fine through all 14.5 betas till now.

I use Hiya and have to go back through the app to turn this feature on.

Feedback reported.

 

[zkap]

iOS 14 is a cluster****. This is the first time that I'm really waiting for the next iOS before the Summer starts. It's also the first time I regret upgrading to an iOS version.

 

[gank41]

iOS 14 is a cluster****. This is the first time that I'm really waiting for the next iOS before the Summer starts. It's also the first time I regret upgrading to an iOS version.

And to me, the entirety of iOS 13 was a dumpster fire and the first beta of iOS 14 was amazing. I guess different experiences for everyone it seems.

 

[dk001]

Here is a new one ..... not really sure if it is beta 7 or not. This happened 2x today.
I am using a 12 ProMax on beta 7.

Doing a phone call (LTE) with another person - I was the recipient, not the originator. Never had this before that I am aware of.
First caller was using a Pixel (in New Jersey), second was using a Galaxy (in California).
During the call, the call would suddenly disconnect. On my side the call just ended - black screen. After I called them immediately back, the other caller (both times) had a message that read "Disconnected due to limited internet connection". Neither of us have seen this before. I know both parties well.

No Feedback reported yet ....

 

[macintoshmac]

Alright, at this point I’m gonna pronounce that this bug has been fixed.

Haven’t had a single link show up incorrectly on Beta 7.

Finally! Did Apple respond to any of your feedback messages saying they made changes that might have fixed it?

 

[PeterK44]

So when 14.5 b1 came out this issue did not exist for normal to bright light. Somewhere around 4 it became an issue. I would have to go back to see when I caught it. You can force the use of the 2.5 on Camera and the the issue still exists. The default to the primary lens vs the 2.5x lens was supposed to be for low light photos however the last few betas this is now occuring for even well lit pics.
Whether I use the primary or 2.5 lens, I still see the same issue which is why I feel it is more than just lens selection.

I believe this problem began before the 14.5 betas. I’m using iOS 14.4.2 on my iPhone 12 Pro Max. I selected 2.5x lens with either stock Camera app or ProCam 8 and took photos of the same object indoors with incandescent lighting. ProCam used f/2.2 lens but Camera used f/1.6 lens.

 

[s2mike]

Finally! Did Apple respond to any of your feedback messages saying they made changes that might have fixed it?

Haha of course not.