Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

iBug2

macrumors 601
Original poster
Jun 12, 2005
4,540
863
I got my 14 Pro and compared the camera outputs to 13 Pro and Nikon Z9 (46MP)

As expected there's hardly any difference between the 48MP ProRaw output of 14 Pro and 12MP ProRaw output of 13 Pro. But there's noticeable difference between them and the Nikon Z9.

The first one is 13 Pro, then 14 Pro and then Nikon Z9. Obviously 13 Pro image is smaller so I scaled up the 13 Pro image by 200% to match the others.
iPhone 13 Pro.jpg


iPhone 14 Pro.jpg

Nikon Z9.jpg



So I think it's pretty obvious that at this sensor size, high megapixels hardly make any difference, as expected I might add.
 

Attachments

  • iPhone 13 Pro.jpg
    iPhone 13 Pro.jpg
    171.1 KB · Views: 81
  • Like
Reactions: servenvolley
I don’t think this is a large of sample size to make any determination. It’s one picture of a sign. You would have to take photos in low light, photos where there are vast differences in lighting, photos with different textures and colors. There are so many factors that can effect how the picture comes out. Especially when the phone is depending on AI to analyze the picture so the content of that picture is really important.
 
I don’t think this is a large of sample size to make any determination. It’s one picture of a sign. You would have to take photos in low light, photos where there are vast differences in lighting, photos with different textures and colors. There are so many factors that can effect how the picture comes out. Especially when the phone is depending on AI to analyze the picture so the content of that picture is really important.
That's why I chose a text. It's the easiest thing for the A.I. to sharpen up. Obviously it's not a good enough sample size, I can try other scenes, but the result will be quite similar. The most important factor is always the sensor size and the size of the glass in front of the sensor. You just cannot get enough detail with a small sensor, even if it's 200MP.
 
There's going to be a large advantage at base ISO, but the gap closes at higher ISO.
I'd say image quality is approaching APS-C for ISO 1000 according to my tests.
Also I'm not manually setting ISO o the phone obviously, I simply go to a darker area such that the camera starts to use higher ISO, or I close down the aperture.
 
There's going to be a large advantage at base ISO, but the gap closes at higher ISO.
I'd say image quality is approaching APS-C for ISO 1000 according to my tests.
Also I'm not manually setting ISO o the phone obviously, I simply go to a darker area such that the camera starts to use higher ISO, or I close down the aperture.
The Nikon photo is at ISO 560. I didn't use base ISO. 13 Pro ISO 160, 14 Pro ISO 200.
 
The difference between 14 pro (left) at 48mp and 13 pro (right) seems obvious to me in real world photos. Sure your Z9 would be a lot better but the 14 pro is a significant leap from the 13 pro.
n9lr37h46zp91.jpg
 
The difference between 14 pro (left) at 48mp and 13 pro (right) seems obvious to me in real world photos. Sure your Z9 would be a lot better but the 14 pro is a significant leap from the 13 pro.
n9lr37h46zp91.jpg
In the right photo it's almost like 13 Pro could not focus. This is a significant difference. I'll test with a similar scene later.
 
14 Pro versus Sony A1
Screen Shot 2022-09-26 at 10.31.00 AM.png

Not as much detail but not far behind.
which is why I consider this performance close to APS-C cameras
 
  • Like
Reactions: ReliableSource
A1 is significantly better, and you can easily apply some NR to the A1 here. I think it's quite similar to my test.
Well I kind of like a little noise, it's just that I couldn't disable NR on the iPhone because it's already baked into the RAW.
I'm only looking at the amount of detail that can be distinguished so noise doesn't really matter.

Sure it's not beating out FF, it better not be, I just think it's way closer than any compact camera, maybe except for a GRIII.
 
Well I kind of like a little noise, it's just that I couldn't disable NR on the iPhone because it's already baked into the RAW.
I'm only looking at the amount of detail that can be distinguished so noise doesn't really matter.

Sure it's not beating out FF, it better not be, I just think it's way closer than any compact camera, maybe except for a GRIII.
It seems like 14 Pro can "fake" the detail at the eyes, but if you look at the notepad, A1 has tons of info on what's on it, and it's all mush with 14 Pro.
 
I don’t think this is a large of sample size to make any determination. It’s one picture of a sign. You would have to take photos in low light, photos where there are vast differences in lighting, photos with different textures and colors. There are so many factors that can effect how the picture comes out. Especially when the phone is depending on AI to analyze the picture so the content of that picture is really important.
A good test would be some sort of outdoor image that has grass in it. Grass looks "mushy" when zooming in at all on a ProRAW image on a 13 Pro.
 
  • Like
Reactions: russell_314
A good test would be some sort of outdoor image that has grass in it. Grass looks "mushy" when zooming in at all on a ProRAW image on a 13 Pro.
I think just different tests and different scenarios if we’re just comparing how the camera performs. I see so many of these posts with one or two pictures and you’re supposed to determine what is better. I’ve seen phone reviews where they show pictures taken by two different phones and what’s interesting is sometimes the pictures look better taken by one phone but other times it’s the other one that looks better.

I honestly don’t think the 48 megapixel isn’t going to help in most situations. I think it’s going to be some edge cases where it allows the phone to do a little bit more AI processing.
 
I think just different tests and different scenarios if we’re just comparing how the camera performs. I see so many of these posts with one or two pictures and you’re supposed to determine what is better. I’ve seen phone reviews where they show pictures taken by two different phones and what’s interesting is sometimes the pictures look better taken by one phone but other times it’s the other one that looks better.

I honestly don’t think the 48 megapixel isn’t going to help in most situations. I think it’s going to be some edge cases where it allows the phone to do a little bit more AI processing.
Since the pixels are incredibly small, the best results probably will be with a tripod. But I never ever use my iPhone on a tripod. And if the pixels are getting mushy even with a small handshake, then the advantage of 48MP disappears really fast.
 
  • Like
Reactions: russell_314
Since the pixels are incredibly small, the best results probably will be with a tripod. But I never ever use my iPhone on a tripod. And if the pixels are getting mushy even with a small handshake, then the advantage of 48MP disappears really fast.
From my understanding, the advantage isn’t with taking 48 megapixel photos but having four times the data for the AI to process. I don’t know how this translates into real life.

Unless you’re just a total camera nerd or just love yearly upgrades, I wouldn’t advise anyone to upgrade from the 13 Pro. There’s always incremental upgrades, but if you upgrade every 2 to 3 years, then you’re more likely to notice them.
 
From my understanding, the advantage isn’t with taking 48 megapixel photos but having four times the data for the AI to process. I don’t know how this translates into real life.

Unless you’re just a total camera nerd or just love yearly upgrades, I wouldn’t advise anyone to upgrade from the 13 Pro. There’s always incremental upgrades, but if you upgrade every 2 to 3 years, then you’re more likely to notice them.
That's indeed the case. One generation of camera improvements are miniscule these days. I only upgrade every year because I can sell the old one for a really good price. And now that I have my 14 Pro, it's basically the same phone as my 13 Pro. :)

And the A.I. can only work wth 4 times as much data if there is 4 times as much data. That's the issue with megapixels. You just know the count of them, but not the percentage of "correct" ones. If in the 48 MP image, half the pixels are incorrect, then the A.I. is working with more noise, not data.
 
That's indeed the case. One generation of camera improvements are miniscule these days. I only upgrade every year because I can sell the old one for a really good price. And now that I have my 14 Pro, it's basically the same phone as my 13 Pro. :)

And the A.I. can only work wth 4 times as much data if there is 4 times as much data. That's the issue with megapixels. You just know the count of them, but not the percentage of "correct" ones. If in the 48 MP image, half the pixels are incorrect, then the A.I. is working with more noise, not data.
True, but in that, same scenario with only 12 megapixels, if half of them are bad data that means it only has 6 megapixels to work with. In that same scenario with a 48 megapixel sensor it would have 24 megapixels. Of course it’s a lot more data to work with so the processor had to be capable of sorting out what is good data and what is bad data.

Of course I’m sure this is extremely oversimplifying it. It’s possible that Apple went with the 48 megapixel sensor, knowing that it wasn’t going to be better but they’re under marketing pressure. Most consumers will see 12 and 48 and think well 48 is 4×12, so it’s four times better. Most android phones have a ridiculous megapixel count just so they can put it in the spec sheet. Apple really tries to avoid the spec wars but with cameras it’s hard to.
 
In the right photo it's almost like 13 Pro could not focus. This is a significant difference. I'll test with a similar scene later.
I’m sure the 13 Pro did focus, it’s just that you can only zoom into a 12mp image so far before it gets mushy, the 14 Pro captures an image 4 x the size so can zoom in more before it starts to get mushy, so the 48mp image retains more detail at the same crop as the 12mp image.
 
I’m sure the 13 Pro did focus, it’s just that you can only zoom into a 12mp image so far before it gets mushy, the 14 Pro captures an image 4 x the size so can zoom in more before it starts to get mushy, so the 48mp image retains more detail at the same crop as the 12mp image.
Obviously, but in my test the difference in detail was pretty minor. I still don't think we'll get this amount of difference in detail in most photos. The sensor size is quite similar to 13 Pro and that's the most important factor.
 
True, but in that, same scenario with only 12 megapixels, if half of them are bad data that means it only has 6 megapixels to work with. In that same scenario with a 48 megapixel sensor it would have 24 megapixels. Of course it’s a lot more data to work with so the processor had to be capable of sorting out what is good data and what is bad data.

Of course I’m sure this is extremely oversimplifying it. It’s possible that Apple went with the 48 megapixel sensor, knowing that it wasn’t going to be better but they’re under marketing pressure. Most consumers will see 12 and 48 and think well 48 is 4×12, so it’s four times better. Most android phones have a ridiculous megapixel count just so they can put it in the spec sheet. Apple really tries to avoid the spec wars but with cameras it’s hard to.
In the case of 12 million pixels, the probability to get an incorrect one is less because the pixels are almost 4 times as large, so more light gathering per pixel = less probability of noise per pixel. When you use the same sensor area to get 48 million pixels, the probability of noise per pixel goes up.

I think Apple is succumbing to the marketing pressure. And in some test cases, there will be more detail for sure.
 
If I am taking ProRaw shots (landscape photography mostly) with no intention of zooming in to pixel peep- mostly viewing them at regular size on a computer screen, and in some cases maybe printing out 4x6 or 5x7 prints- would I truly notice a difference with the 12mp ProRaw images from a 13 Pro vs the 48 mp ProRaw images from a 14 Pro? Or is the noticeable difference truly only when blowing an image up really large, or of course zooming in to pixel peep?
 
If I am taking ProRaw shots (landscape photography mostly) with no intention of zooming in to pixel peep- mostly viewing them at regular size on a computer screen, and in some cases maybe printing out 4x6 or 5x7 prints- would I truly notice a difference with the 12mp ProRaw images from a 13 Pro vs the 48 mp ProRaw images from a 14 Pro? Or is the noticeable difference truly only when blowing an image up really large, or of course zooming in to pixel peep?
In general if there's enough difference in detail at 1:1, they should be also be visible when you view them at smaller screens. If the difference is minor, they won't be visible at all when you scale them down.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.