Samsung Galaxy S8 comes close to the best in DxOMark Mobile testing

OK, Not a fan of ratings, but I actually do use HTC 10 and can tell you that the camera on that phone is non plus ultra. Overall, it beats, or at least it did in 2016, all other phone cameras out there, especially if you use Pro mode and tweak the DNG files a bit.
Hopefully U11 is even better in this department.
Samsung and Apple beat the PR drum so hard that the general public starts to believe theta there's nothing better out there. Mistake...

yep, dxo has always been sony's kid..

I wouldn't be surprised if lense assembly comes from same facility, too. More or less there are changes in image processing, that's all, imo.

It is an over-aggressive jpeg engine, certainly. Given the various work-arounds for addressing the shortcomings, it makes the omission of the dual camera 7+ all the more mystifying. But look into that list of figures linked and it throws into sharp relief the term 'marginal' when you observe those for ANY current large sensor camera against ANY smartphone.

The PDR figures for the iPhone 7 are of course BETTER than those of the S7 by virtue of engaging very low ISO. It's swings and roundabouts with these things. It doesn't disturb me in the slightest to think there may be better camera phones than the 7+. It's not leagues of difference.

The figures combined with the aperture difference, account for a total difference that you find between FF and APS-C of the same generation/efficiency. For many that is a much more than marginal difference when talking about larger sensor cameras (hence why FF exists still), but for light deprived small sensor cameras, that's arguably an even greater " loss" . And a lower noise floor at extended (lower) ISO's is only helpful in bright daylight or long exposures, similar to exposing any of the other cameras half a stop " to the right" . That's not really a benefit for the iPhone module either.

"similar to exposing any of the other cameras half a stop " to the right" That's not absolutely true is it? It may give a benefit in terms of the noise floor in larger format cameras, (which is hardly necessary) but the only one I can think of that doesn't actually lose out in terms of DR when set under ISO 100 is the Nikon D810. In daylight it isn't all that impractical to have ultra low ISO in tiny sensors and given that they suffer from noise issues even at base sensitivities, it is actually useful. The advantage may slip away in lower light conditions, but that doesn't mean it's not worth having.

You seem to be missing the point I'm making that there's a huge gulf in performance between cameraphones and large sensor cameras. In comparison [as you might expect], the differences between different cameraphones are small. With ILCs I assume the tests are based on similar lens useage to give a level baseline on sensor performance.

@grasscatcher
If you want premium audio quality from your B&W's [or whatever], there's always the option of that [included] 3.5>lightning adapter. The horror, the horror!

I was using wireless [mainly] for quite a while before getting an iPhone 7. There are some issues and features where other smartphones have the edge, but I just don't get that this is one of them.

Maybe if there wasn't a 5 second delay from pressing the shutter to the camera taking the shot it would be useful
. With phones you tend to get the shot the camera wants not what you asked for

Sorry, but there is basically no delay on my S7 when pressing the shutterelease. Of course if the flash is on there might be 0.5sec or so delay, otherwise none

I just got a BlackBerry KeyOne! Couldn't care less about the camera though :)

"There are reasons for upgrading from the S6"
Yes...but I'm in no hurry.

yes, these reasons are called "clever marketing"

jesus, when I hear that I could cry!
"You have to upgrade because there is better" OMFG!
do you need all those features you mentioned?
like memory? what crazy amount of data do you need to take with you all the time on your 4G data flat super computer?
Chuck Berries whole collection?
Your portfolio of indoor shots of your stamp collection in various lightning?
And water resistance? Really? To be secure against the potential water damage? My first 2 smartphones were water resistant. Didn't need it much (and I cycle for 1 h no matter what weather)

Seriously. If a device is not broken or you need a function, don't upgrade. GAS is real and it is not sustainable.

Try decaf kreislauf. There's a large curve between no phone, and Deep Thought in my pocket. Options are good.

No problem...I basically agree.

I can relate and agree. I reluctantly upgraded to the Note 7 and my gOD it was a beast. Awsome except for curved screen and sealed battery and glass back. Screen broke then they took it off me...
The canera was stellar though

@Gmon750 Unless of course there's some change in the paradigm. Currently that's being addressed by multiple capture and merge [which can apply to cameras too, but does have its limitations [mostly with anything moving!]

The other approach is multiple simultaneous capture via sensor arrays, giving increased resolution, lower noise, better DR and zoom range, as well as depth effects - although not all at the same time.

The digital ILC is now quite mature technology and while there are significant developments in all sorts of areas, IQ is arguably increasing only by very small increments. Multiple array sensors will need massive computational power, but the decrease in size/increase in power is not stalling, so it's likely we will see more of this. Without physics getting in the way. At the moment the gulf is massive though.

In the end the gap is a physical one, because any advances on phone sensors can be applied on ilc sensors. ILC sensors are larger. The improvements of phone sensors are huge but mainly because they were very very bad.

@Dante Birchen
You could do [and indeed it is done for the gigapixel images] multiple arrays with ILCs, but they're hardly practical for daily use. Multiple array cameras using many small sensors and lenses could be cheaper as well as much slimmer. I have a variety of cameras up to and including 35mm format, with a huge chunk of money invested in systems and lenses. It doesn't give me any pleasure to contemplate the obsolescence of my gear [it's not going to be for a while], but I do think it's coming.

In terms of the physics there's no absolute reason not to calculate for the total area of multiple sensors for the impact on IQ.

What size monitor did you do the comparison on? Was it calibrated? What codec was used to capture the GH4 footage? Was the GH4 footage straight out of the camera or was it graded? Was it recorded externally or internally? If you can't answer those questions, your claim is worthless.