The Smartphone vs. the Camera Industry

There is zero reason a DSLR camera is needed for this setup. Ten years ago it was the best option, but for our application we really don't need it. Sadly, the software we use is 10 years old, usupported and half written in German. Calibrating was a fucking nightmare, more than half the variables in the configuration file were German and the machine support people stopped supporti.g it years ago and don't have any staff familiar with it any more. Anyway tangent aside, a high quality cell phone cameram that's easy to interface with and doesn't have the problems associated with a ten year old DSLR camera would be so much of an improvement.

I also learned because of all of this, cabin cameras have a finite internal memory because of the way pictures are stored internally on the camera's storage and the only way to get more than 10,000 pictures, even if the pictures aren't being stored on the camera is it insertt an SD card which will work once to reset the camera's interal counter which should give you another 10,000 pictures.

One is the smartphone user who desires good quality for whatever the situation, who just wants a photo uploaded to Instagram or whatever, NOW. They don't want to learn how to adjust exposure compensation to expose-to-the-right for optimal raw image quality.

This user wants lots of R&D spent on computational photography, lots of processing power on the phone, and a simplified UI that just gets out of the way.

The other is the enthusiast photographer who shoots for either pleasure or maximum quality.

This user wants all the control, all of the time: they've invested time and effort into nailing exposure, mastering autofocus, all while composing beautiful shots. A smartphone interface is not what you want: the best images are often captured in difficult environments where a touchscreen just gets in the way and you must have hard control points.

As far as processing goes, they don't really need any of it; they'll often just process in post. Just gimme the raw data and that's all I need.

Can any manufacturer manage both? Much less in the same camera? For a competitive price?

I'm not sure.

My process was take a bunch of pix, wait till I got back to my hotel room, plug the camera's memory card into my laptop, transfer the pictures, open up email, attach pictures, email them.

This has been tried and it didn't sell well, see the samsung galaxy cameras, and the nikon coolpix s800c. Also smartphone cameras have gotten really good, and in practice a fully fledged cell phone with a good enough camera is not that much more

I feel Sony are the only company to be looking far enough forward, catering well to the overall industry direction. But even the very best very smallest camera system doesn't fit in your pocket.

I have a bulky DSLR and a mirrorless that I use for bird photography. Most of the time, when not doing that specific task, the camera phone will be good enough if I didn't want to bring my other cameras. And mirrorless cameras are tiny! I can fit a professional quality camera system in a satchel no bigger than my cupped hands, yet the convenience of just keeping my phone on me is impossible to beat.

Now somebody should simply come up with the idea of a camera that is also a phone. I mean you did but somebody should make that idea a reality. Hey, my camera is ringing ...

It’s a good camera but it’s frustrating in some ways. No GPS. No easy way to transfer pics to my phone. I probably care less about in camera processing but one has to believe Canon and Nikon could do far more interesting things in camera than they do given the hardware resources and glass they have to work with.

And yeah. Even if I have my DSLR with me I find myself shooting pics on my phone so I can share them now and not in a month when I get around to processing the ones in my camera.

This, and as the old adage goes: "The best cameras is the one you have with you". I, for one, have taken technically superior photos with my camera - unsurprisingly - but certainly the more interesting photos with my smartphone. Now you can argue what is better, but I'd say interesting beats technical superior any day.

> Sony and Fuji are really the only two companies that really get the current camera market IMO.

So much this, and if you compare the two: Sony is so much ahead of Fuji that they play in a different league in my opinion. As much as I hate Sony, I think they nailed it. In some sense I'm tempted to add Leica, because I think the Q and Q2 show that they get the current market too, but considering the price, the price/performance ratio, comparing that to Sony and then comparing it to my smartphone ... just nay.

Another interesting data point is that younger artists, like Eric Kim, don't have a problem to add smartphone photos to their portfolio.

What breaks away is basically the customer base focused on documentary photography, not the artists. They may even gain more artists that get hooked through smartphone photography and convert into enthusiasts.

The documentary focused customers could be won back with 3D technology and virtual tourism in VR, "re-experience your photographed/scanned holiday shots".

Not everything can fit into smartphones, and I'm suprised that the camera industry isn't taking advantage of that.

https://www.playmemoriescameraapps.com/portal/

They are almost universally features you can get on your phone cheaper or even free with much better UX. Maybe at the resolution or raw area or some other dimension they might be better. You can also tether to your phone but why would you? It's just too tedious.

I don't think anything is going to save cameras. I keep one which I pretty much only use for super telephoto (400-800mm) and for ultra wide/fisheye. If a phone optically offered both of those things built in I'd stop using that camera completely.

Note: A snap-on 3rd party lens system would not get me to switch but a built in mount might (if Apple or Google or Samsung, ideally Apple, designed some kind of mount for lens I'd totally by them). In fact I'm a little surprised Apple hasn't already done this. I can imagine 2 or 3 tiny notches or mag-divits or something in the back or edge of the case near the lens that external lenses attach to. Apple could offer 2 or 4 lenses and maybe a ring light and easily pull in another billion a year as all the instagrammers go buy them all.

I think this is the real answer for the standalone cameras: treat it as an extension of the smartphone.

It's embarassing that in 2019 it's _still_ not possible to wirelessly transfer RAW files from a camera to a phone. I haven't seen a single camera that can do this. iOS is perfectly ready for it too: there's platform-level support for RAW images and processing and plenty of apps take advantage of it, including the built-in Photos app.

Related: does anyone know why cameras typically have horrendously slow wireless transfer speeds? My iPhone can easily reach 600Mbps over WiFi yet even third-party devices made solely for the purpose of quickly transferring photos can only manage ~60Mbps.

I'll take that one step further (and with a family member in professional photography it's a painful conclusion): the convergence between photography and mobile communications devices will make the photographic camera industry disappear because the numbers won't be there to make the devices economical enough. It's basically a continuum between 'doesn't work at all' to 'works for almost all cases'.

System cameras will at some point be as rare as hole through CPUs.

I shot Sony since the NEX series came out, starting with the 5000R. Paired with a fixed Zeiss 24mm this camera took absolutely astounding images. Pair a Sony with a smartphone and you get easy upload to social media. What's missing, exactly?

Incidentally I sold all my Sony gear last year. The UI and handling of these cameras is just as ugly as any smartphone can be. It takes seconds to start. The animated/touch UI is just fancy, but slows any operation down. Most modern cameras have humongous menus which are hard to navigate and make little sense as dozen of features get added, but Sony is really on the worst end when it comes to placement and behavior of all the functionality they pack into their cameras, which is sad.

Their "app" store is the most useless thing I've ever seen, including stuff which should be bundled in the camera but when used precludes the functionality of the camera itself.

I'm lucky I always had a great group of friends from which we exchanged cameras and I was able to use most of the others you mention too. If you pick any Nikon, Canon or Fuji camera and start taking pictures you immediately realize why menus are placed in a certain way, the UI is intentionally minimal, and why the body is shaped the way it is. It's a night and day experience if you know what you're doing. No operation takes more than half a second to perform. It's form-follows-function. These cameras cater to pros _first_. You don't need to upload to social media and/or you basically never want to have smartphone-like post-processing in camera because the normal workflow is to either take the built-in jpeg because it's already good enough, or process the raw yourself. Transferring the image via wifi/ble/nfc is decent enough in most 2017+ mirrorless cameras.

Granted, Sony made _huge_ steps forward since the NEX line, but you can definitely see how the camera is influenced by the smartphone market. Sony even released the QX1, which essentially is a camera _for_ the smartphone.

Somehow I have the impression that smartphones themselves will eat away all the middle market as the small optics keep improving. If you want images for instagram you'll probably never care about the improvement in image due to a larger sensor anyway. But a pro will definitely seek superior handling and behavior in a camera, and that is almost tangential to how a smartphone operates.

https://github.com/ma1co/Sony-PMCA-RE

Computer + X = Computer

It doesn't matter if X is a screen, an audio system, a network terminal, a phone, a GPS locator, a fingerprint reader, a 3D scanner or a camera. What matters is the fact it's a computer, and the standalone thing will either retreat into a specialist niche or die. Even train tickets and credit cards are going the same way.

These computational imaging capabilities are good enough for 99% of people because they're taking pictures on their phone and viewing them on their phone. They are NOT good enough for professional uses.

https://spectre.cam/full-res/Full%20Size%20Comparison%20for%... <--- this is from an 11 pro, and was posted on a related article several days ago. The left side has had contrast tweaked to make it more obvious, but on both you can see that the iPhone 11's computational photography is doing very weird things to the picture that become pretty obvious when you start to look at them closely. You'd never notice on a phone, but on a monitor, or if this were to be printed for display, you notice that they look almost more like photorealistic paintings than they do actual photographs.

This sort of thing would be completely unacceptable on a prosumer+ camera. There's some programs out there that can do similar things on your computer, like Luminar, that are significantly better at it - but are also using significantly more computing power to do it. Doing the same sort of processing in Luminar on a raw out of my a7r 3 takes 30-40 seconds to apply on an i7-7700k running at 5.2Ghz, and the results still sometimes come out looking similar to the iPhone shot.

The tech isn't nearly there for this to be used on standalone cameras. Maybe in 5-10 years, but certainly not today.

>Imagine how many iPhones Apple would sell if the sensor were not much changed from the 6s and the camera software not changed at all.

Honestly, very little of the advances in cell phone cameras since the 6s have come from sensors; it's all computational. We're up against the brick wall that is photon noise.

Finally, the A6600 is very much not a flagship. The A9 II is the flagship, at three times the price ($4500).

how often does Sony release new APS-C sensors?

I've noticed this too and I have a pet theory about why.

It all comes back to the Canon 5DMk2. The Mk2 was an industry first: integrating video into a full frame sensor in an affordable body. It was an enormous hit. An entire generation of filmmakers grew up using the Mk2 and doing shots that were basically impossible on movie making equipment in the same price range. It made entire movies possible.

The problem is this: Canon also makes extremely high margin movie cameras and lenses. This equipment costs a lot more (and is probably what you want for a full production but I digress) and Canon makes a lot more on it. My impression is that the Mk2 killed an entire segment of Canon's product line and scared the shit out of them.

Since then, Canon has been extremely careful to roll out video features on prosumer cameras (like the Mk2) cameras slowly. They will only support lower quality video codecs on these cameras. They have made a new line of cheap-er video-first cameras.

I think that when the camera makers cannibalized their own products, a common move in smart devices, they got incredibly scared and ran away. The only one really making moves in this area is Sony and that's because Sony has nothing to lose - they had no business when they started making their name by releasing cameras with a lot of tech. Even so, they are not very good.

And as is typical one ends up with a half-assed solution that will stop working in 2 years when Apple/Google change some API or the app isn't maintained any more.

No matter how awesome it seems at launch and how good the quality, anything relying on smartphones is basically disposable.

My 2008-era 1Ds3 still crushes a modern phone (I have a Pixel 2) in dynamic range when processed properly, and a D7500 is even better still, not to mention the latest full-frame models like the D850 and such.

Phones do, counterintuitively, have the advantage in extreme low light because they're able to stack multi-second exposures very reliably. But then again, I'm comparing my phone against a 2008 camera... the very best latest full frame cameras might be on par (at an extreme depth of field disadvantage though).

To some extent, I fear that "the eye" of the public has simply forgotten what sort of photographs are possible, and are happy with snapshots taken in perfect conditions, which do look OK on phone cameras.

Remember the old architectural style with tilt-shift lenses, where tall things would imposingly loom over you with perfectly parallel lines? You can remove the "distortion" from tiling your camera up in software, but it tends to look pretty terrible compared to a camera on a tripod that is perfectly level and is just taking the image from the bottom of the image circle.

Remember how it's nighttime half the time, but you never see good pictures taken at night? Stars exist, waves bend into glass, and the moon can glow over your mountains and valleys. But not if you're holding your camera in your hand and have a fingerprint on the lens. So we don't see those pictures anymore.

Even product photos in ads look pretty bad a lot of the time. You can get a white table and take a snapshot with your cell phone, but it looks like you did. Maybe it doesn't matter, though.

Ultimately, the problem the camera industry has is that very few people need to take perfect pictures in tough conditions. Fine art, wildlife, sports... phone cameras just don't yield as good of results as a larger camera. But not many people are taking that kind of picture. They just want to say "hey friends, I went to this place, look at it" or "look at my friend here". And a phone camera is great for that. It's the perfect balance; it takes up 0 space (since you were carrying a phone anyway), and the image quality is good enough. Smaller dedicated cameras (like an RX100) just don't provide much of a leap in quality.

RAW can be done on phones (I use it on my iPhone), but the result seems to be extremely noisy (but at least you get the detail, which generally gets lost in jpegs from the built-in Camera app).

I was shooting Canon for about 12 years and now have a Sony A7rIII because of it's amazing eye-autofocus, and a heap of lenses. With the DSLR I take the picture, download it to the computer and then play with it on Capture One, often using the great smart masking feature to add some punch to people's faces vs background stuff.

Now on the iPhone 11 range I shoot in portrait mode, which now also works with the 26mm lens (previously just the 52mm). Basically the portrait mode uses semantic segmentation to identify people and clothes. I then use iOS 13's picture editing to get the look right on the picture (usually a little too yellow), add vignette, etc. and then use the portrait lighting to brighten people. There is a huge difference in the before/after of the photos, and they look really impressive. No manual masking, no delays.

Ok, so I can notice the funky bokeh effects sometimes, and it's not on par with the Sony + 35mm f1.4 , or even the 35mm f1.8, but the ease of getting impressive shots is amazing if you know how to process them.

For a given field of view your focal length must change for different sensor sizes so effectively you see a shallower depth of field.

While computational imaging is doing amazing things right now, I've yet to be impressed by any sort of portrait mode, they simply do a mediocre job at creating a realistic sharpness falloff and fail far too often at identifying the edges of a subject. Once they get it right however, they could theoretically do some absolutely amazing things by simulating different bokeh and as small of a DoF as you could ever desire.

Now I spend mere seconds processing each raw image, and they look way better than I can get with the default processing on a smartphone.

When you have the benefit of a large sensor, you don't actually need multishot HDR capture; even older cameras have plenty of dynamic range, as long as you expose properly.

Now if only there were a camera with a beefy processor that I could port my software to run on...

[0] https://github.com/CarVac/filmulator-gui

For pro tier cameras, the problem is that no software manipulation can ever read the mind of the photographer and what they want, so you have to have a way to opt out of in-body post processing or there will definitely be times you're fighting the camera to get what you want. On professional shoots or at fast moving events like a wedding, no professional will tolerate their camera having a mind of its own and editing the files in a possibly-unpredictable way.

So if that's the case, why even include the options at all? Professional photographers have the technical skills to get what they want from the camera by turning the normal ISO/shutter/aperture knobs and don't need those headaches. Casual users don't know or care what's possible and just want a photo that "looks good" which usually means in focus and a little extra saturation. "Prosumers" or "techy users" are the hardest to pin down, but that market may be disappearing anyway.

ehh, I have a photo from the iPhone 4 printed out and hanging on the wall, still very happy with some shots I took with it :)

> too much noise when it gets darker

To be fair, many older DSLRs (Canon 600D) have horrendous performance at high ISO too, while smartphones have Night Sight type stitching of multiple exposures, but these require steady hands and non-moving subjects of course.

Creating a device that every person needs to replace after a few years is a very sound business model. The issue is how many years you generally need to replace it. Apple used to be able to convince folks to replace their phone every two years. Now a two year old phone is not bad at all, and many people are moving to a three year cycle or even longer. This is significant, if 50% of folks move from replacing their phones every 2 years to 3, that's a 16% drop in sales.

Apple's real challenge is to keep innovating to convince folks to keep to shorter phone life-spans. However, as Moore's law keeps it's pace towards the graveyard, that will be be harder and harder to do.

[1] Most recently, https://jefftk.com/pictures/2018

I'm much happier with the ergonomics and quality of my phone camera than I ever was with my dedicated cameras, even a dSLR I owned for a while. On top of that, I always have it in my pocket and the response time is insanely short compared to a dedicated camera.

A decade ago I printed a bunch of photos taken on my 2006 Sony Ericsson K800 (one in a line of cell phones to come with a real Xenon flash instead of the LED rubbish we suffered with after) and they looked fine as prints - it's on a bright, high-res computer screen that they look horrid.

How will we look back on today’s artificially blurred backgrounds and impossibly HDR photographs in a decade or two?

Samsung tried building Android-based cameras a few years back, struggled to gain any sort of market share, and got out of the interchangeable lens camera market altogether. It’s a shame too because the cameras got really good reviews.

Nikon's stock is down over 2013 but I'm not seeing the end days reflected in that. AMD was down for ages but computers didn't go out of fashion.

I recon smart phones will kill for-purpose cameras the same year they kill the PC gaming industry.

Beyond prosumer DSLRs and prosumer Mirrorless there is an opinion among professional photogs that these software enhancements are gimmicks and are not pure photography, although I kind of doubt that being anything but a footnote in decisionmaking since if you feel that way you can simply not use that feature.

Not sure exactly what you meant, but that sounds terrible.

On my Nikon DSLR I can adjust just about any setting using physical buttons, very fast, most often without even taking my eye off the viewfinder. The camera UX is optimized for speed and efficiency.

On a phone, I'd be scrolling through touch screen menus to do anything. By the time a shot is taken the subject is long gone (unless it's static scenery).

The Fuji X100F is about 3 times as heavy then a usual smartphone. So carrying it around would quadruple the weight of my "camera equipment".

I wish somebody would run around with a pocket camera and a phone for a day and then write a blog posts with comparison shots.

I disagree here entirely. The DSLR "app" market predates and has been the inspiration for all of these apps we see on phones. Photoshop filters, lightroom presets and actions, etc. All of the stuff that that inspired things like instagram filters have been around for decades now, and have consistently always been a step ahead on power and quality. You don't run them on camera, but... you wouldn't want to! When you're shooting with a DSLR you're shooting for consumption at higher resolutions than you can see on your LCD display. You want to be able to look at things in more detail than you could ever do on the camera itself.

>The other reason is much more banal. For various reasons, people jumped to the DSRL bandwagon thinking that would make for better photos compared to more compact cameras (false for most people)

You're right here, but I don't think people actually wised up to this.

If we're talking specifically about DSLRs, mirrorless cameras are just better for most people at this point - Sony has almost single handedly rendered DSLRs obsolete for the vast majority of use cases where people need prosumer or professional cameras. If we're talking about prosumer+ cameras in general, it's even simpler than your reasons: Most of the people taking pictures with their phones are doing so for people who are consuming them on their phones.

You simply don't need a DSLR or mirrorless or medium format camera when everyone viewing your pictures is also using a phone. People aren't getting photos printed, and they're not viewing them on monitors. I'm really impressed with my iPhone 11 Pro camera - it works fantastic! - for viewing photos on the phone. But throw it up on my 4K monitor, and the pictures don't compare at all to my old D5300, much less my a7r 3 or even my rx100 m7 (assuming good shooting conditions and manual mode - the 11 Pro auto settings are quite good and significantly better than the 5300, and in some situations seem to be about as good as the rx100). I haven't gotten any printed to put up on display, but just seeing how poorly they compare when I'm looking at them in Lightroom, I wouldn't want to.

The way the majority of people consume photography has changed, and the requirements for producing photos for that medium has changed with it.