[BruceBanner]

I've concluded that it is impossible to have that discussion on a DSLR forum like this one. Within a few posts the topic swerves away from the interesting things you could do with large-sensor computational photography into things like "even if you could do that it would have to be mirrorless... why do you hate Pentax?" and "computational photography (besides pixel shift) is fake photography" and best of all "if you think phone cameras are so great why don't you stick to your iPhone and sell me your K-whatever and all your lenses for $300". Bumping this thread is a lesson in continued futility.

Honestly, it feels like that. But, whether we like it or not, painstakingly editing every picture, choosing this or that lens for their specific features and rendering instead of the spec sheet and this and that... that's a thing for serious hobbyists; not the kind of people that keep your business afloat in these times. Adding better connectivity, computational photography to, for example, get better portraits out of the kit lenses and this and that would help sales...

I just linked this video (and others before) to actually just (hopefully) educate and cement the point I was trying to make here, things are getting seriously cool on phones and little ol' me was wondering how and why the REAL cameras were getting left behind in terms of tapping into all of this computational advantage stuff. Perhaps I used the forbidden word in my headline (*Android*) and that rallied people into a craze! Honestly I don't care what OS the camera is running, as long as it improves upon the current shooting experience, not replaces, not something that takes away, but something that adds on, that's what I would like to see. There just seemed to be so much resistance and lack of vision to the idea and how it might be implemented. For example I imagined that the camera may have 'two modes', one that is default is the shooting experience we have now, quick start, no boot time, just straight into Pentax Firmware stuff, and another mode that is like the Playback mode but a little more indepth, this mode may take 20 secs or so to load up, but once in you might be in a iOS/Android like environment where wifi transfer and such things are not so important anymore, you can now post and edit images better directly from this mode. And then perhaps this mode also has some specialised Computational Photography modes as well. As long as this mode doesn't interfere with the current shooting experience, doesn't lag the camera mode more when in its default mode, then I see this as just an addon bonus, not a replacement. If you don't want any of these features then you don't need to visit this mode

I actually think something like this will happen, I don't know how it will be done but I cannot see the dslr market turn a blind eye to all the computational photography benefits that currently exist. Either that or Apple will actually enter the dslr market and combine what it already has achieved (advanced computational photography) with proper bodies and lenses... you may laugh...

I know so many people who were gifted cameras and gave up, because it was all too complicated and difficult. At the same time when they show me one of their pictures from their smartphone (that essentially did it all for them) they are content (up until the part I maximise the image and compare the weak parts of the smartphone tech with a real dslr). Because so many view their images on tiny screens they are ignorant to the major drawbacks to smartphone images. If someone comes up with a dslr/mirrorless camera that behaves like a smartphone (in terms of ease and computational photography advantages) but now is having proper glass attached... then I think that will make a major dent in the market. I know of two people who are not photographers who got gifted entry level dslrs this xmas... they won't use them more than 2-3 months tops, they'll go back to their iphone 11's guaranteed and it will be because the camera is confusing and the iphone 11 clever. Whichever company can produce an entry level dslr that is helping the end user out more like how smartphones are... then that will likely make a large impact in the market.

[ThorSanchez]

I just linked this video (and others before) to actually just (hopefully) educate and cement the point I was trying to make here, things are getting seriously cool on phones and little ol' me was wondering how and why the REAL cameras were getting left behind in terms of tapping into all of this computational advantage stuff. Perhaps I used the forbidden word in my headline (*Android*) and that rallied people into a craze! Honestly I don't care what OS the camera is running, as long as it improves upon the current shooting experience, not replaces, not something that takes away, but something that adds on, that's what I would like to see. There just seemed to be so much resistance and lack of vision to the idea and how it might be implemented. For example I imagined that the camera may have 'two modes', one that is default is the shooting experience we have now, quick start, no boot time, just straight into Pentax Firmware stuff, and another mode that is like the Playback mode but a little more indepth, this mode may take 20 secs or so to load up, but once in you might be in a iOS/Android like environment where wifi transfer and such things are not so important anymore, you can now post and edit images better directly from this mode. And then perhaps this mode also has some specialised Computational Photography modes as well. As long as this mode doesn't interfere with the current shooting experience, doesn't lag the camera mode more when in its default mode, then I see this as just an addon bonus, not a replacement. If you don't want any of these features then you don't need to visit this mode

I actually think something like this will happen, I don't know how it will be done but I cannot see the dslr market turn a blind eye to all the computational photography benefits that currently exist. Either that or Apple will actually enter the dslr market and combine what it already has achieved (advanced computational photography) with proper bodies and lenses... you may laugh...

I know so many people who were gifted cameras and gave up, because it was all too complicated and difficult. At the same time when they show me one of their pictures from their smartphone (that essentially did it all for them) they are content (up until the part I maximise the image and compare the weak parts of the smartphone tech with a real dslr). Because so many view their images on tiny screens they are ignorant to the major drawbacks to smartphone images. If someone comes up with a dslr/mirrorless camera that behaves like a smartphone (in terms of ease and computational photography advantages) but now is having proper glass attached... then I think that will make a major dent in the market. I know of two people who are not photographers who got gifted entry level dslrs this xmas... they won't use them more than 2-3 months tops, they'll go back to their iphone 11's guaranteed and it will be because the camera is confusing and the iphone 11 clever. Whichever company can produce an entry level dslr that is helping the end user out more like how smartphones are... then that will likely make a large impact in the market.

I'm going hazard a guess that nobody is going to make a simple entry level DSLR that will sell a lot of copies and make money. Because there are already simple DSLRs and mirrorless cameras. My K-3ii can be very simple. Stick it on green mode and fire away. The hurdle isn't the complexity of taking photos. Pressing the shutter button in green mode isn't more complex than pushing the shutter button on a phone. It's the size, and the cost, and the fact that you look like a photographer, and figuring out lenses (and the cost of the lenses) and lugging around a bunch of stuff. And the fact that most people view most pictures on a 4" screen, and most people take pictures to share with their friends not as art. For most people there's no going back, it's phones forever.


The one thing that perhaps could bring more people from phones to ILCs is the connectivity. Even for me, as a lugger of DSLRs, I would like to be able to share photos much more easily. Good internet connectivity would be wonderful. Often my wife will post a phone picture of something we're doing seconds after we've done it. My technically superior K-3ii photo won't show up until hours or days later, there's just no way to get it off the camera until I get to a computer. That will be an innovation that has to be standard in coming cameras.

But all the computational stuff? That's for me and you and the others who're into this enough to be on a forum like this. It's not going to make a bunch of phone photographers buy an ILC. And apparently it's for a small part of our peer group - because it certainly appears that many/most Pentaxians see computational photography (besides pixel shift and whatever Pentax has already implemented) as foolish.

[reh321]

(a) Valid points. It has to be about implementation, it has to enhance with NO downsides to current experience.

(b) I shoot weddings, I need the camera on ASAP and working as efficiently as possible.

I believe you are on the edge of truth, but not quite there.

(a) When Pentax introduced the 'accelerator" in the K-70, then the KP, everyone loved the implementation with good graceful high ISO, but they did not think about it - not until the K-1ii came out, that is. In fact, people want their 'raw' files to be exactly what the sensor saw, they do not want any sort of processing. It is not about implementation, it is about where the image came from. They want to control any 'computation' that occurs, which would essentally doom any Android inside the camer, because once you are performing operations in this way. you might as well build them into a 'raw' editor in the computer.

(b) I worked once on a research project on digital radios. Once, at the beginning of a demo, a US Air Force General told me something like "Starting this radio is too slow. I don't want to sit on the end of a runway while my radio starts". An Android-based camera will have the same problem. I wait impatiently the few seconds while my KP starts - I can just imagine the reaction to how long an Android-based camera would take.

[ZombieArmy]

While I think an open source camera OS being made would be cool, an android based one is basically DOA. Android is not light by any means, the hardware required to get it running seamlessly is pretty funny and I don't think people want really heavy weight hardware in their camera sucking down their battery further.

The OS would have to be built from the ground up to:

1. Start up instantly

2. Be seamless with physical controls

3. Be easy to cross compile to different processing chips

I think it's pretty unlikely we'd see it any time soon.

[gatorguy]

While I think an open source camera OS being made would be cool, an android based one is basically DOA. Android is not light by any means, the hardware required to get it running seamlessly is pretty funny and I don't think people want really heavy weight hardware in their camera sucking down their battery further.

There have been a very few Android-based standalone cameras (do a search), and if you remember Android was originally designed to be a camera OS anyway. I wouldn't discount the possibility of a mid-range DSLR or mirrorless running Android as the operating system in the relatively near future.

EDIT: Just saw this...
https://www.provideocoalition.com/zeiss-zx1-and-yongnuo-yn450-are-android-cameras-the-future/

FWIW there are also quite capable versions of Android that are very "light" and require little in the way of resources, for example Android Go. There's also tons of devices now with embedded versions of Android. I'd personally guess that somewhere in the bowels of Canon and/or Nokia engineering there are already working camera prototypes

[BruceBanner]

I'm going hazard a guess that nobody is going to make a simple entry level DSLR that will sell a lot of copies and make money. Because there are already simple DSLRs and mirrorless cameras. My K-3ii can be very simple. Stick it on green mode and fire away. The hurdle isn't the complexity of taking photos. Pressing the shutter button in green mode isn't more complex than pushing the shutter button on a phone. It's the size, and the cost, and the fact that you look like a photographer, and figuring out lenses (and the cost of the lenses) and lugging around a bunch of stuff. And the fact that most people view most pictures on a 4" screen, and most people take pictures to share with their friends not as art. For most people there's no going back, it's phones forever.


The one thing that perhaps could bring more people from phones to ILCs is the connectivity. Even for me, as a lugger of DSLRs, I would like to be able to share photos much more easily. Good internet connectivity would be wonderful. Often my wife will post a phone picture of something we're doing seconds after we've done it. My technically superior K-3ii photo won't show up until hours or days later, there's just no way to get it off the camera until I get to a computer. That will be an innovation that has to be standard in coming cameras.

But all the computational stuff? That's for me and you and the others who're into this enough to be on a forum like this. It's not going to make a bunch of phone photographers buy an ILC. And apparently it's for a small part of our peer group - because it certainly appears that many/most Pentaxians see computational photography (besides pixel shift and whatever Pentax has already implemented) as foolish.

Yes, you're right on all accounts. I think what I meant to say is that they think that when they see a good photo come from someone using a dslr they think that the camera is 90% the result, the user 10%. And so they buy one (even the same one) and then realise "oh wow... my shots are not like theirs at all.." and give up. There is quite a bit more to taking and deriving a good photograph than simply owning a nice and adequate dslr. The beginner realises this quickly and either gives up or decides they do actually have some homework to do. I see computational entry level photography cameras as a way to circumnavigate that hurdle so that straight from the get go things are looking really good, like iphone 11 on steroids good.

I believe you are on the edge of truth, but not quite there.

(a) When Pentax introduced the 'accelerator" in the K-70, then the KP, everyone loved the implementation with good graceful high ISO, but they did not think about it - not until the K-1ii came out, that is. In fact, people want their 'raw' files to be exactly what the sensor saw, they do not want any sort of processing. It is not about implementation, it is about where the image came from. They want to control any 'computation' that occurs, which would essentally doom any Android inside the camer, because once you are performing operations in this way. you might as well build them into a 'raw' editor in the computer.

(b) I worked once on a research project on digital radios. Once, at the beginning of a demo, a US Air Force General told me something like "Starting this radio is too slow. I don't want to sit on the end of a runway while my radio starts". An Android-based camera will have the same problem. I wait impatiently the few seconds while my KP starts - I can just imagine the reaction to how long an Android-based camera would take.

a) Well currently computational photography is bias towards a lot of automation, but that certainly doesn't need to be where it stays. I would hope it goes without saying that the advantages of such technology to the informed user is really exciting, this is not about taking control away from the photographer and letting the camera do it all for you (in the same way that having pixelshift on is only the beginning of deriving a good pixelshift shot, it most certainly doesnt start and stop at that process). I'm excited for what that technology could bring to the seasoned photographer!

b) Which is why I stated that perhaps an interface providing two modes would be necessary, not a replacement interface with smartphone like boot up times, but something that is an addon/toggle into mode (for when time is not so crucial). Heck.. it could still be the same mode as we all have come to know and love, pixelshift for example is still computational photography! I just want the other stuff too, and I don't care what box they package it in

[ZombieArmy]

There have been a very few Android-based standalone cameras (do a search)

And none are ones I'd own. But maybe I'm just out of touch?

There's also tons of devices now with embedded versions of Android.

I don't think I've ever used an embedded version of android that didn't make me want to die.

[photoptimist]

I question the premise that this is needed for two reasons.

First is the difference between smartphone users and DSLR users. Smartphones are for the vast majority of people who simply want to take photographs. DSLRs are for niche group of people who want to make photographs. It's two extremely different populations with two extremely different workflow needs.

It's like the fact that most people simply want to take a slice of bread. Only a few people want make a loaf of bread. For the people that just want some bread, the hassles of flour sifters, mixing bowls, rolling pins, bread pans, ovens, flour, eggs, salt, yeast, oil, etc. is too complex. And the people that actually want to make bread want the control over ingredients, tools, and process to make it the way they want.

Second, DSLR users already have access to computational photography. It's in the extremely powerful PC-based applications they use for post processing. And for the ways that DSLR users work to carefully make photographs, that's the more natural solution than some camera-internal app with automagical presets operated through a puny screen on the back of a camera. A PC has more power, larger screen, better user-input interfaces

[gatorguy]

And none are ones I'd own. But maybe I'm just out of touch?



I don't think I've ever used an embedded version of android that didn't make me want to die.

What devices do you think you used with embedded Android?

It's a very capable and adaptable OS and not simply something running smartphones. You may be using it now and not even realize it since it's used in devices ranging from routers and VOIP phones to automotive systems, and home appliances. Tesla makes extensive use of it, as does Volvo and Audi. Beginning late this year GM will be relying on Android code as well.

Don't be shocked to see a fully-capable prosumer camera making use of Android code.

Remember the Samsung NX camera line? They were pretty well reviewed and many shooters quite like shooting with them. It ran Android.

[reh321]

I question the premise that this is needed for two reasons.

First is the difference between smartphone users and DSLR users. Smartphones are for the vast majority of people who simply want to take photographs. DSLRs are for niche group of people who want to make photographs. It's two extremely different populations with two extremely different workflow needs.

This simplification is not true at all as stated. I am one of many DSLR users who take photos - yes, I will do a few very simple "PP" operations, just like news and sports photographers do; I cannot imagine most who photographs sports for a living switching from ILC to a smartphone, either. When I first switched to digital, I used a compact camera, which is much more powerful than any camera-based phone is, but it was much too inflexible for me. I need an ILC, for example, and I really don't expect a phone to provide that anytime in my lifetime {I passed 72 in the past month}. But you are correct in that I wouldn't want Android, as I have already indicated.

[BruceBanner]

I question the premise that this is needed for two reasons.

First is the difference between smartphone users and DSLR users. Smartphones are for the vast majority of people who simply want to take photographs. DSLRs are for niche group of people who want to make photographs. It's two extremely different populations with two extremely different workflow needs.

It's like the fact that most people simply want to take a slice of bread. Only a few people want make a loaf of bread. For the people that just want some bread, the hassles of flour sifters, mixing bowls, rolling pins, bread pans, ovens, flour, eggs, salt, yeast, oil, etc. is too complex. And the people that actually want to make bread want the control over ingredients, tools, and process to make it the way they want.

Second, DSLR users already have access to computational photography. It's in the extremely powerful PC-based applications they use for post processing. And for the ways that DSLR users work to carefully make photographs, that's the more natural solution than some camera-internal app with automagical presets operated through a puny screen on the back of a camera. A PC has more power, larger screen, better user-input interfaces

With all due respect i think you're over complicating things. Whether a photo is taken with a smartphone or a dslr, its still just a photo (and the mass medium of consumption of images right now is small screens and not print, that's just how things are nowadays). People just want 'nice photos' and that is derived from various methods, some advanced and some automated. There are many sides of computational photography that perhaps you are missing out on, we cannot for example change focus of where the shot is taken from with our dslrs, maddening to think we can right now with smartphones. Imagine just fixing focus with our dslrs because at the the time we caught the eyebrow vs eyelash/eye (and nothing in PP can do this currently, or if we are its sharpening tools which degrade IQ). There's nothing magical going on here, just really clever computational photography that would have endless benefits to the dslr shooter without being 'automated' per se. You have to use a degree of imagination when approaching this subject. For example, much bokeh is 'fake' in smartphones, out dslrs can navigate that hurdle competently and leave resources to other more important tasks.

This is what will happen onwards, make no mistake, change is coming in this direction to the dslr market whether one likes it or not, that much I am sure. Why? Because when tech advances no one ignores it, you go out of business ignoring massive advantages. Smartphones are showing what is possible in a small compact form, it's mind blowing to think what will come once this is implemented into proper cameras.

Right now it goes something like this;

Smartphones, 80% software, 20% tech, making awesome photos from clients that have no idea about photography at all.

vs

Dslrs/Mirrorless/whatever, 80% tech, 20% software, struggling in current market for profits, higher skill set required, higher prices and higher post processing time costs.

Now that smartphones have demonstrated the advantages of computational photography, how it can help with revenue and sales... how can the dslr/mirrorless market ignore this? That is my point. I sincerely believe they won't. Right now its a gold rush for the first dlsr company to strike gold and bring the best of the two worlds together. Make a camera that is successful for the entry noob and a powerhouse for the seasoned pro. I just cannot fathom what smartphones are doing right now to be ignored... like... I have never encountered an example of a tech market just go "yeah nah... lets not bother trying to do that with all the $$ it could bring". As I said... the entire mp3 market adopted Android to stay alive, and it worked. That doesn't mean I think all cameras will become Android lol, (and maybe my title to this entire thread is ill placed), but something is going to happen to our cameras to employ these benefits more than 'pixelshift'.

[redcat]

The full Android camera is in production many years ago and is voted "camera of the decade" !!!!

Youtu.be

Youtu.be

[BruceBanner]

The full Android camera is in production many years ago !!!!
Camera of the DECADE (you'll HATE it) - YouTube

Just watched that now, wow... I feel uncomfortable kinda agreeing with Tony.

I think there's a lot going on with the Galaxy NX that contributed to it's demise, Tony touched upon a few of them but I would add these also;

1) The Android interface replaced something, not actually becoming an additional feature. That hit them hard. Notice the absence of dials and a slick rear 'touch only' interface. That is not and never will be what camera users really want. It had to be an 'and/or' option. We need all the tactile buttons, dials and wheels that we have come to need and appreciate in the field, not for them to disappear and be replaced with only a screen.

2) Computational Photography was weak back in the NX release time. Things have changed massively and so one of it's major benefits of having a camera with Android OS was losing huge potential, effectively making it feel like all it had to bring to the table was 'Social Media' options.

3) I owned a Galaxy K Zoom smartphone, 20mp, xenon flash and 10x optical zoom. It was great at the time, but you do suffer from things like start up times, battery life etc. If any modern 2020 camera that wants to take advantage of computational photography (in the shape of offering an OS that can run the necessary apps) then it has to be implemented cleverly. We camera users will never take a proper camera seriously if it takes 20-30secs to fire up before we can squeeze off even a basic non computational photography shot. I'm no expert but it feels like a camera that can succeed where others have failed must have two modes, one that turns the camera on quickly like what we have come to accept and appreciate and another mode/button that will take you to a deeper OS that is richer but will take slightly longer to move/toggle you there, but upon entering that mode you will have;

- potential to make a phone call
- instant messaging
- a more powerful review section
- specific computational photography modes
- ability to access more editing modes for the photos on the camera (Lightroom CC etc)
- backup photos to the cloud out in the field
- instantly upload images to social media outlets or forums
- access certain useful apps such as Golden Hour

and I am sure there are other benefits I have missed.

But all of this must be included well, if a single component of this replaces a feature of the basic camera we have come to love and appreciate then it will be a hard sale, because... for many of us our phones are always with us anyway. Really I see 'computational photography + proper camera' as being its largest benefit that should not be dismissed lightly, the other stuff is a welcome addon, or perhaps helps people travel lighter in the field by even allowing us to leave the phone at home.

[hvz448]

Guess OP can get one now


Zeiss' ZX1 Android-Powered Camera Costs $6,000