[Poit]

I'm just putting it out there...and feel free to tell me why it could never happen (no doubt people will anyway)....but what if a DSLR was enabled for third party application developers, and those developers were able to improve the performance of the camera (such as AF). Would that be a bad thing?

If it could be done, and Pentax was brave enough to do it, would that not help to resolve some 'tricky' issues without employing their own additional development staff?

That's what I was thinking when referring to 'apps'. It's also what I meant when I said Pentax cameras would 'fly off the shelf'. A world in which bright, independent developers actually improve our cameras. That would be very cool.

Dare to dream, folks. 15 years ago we were all claiming that digital would never be as good as film...

---

Last edited by Poit; 11-11-2016 at 10:18 PM.

[lithedreamer]

I'm just putting it out there...and feel free to tell me why it could never happen (no doubt people will anyway)....but what if a DSLR was enabled for third party application developers, and those developers were able to improve the performance of the camera (such as AF). Would that be a bad thing?

If it could be done, and Pentax was brave enough to do it, would that not help to resolve some 'tricky' issues without employing their own additional development staff?

That's what I was thinking when referring to 'apps'. It's also what I meant when I said Pentax cameras would 'fly off the shelf'. A world in which bright, independent developers actually improve our cameras. That would be very cool.

Having done some app development myself, providing an ecosystem that supports outside developers without just leaving consumers with bricked cameras or no additional meaningful functionality would actually require a lot more support than Ricoh currently provides Pentax cameras.

[Poit]

Having done some app development myself, providing an ecosystem that supports outside developers without just leaving consumers with bricked cameras or no additional meaningful functionality would actually require a lot more support than Ricoh currently provides Pentax cameras.

Oh ok. Thanks for the constructive feedback.

[lithedreamer]

Oh ok. Thanks for the constructive feedback.

Thanks for being understanding.

It might be a worthwhile investment anyway. Smartphones are largely successful due to third-party apps after all. Camera manufacturers are just used to a hardware-based way of working, and with falling sales, have moved to conservative business models, so I don't see it happening anytime soon. Sony does it closest, with 'apps' you can pay to download from their App Store, but all of those are developed in-house, as I understand it.

[Poit]

Thanks for being understanding.

It might be a worthwhile investment anyway. Smartphones are largely successful due to third-party apps after all. Camera manufacturers are just used to a hardware-based way of working, and with falling sales, have moved to conservative business models, so I don't see it happening anytime soon. Sony does it closest, with 'apps' you can pay to download from their App Store, but all of those are developed in-house, as I understand it.

No probs. I'm fine with civil conversations. Just the endless defensive 'you're an idiot' style comments bother me.

And yes, it might be worthwhile. That was why I suggested it in the first place. It would certainly be very exciting and a great move for whoever does it successfully. But, with your comments, maybe the investment in the ecosystem would be too great.

Yeah, I know Sony does it but that's in-house, as you say. I'm suggesting allowing independents get involved.

[lithedreamer]

No probs. I'm fine with civil conversations. Just the endless defensive 'you're an idiot' style comments bother me.

And yes, it might be worthwhile. That was why I suggested it in the first place. It would certainly be very exciting and a great move for whoever does it successfully. But, with your comments, maybe the investment in the ecosystem would be too great.

Yeah, I know Sony does it but that's in-house, as you say. I'm suggesting allowing independents get involved.

Yeah, it's definitely refreshing to have some civil conservations, especially with the way the States have been blowing up this week. I think Ricoh has it in its DNA to change, it just doesn't have the capital to be the first. You see it in the way they create things like astrotracer, pixel shift and the funky tilt-shift screen on the K-1. If Samsung was still in the camera business, they could have been the first. Instead, I'd expect Sony or Nikon (they've been having the best luck with wifi with their Snapbridge [wifi+bluetooth] technology, even if they're fumbling with updates) to make the first move in that space if it's an existing company.

Otherwise, expect an outsider to enter the market eventually from an adjacent space, like smartphones or drones, if the current manufacturers never get their act together.

[Kunzite]

Having done some app development myself, providing an ecosystem that supports outside developers without just leaving consumers with bricked cameras or no additional meaningful functionality would actually require a lot more support than Ricoh currently provides Pentax cameras.

I would add that things like improving AF speed (or performance, in general) would be very difficult, even for experienced developers. On the contrary, I have every reason to suspect that the performance could get worse (of course, I don't know until it's done and measured)

Opening every aspect of their protocols and algorithms and whatever would be problematic, too. User applications would rather target higher level operations, for example automatic panorama stitching, or an advanced intervalometer, or easy upload to social media. Like the apps Sony has on its Alpha application store. And Samsung had Android-based cameras. Yes, it's been done, with no accompanying dramatic increase in sales.

I wish people would stop saying that cameras would fly off the shelves only if X would be done, especially when they don't know what X is. There is no silver bullet; no secret recipe which could push the sales with something like an order of magnitude.

[Rondec]

I would add that things like improving AF speed (or performance, in general) would be very difficult, even for experienced developers. On the contrary, I have every reason to suspect that the performance could get worse (of course, I don't know until it's done and measured)

Opening every aspect of their protocols and algorithms and whatever would be problematic, too. User applications would rather target higher level operations, for example automatic panorama stitching, or an advanced intervalometer, or easy upload to social media. Like the apps Sony has on its Alpha application store. And Samsung had Android-based cameras. Yes, it's been done, with no accompanying dramatic increase in sales.

I wish people would stop saying that cameras would fly off the shelves only if X would be done, especially when they don't know what X is. There is no silver bullet; no secret recipe which could push the sales with something like an order of magnitude.

That's what I struggle with. The idea that some one or two person developer is going to whip up an algorithm on their desktop computer at home that will let the K-1 auto focus 50 percent faster is unlikely to say the least. The possibility is a lot stronger that they could somehow make the camera unstable.

I do think it would be a lot more likely that developers could improve video performance because Pentax has sunk very little effort into that and doing things like enable mechanical SR during video or, adding focus peaking during video would seem like relatively minor things. But even there, I don't know how much I would trust a third part firmware, particularly if it voided my warranty to download it.

None of these things will increase the number of ILCs sold, though. At best, it could shift a small percentage of people from one brand to another who are already interested in buying a SLR or mirrorless camera.

[RobA_Oz]

The next generation of DSLR users might just expect that these things are available for them.

[Kunzite]

doing things like enable mechanical SR during video or, adding focus peaking during video would seem like relatively minor things.

They don't look minor to me

But even there, I don't know how much I would trust a third part firmware, particularly if it voided my warranty to download it.

And this is another issue.
I really doubt a camera maker would sell a camera without guaranteeing it's functioning. But how could they do that, if anyone can change anything? How would they even advertise a feature while saying it would void the warranty?
Sony's Play Memory is a tightly controlled application store, and they don't offer anything like that.

None of these things will increase the number of ILCs sold, though. At best, it could shift a small percentage of people from one brand to another who are already interested in buying a SLR or mirrorless camera.

I agree.

[Nicolas06]

That's what I struggle with. The idea that some one or two person developer is going to whip up an algorithm on their desktop computer at home that will let the K-1 auto focus 50 percent faster is unlikely to say the least. The possibility is a lot stronger that they could somehow make the camera unstable.

I do think it would be a lot more likely that developers could improve video performance because Pentax has sunk very little effort into that and doing things like enable mechanical SR during video or, adding focus peaking during video would seem like relatively minor things. But even there, I don't know how much I would trust a third part firmware, particularly if it voided my warranty to download it.

None of these things will increase the number of ILCs sold, though. At best, it could shift a small percentage of people from one brand to another who are already interested in buying a SLR or mirrorless camera.

I think it is quite likely that 1-2 person developper if motivated and provided the right incentive/environment to work could do it. There well known experiment among companies of asking people to improve their algorithms (netflix algorithms to find the movie you like most come to mind).

It is well known in the area that the difference in productivity can be at least a factor of 10. That why startup manage to make thing that thousand people fail to do and so on.

But for that to happen, and that may not be that easy for AF, you have to provide an env where you can test, operated and evaluate AF on a computer, without having to shoot the real scene. You need to make it like a challenge for the brightest people of this world to put their brain on. If they achieve what you ask, it should be made public so they can put it on their CV and their current or future employer would be aware of it. Potentially the next employer should be your company, with enough freedom of mind for the winner to want to accept the offer while he can likely work for google, facebook, microsoft or Canon. And of course they should be some money incentive like at least $10000 for the best "teams", maybe $100000 for the winner. You should advertise that quite loudly even if that would be cheap counting the target audiance: researcher, student engineers and alike. For universities you could propose to finance a thesis or something. What count is that you motivate the smartest people with a knowledge in the area your are interrested to work for your problem and propose things.

Finally, this would not be apps, but likely only algorithms, Ricoh would then have the choice to integrate or not the best algorithms that improved the most.

But this would likely need huge preparation, some time to setup, analyse, and then put back into camera. And this would not be really apps.

As for what even a normal app dev could do, I am sure a single guy can make the video experience many time better like the guy of magic lantern do on Canon. and I am sure you can get interresting stuff like nice panoramas done. Why not partner with Microsoft research that an app that does just that on phones and also on desktop? A cooperation between companies would likely bring interresting stuff.

[Kunzite]

Your 1-2 person team won't even fully read the detailed technical documentation (assuming it would be available). And 1 people making the job of 10 men can only be possible with completely compromising code quality. Performance tuning? Don't even think about that; no time when you have to be 10 times more "productive" - and you're making a good point about testing. Maintenance? Not possible, because code hastily written is a mess. Making sure it works with every Pentax lens? Nope, won't happen, at best hope decent compatibility with the lenses the developer(s) happened to have.

In the meanwhile, the people who could improve the AF/performance/whatever (i.e. Ricoh Imaging's) would be too busy to rewrite the code base just to allow third parties to mess with it.

I'm fine with people dreaming; except if they're dreaming that difficult things are easy. And if they're puzzled when Ricoh Imaging doesn't easily implement those difficult things.

[Rondec]

They don't look minor to me


And this is another issue.
I really doubt a camera maker would sell a camera without guaranteeing it's functioning. But how could they do that, if anyone can change anything? How would they even advertise a feature while saying it would void the warranty?
Sony's Play Memory is a tightly controlled application store, and they don't offer anything like that.


I agree.

I had in my mind that there was code already there for use of these things in live view and it would just have to be activated for use in video. But I am not a software developer. My fear in general is that small changes in one area of the firmware could lead to instability in other area. Something like the whole uncontrolled mirror issue that some users experienced on the K3. Those sorts of things would kill any small benefits you would get from alternate firmware versions.

[Kunzite]

You are not a software developer (I am, though I'm not writing camera firmware), but your concerns are entirely valid. That's why I'm saying Ricoh won't ever give people full access to the hardware - the equivalent of kernel mode / driver programming. User applications runs in a more restricted environment, and have fewer capabilities.