> Imagine if Apple and Google made full frame cameras.
I did – briefly – and am still recoiling from the horror.
> Currently we have great hardware with terrible software and vice-versa.
The software on even older DSLRs isn't terrible. It does the job it was intended to do quite well. It is learn-able, and it doesn't change its behaviour with an OTA-update you can't easily prevent.
More thought has gone into the UI design of even the earliest, rushed-to-market DSLR firmwares than some of the UI changes that Apple and Google have inflicted on their victims.
That's exactly the problem. If you put Apple or Google in charge, you'll get a true software platform acting on all those people. That's going to make everything slow and terrible.
The beauty of a DSLR is that you turn it on, and it's ready to go. Take pictures in raw and process them when you get home. If you want to adjust the exposure and what not, you can; if you do that a lot, you can get the higher model cameras with more dials, and customizable buttons so you don't have to go through menus as much.
I don't know about Android, but the Camera app on iOS is extremely responsive.
And I envy the amount of processing that goes on when you press the button.
From turning a noisy mess from a fingernail size sensor into pretty good JPGs to auto choosing the best photo from crazy fast burst mode, even before you press the button!
And audio/video, oh boy. Hyper slow motion 4k, noise canceling…
Imagine what Apple/Google could do with something the size and weight of a DSLR and 3k US$ budget.
I love my iPhone and I'm happy to go to bat for Apple when I agree with them, but the Camera app on iOS is painfully slow compared to a Canon DSLR. The Canon can wake from sleep and take a photo in the tiniest fraction of a second.
What do you mean locked? I can go from powered off to shooting my Canon in a fraction of the time than my iPhone from locked. And importantly a good photo because I can zoom and frame while the camera is getting the software ready.
Where the iPhone wins is when it’s already in my hands.
I have an iPhone 7. I’ve never quite gotten the gesture right in the moment. It feels like it has some degree of spurious input rejection on the unlock screen.
I hate my iphone x camera because it's slow. I have to double and triple check that it's focusing on the right thing and hold the phone really steady. It is adequate for me because I always carry my phone with me, but I would never want a general approach to a real DSLR or mirrorless camera.
Apple/Google would add too many features, focus on wifi connectivity, and generally ruin the one amazing thing about cameras–the ability to turn them on and immediately take photos, no fuss. No waiting for software updates, or alerts, or any of the million other things that do not belong in a camera.
The 1DX Mark II (admittedly a very expensive pro camera) goes from physically switched off to first shot in 0.8 seconds, and from standby to image capture with full autofocus in 0.085 seconds: https://www.imaging-resource.com/PRODS/canon-1dx-ii/canon-1d...
My D750 can go from OFF -> ON -> Shutter press -> focus -> picture taken in cca 0.5s in normal light conditions. There are so many moments that I managed to capture only because of this speed, they would be gone in those 3s.
Not even going into "details" like having 5x mechanical zoom on it (24-120mm), I can snap distant scene in 1s in above scenario. And I do snap those scenes quite often.
> I'm imagining a processor like the one on the iPhone/iPad Pro acting on all those pixels. And a true software platform running on top of it.
I think a lot of professionals and prosumers who are still buying interchangeable lens cameras don’t care. Folks generally shoot in RAW and edit their photos on a PC with Photoshop and Lightroom, which affords people much more control and is much more powerful than whatever is in an iPad.
The only photographers shooting JPEGs are sports and news photographers, and they have specialized use cases. That’s why the pro cameras (1DX2 and D5) come with Gigabit Ethernet and auto-upload to an FTP server through that.
People want the pre-capture sequence to be reliable and easy to use (so lots of buttons and dials) and in general once the photo is captured and saved, they don’t care and don’t want the camera involved any more.
Of course, if people really want to upload to Instagram immediately, most cameras nowadays come with Wifi so they can just transfer photos to a smartphone and do whatever they need to do there.
Do you know what is being done to the data between coming off the sensor and being saved to a RAW file? My impression is that other than dead/hot pixel mapping and some degree of noise reduction, not much else is done (at least in Canon-land).
Even if the camera can do more post-processing, I am not sure I want it to. I would rather the camera do less after exposure is complete, so I can get more control over it while editing.
At the very least there’s signal amplification (and possible highlight clipping) at higher than baseline ISO. Otherwise you'd get a very dark image indeed.
Then consider advances in optics, CAD, sensors, and then marvel at why some people are desperate for an arbitrary sensor size that plucked out of the air more than a hundred years ago.
I did – briefly – and am still recoiling from the horror.
> Currently we have great hardware with terrible software and vice-versa.
The software on even older DSLRs isn't terrible. It does the job it was intended to do quite well. It is learn-able, and it doesn't change its behaviour with an OTA-update you can't easily prevent.
More thought has gone into the UI design of even the earliest, rushed-to-market DSLR firmwares than some of the UI changes that Apple and Google have inflicted on their victims.