Canon want's a new feature...you decide.

Ron Lacey wrote in post #8691175
Don't they have that now? I'm pretty sure my 50D does and I assume all the other live view bodies do too.

See i told you it was a great idea......

Expanded live view functionality
Good for shooting subjects which don't move much.

- Electronic horizon (both in live view and in viewfinder)
- Ability to choose smaller areas for contrast-detect AF

Contrast-detect AF without using Live View
No more AF adjustment and no more being limited to whatever AF points are provided in-camera.

Contrast detection, as used in Live View, is more accurate than the current phase detection system and you can focus on any part of the image, not just the fixed AF points. Trouble is, it's slow and you can't use the viewfinder at the same time.

The viewfinder problem can be solved using a pellicle mirror, allowing light to reach the sensor for contrast detection while you are looking through the viewfinder. This can be swung out of the way just like a conventional mirror during exposure, to eliminate the loss of light and extra source of flare that would result if the mirror were left in situ during exposure. The speed issue can be resolved with faster processors; also, with the phase detection system still in place, you could choose to turn off contrast detection when shooting fast-moving things and just use the faster, but less-accurate and less flexible system.

Low ISO support
ISO 0.2 is just as useful as ISO 6400. It's the equivalent of putting a 9-stop ND filter on a camera shooting at ISO 100. ND filters are one of the most useful filters out there, for flash photography and long-exposure shots, but their need could be entirely eliminated by having in-camera low ISOs.

Moveable sensor
Pentax has inbuit sensor rotation to correct for uneven horizons. The same technology could be applied to tilt the sensor, giving the Tilt function to all lenses, TS-E or not. By making the whole sensor mobile, you could give the Shift function to all EF lenses on crop bodies (full-frame bodies would still be limited by the size of the image circle).

Electronic shutter for unlimited sync speed
Currently, sync speed is limited by the movement of the shutter across the frame.

With the option to use an electronic shutter, you could shoot at whatever sync speed you wanted, making it easier to capture action and possible to out-flash the sun at lower flash power. This would be a simple firmware fix.

Concurrent exposure bracketing
Instead of taking shots one after the other (and giving subjects time to move between each shot), why not take them concurrently, taking measurements after each shutter time has elapsed?

For a 5-shot sequence, for example, instead of taking 5 separate shots at 1/1000, 1/500, 1/250, 1/125 and 1/60 (limited also by the camera's frame rate), just take a single shot at 1/60 and read from the sensor 5 times, at 1/1000, 1/500, 1/250, 1/125 and 1/60.

Logarithmic response curve
One does not need 8000+ luminance levels in the brightest stop of DR. Similarly, one needs more than 2 luminance levels in the dimmest stop. A logarithmic curve would make the shadows just as detailed as highlights and eliminate the need to shoot to the right (with all the risks of blowing highlights and the difficulties in shooting scenes with high dynamic range using this method).

16-bit output
Combined with a logarithmic response curve, this would be invaluable for eliminating posterisation while pushing and pulling images.

True underwater seals
They seem to be able to do it with every other piece of electronic equipment out there...

Inertial movement detection
Put an accelerometer into the camera and you can record how and where the camera moves during an exposure. Couple it with the lens' IS mechanism and you can also record the compensation provided by IS. With these two pieces of information, one could compensate in software for movement and vibration during an exposure, eliminating motion blur as a cause of softness.

Software/firmware improvements
- Motion blur compensation - using the accelerometer data recorded at the same time as the image, it is then possible to correct for motion blur. Other developmental software can do it blind, with no accelerometer data, but is much slower and less accurate.
- Diffraction compensation - Airy diffraction patterns are well-described; knowing the mathematics behind it, and given a particular sensor's pixel pitch and the f/stop of a particular exposure, one can correct for diffraction in software, making large DOFs for landscapes and group portraiture much more feasible without losing sharpness and detail resolution. This would also make the ever-increasing megapixel count much more useful... Shooting at f/128 on full-frame? Not a problem!

Dump the video, pop-up flash, the useless print button, all ISO over 1,000 and kill the green box and other basic zones from the mode dial.

Give me a built in grip and a sensor having a true ISO 25 capability.

Naturalist wrote in post #8692195
Dump the video, pop-up flash, the useless print button, all ISO over 1,000 and kill the green box and other basic zones from the mode dial.

Give me a built in grip and a sensor having a true ISO 25 capability.

No point dumping video unless you were to dump live view as well... video requires no extra hardware.

BTW what does a built-in grip give you that an add-on grip doesn't, apart from the loss of an option for a smaller body?

Shadowblade wrote in post #8692223
BTW what does a built-in grip give you that an add-on grip doesn't, apart from the loss of an option for a smaller body?

better fit and finish. better sealing.

Coffee maker.

Modular sensor/processor unit. Instead of buying an entire new camera every time there is a technology upgrade, just pop in a new module and send the old one back to Canon to be recycled.

I guess I miss having a camera that can last 30 years and still use the latest and greatest "sensor" technology (film). Could be some interesting advantages to this approach:

- You could have different modules with different resolutions or full frame and crop modules that could fit in the same body.
- I have to think this approach would create less landfill from disposal of used/broken electronics gear.
- Modules could be made to a standard that would allow swapping between brands. Like Nikon's ergonomics but Canon's sensor ... No Problem!

aroundlsu wrote in post #8690453
Built in radio trigger. Pocket Wizard compatibility would be a plus, obviously.

I'd prefer to see the Radio Popper system integrated due to the ongoing RF issues PW has with Canon flashes.

5DII - make Auto-ISO in M to be Auto-ISO instead of 400 fixed

I know that this won't be very popular on the board, but I think that in regards to NEW ideas, the more toys the better (assuming that they work).

built in GPS (not a new idea) writing to the EXIF is a great idea (though ability to turn off or on). even software to map out the pics in addition. dynamic range and better AF functions have been mentioned here and I don't think anyone would complain, but again these (and many more suggestions) are modifications of existing systems.

Tetris on the other hand is definitely "new" for a DSLR!

Auto focus at hyperfocal distance.

That kind of exists already in the shape of the A-DEP function.
Not many like it.

Built in 3G/4G broadband so I can upload a file to my editor within seconds of taking the shot.

apersson850 wrote in post #8690396
I want automatic AF micro adjustment.

+1..

Shadowblade wrote in post #8692070
Concurrent exposure bracketing
Instead of taking shots one after the other (and giving subjects time to move between each shot), why not take them concurrently, taking measurements after each shutter time has elapsed?

For a 5-shot sequence, for example, instead of taking 5 separate shots at 1/1000, 1/500, 1/250, 1/125 and 1/60 (limited also by the camera's frame rate), just take a single shot at 1/60 and read from the sensor 5 times, at 1/1000, 1/500, 1/250, 1/125 and 1/60.

This seems like a brilliant idea to me!

Also, additional votes for onboard GPS-to-EXIF and wireless data transfer.