A Canon DSLR First Coming to EOS 5D Mark IV [CR2]

[StudentOfLight]

New rear LCD with resolution increased to approx 2560x1704

 

[Sharlin]

An Android OS module (some P&Ss are Android; you can install whatever editing or social media apps you want and use them in-camera).

Related but not exactly along those lines... near field communication could be a nice feature to pair your phone/tablet etc with the body more easily/quickly seamlessly for remote shooting. That would be handy, not very major and would make at least some of us happy.

The 80D already has NFC. I'd wager most if not all future Canon cameras that have WiFi will also have NFC.

 

[PFerrara]

Haven't read through 17 pages, but my guess is the "first" for a Canon DSLR would be using somebody else's sensor. That could make a lot of shooters happy.

 

[Mikehit]

Haven't read through 17 pages, but my guess is the "first" for a Canon DSLR would be using somebody else's sensor. That could make a lot of shooters happy.

I wouldn't call that 'minor' by any stretch of the imagination.

 

[kaihp]

The more I'm thinking about it, the more convinced I am that this "Canon DSLR First" is either a Type-C USB (3.0) connector or powering/charging through USB. Maybe even both

 

[rrcphoto]

things canon hasn't done yet in a DSLR (may have in an ILC ie: M, or powershot) that others have:
- high speed crop mode (haven't seen any patents)
- some form of auto MFA. (canon has patents)
- IBIS (canon doesn't have patents for this .. it isn't happening)
- focus peaking

things that canon (and other DLSR's haven't done yet, but other cameras have).
- hybrid EVF
- EVF hotshoe for the EVF-DC1

 

[Ozarker]

In camera image stabilization that can be turned on and off?

 

[Maiaibing]

Got it! Eye-focus

Finally, Canon will bring back eye-focus too its DSLRs. ;D

It fits the criteria for the rumor (while a lot of the above suggestions do not).

Canon has the tech and it would be a very helpful AF upgrade for a Canon DSLR since Canon does not have focus-point AF or eye-AF etc.

 

[Michael Clark]

It would be a 30.6x30.6mm sensor with a 43.3mm diagonal.

No, it would be a 43mmx43mm sensor, covering the entire image circle. It would either be mirrorless or use a different mirror mechanism (sliding like a shutter rather than flipping?)

This has been covered so many times here, just look at the basic geometry of a circle and a square for goodness sake!

I'm sorry, what was the point?

If you project a 43mm image circle onto a 43x43mm square The circle would only project light on 1452mm² of the 1849mm² square. Why create a sensor when 27% of it will never have any light fall upon it? For an image circle to cover a 43x43mm square it needs to have a diameter of the 61mm diagonal of said square. It isn't exactly rocket science.

 

[Michael Clark]

A round image sensor would work. The RAWs could be circular and you would just choose what you want. Square? Done. Any aspect ratio with the largest possible sensor size for full frame lenses and no part of the sensor outside the image circle.

I probably sound like I'm trolling

No, its not something on my wish list at all, but it could be an option (instead of a square sensor). I'm not sure what implifications it would have on the manufacturing processes.

You'd have to cut the wafers to use more area of the wafer per sensor than the area of the circular sensor. You could stagger each row but you'd still wind up with material between the tangents that would be wasted.

 

[Michael Clark]

Actually the bigger improvement would be curved sensors to reduce/avoid the need for the full panoply of corrective elements needed for a rectilinear lens.

IIRC, Sony patented something like that a while back, though I believe the intent was for a fixed-lens camera rather than an ILC.

Wouldn't the radius of curvature need to equal the focal length? Which means any particular sensor would only work with a single focal length...

 

[Orangutan]

It would be a 30.6x30.6mm sensor with a 43.3mm diagonal.

No, it would be a 43mmx43mm sensor, covering the entire image circle. It would either be mirrorless or use a different mirror mechanism (sliding like a shutter rather than flipping?)

This has been covered so many times here, just look at the basic geometry of a circle and a square for goodness sake!

I'm sorry, what was the point?

If you project a 43mm image circle onto a 43x43mm square The circle would only project light on 1452mm² of the 1849mm² square. Why create a sensor when 27% of it will never have any light fall upon it? For an image circle to cover a 43x43mm square it needs to have a diameter of the 61mm diagonal of said square. It isn't exactly rocket science.

It would be a premium/niche product. Please look back at the thread for context: the point is that if you want to get the maximum light out of the existing EF lens line your options are limited. As it is now, we're throwing away about half the light of the image circle. Of course, that estimate is a little high because the edges of the glass distort light beyond usefulness, but it's still a lot of wasted light. Without going to a MF lens, what else could salvage that light?

 

[neuroanatomist]

Actually the bigger improvement would be curved sensors to reduce/avoid the need for the full panoply of corrective elements needed for a rectilinear lens.

IIRC, Sony patented something like that a while back, though I believe the intent was for a fixed-lens camera rather than an ILC.

Wouldn't the radius of curvature need to equal the focal length? Which means any particular sensor would only work with a single focal length...

No, it wouldn't. Just as current optical designs project a planar image regardless of focal length (although some fail, e.g. the original 24-70/2.8L's notable field curvature), lenses of different focal lengths could be designed to produce the same fixed curvature of the image 'plane'. However, it would be challenging for longer lenses and for zooms.

 

[slclick]

SD Card slot Wifi adapter

 

[tpatana]

SD Card slot Wifi adapter

Now that would be interesting option!

 

[Michael Clark]

Actually the bigger improvement would be curved sensors to reduce/avoid the need for the full panoply of corrective elements needed for a rectilinear lens.

IIRC, Sony patented something like that a while back, though I believe the intent was for a fixed-lens camera rather than an ILC.

Wouldn't the radius of curvature need to equal the focal length? Which means any particular sensor would only work with a single focal length...

No, it wouldn't. Just as current optical designs project a planar image regardless of focal length (although some fail, e.g. the original 24-70/2.8L's notable field curvature), lenses of different focal lengths could be designed to produce the same fixed curvature of the image 'plane'. However, it would be challenging for longer lenses and for zooms.

But isn't the entire point to begin with of a curved sensor to eliminate the need for lens correction? In terms of things such as field curvature, etc.

 

[tron]

Actually the bigger improvement would be curved sensors to reduce/avoid the need for the full panoply of corrective elements needed for a rectilinear lens.

IIRC, Sony patented something like that a while back, though I believe the intent was for a fixed-lens camera rather than an ILC.

Wouldn't the radius of curvature need to equal the focal length? Which means any particular sensor would only work with a single focal length...

No, it wouldn't. Just as current optical designs project a planar image regardless of focal length (although some fail, e.g. the original 24-70/2.8L's notable field curvature), lenses of different focal lengths could be designed to produce the same fixed curvature of the image 'plane'. However, it would be challenging for longer lenses and for zooms.

But isn't the entire point to begin with of a curved sensor to eliminate the need for lens correction? In terms of things such as field curvature, etc.

Different lenses, different focal lengths, different behavior (like field curvature)...

 

[Daan Stam]

maybe bluetooth

 

[Donkey_Kong]

I think we can safely assume the 5DIV will not have a round sensor..

 

[bvukich]

But increasing the height of the mirror by 6.6mm to 30.6mm likely means the current 44mm registration distance would be too short to accommodate a mirror that size. So now you're talking entirely new lens systems with a longer registration distance, a semi-translucent stationary mirror, or the end of the TTL optical viewfinder.

You're on the right track, but it's worse than you think. The mirror is at a 45 degree angle, so to cover a 30.6mm tall sensor, the mirror would need to be a minimum of 43.3mm tall (and clearly have no room to flip up).