Canon EOS 7D Mark II Spec List [CR2]

[Marsu42]

If you want the best information available, CR (the site, not necessarily the forum) is pretty much the place.

Ok, I can subscribe to that view :-)

I don't have anything new to add, but rumor has it (haha!) that Canon is not entirely oblivious to items on this website.

Where did you get this rumor from - CR :-> ? I think the decisive Canon people are in Japan and a small circle of execs (like the CEO in a recent interview) and maybe some tech people, and I would be very surprised if any information from places like this would reach them - most likely they rely on their internal marketing studies and some direct feedback from cps professionals and beta testers.

[DanielW]

Where did you get this rumor from - CR :-> ? I think the decisive Canon people are in Japan and a small circle of execs (like the CEO in a recent interview) and maybe some tech people, and I would be very surprised if any information from places like this would reach them - most likely they rely on their internal marketing studies and some direct feedback from cps professionals and beta testers.

How harmful would it be for them to pay a small team of photo nerds to surf the web all day and find out what buyers want?
(Heck, I'd do it for free if they let me try some prototypes every now and then.)

[Lee Jay]

This camera would have a pixel density equal to a 61mp full frame camera, that is far beyond the resolving power of most lenses.

Grrrrr....would people quit saying entirely wrong stuff like that please?  First of all resolving power doesn't work like that.  Second, even if it did the better lenses can already resolve up into the many hundreds of megapixels on full frame.

Yes, but diffraction softness at this pixel density starts to become a problem, get up to f/5.6 or higher and you start loosing sharpness.

Diffraction is the most misunderstood concept in photography.

Oh, I don't know, DOF is up there pretty good too!

The notion that diffraction is ever a "problem" is just flat out wrong.

Diffraction is always a factor in resolving power.  When I'm shooting planetary photography, diffraction tends to be the driving factor, and it most certainly is a problem.  I wouldn't have just upgraded from a 5" to an 11" scope if it weren't!

Just because diffraction starts earlier with smaller pixels does NOT mean you are resolving less detail.

"Just because diffraction starts earlier with smaller pixels does NOT mean you are resolving less detail."

I just thought that was important enough for a repeat.

I really have to emphasize this: In no way, ever, can diffraction produce worse results on a higher resolution sensor than a lower resolution sensor. EVER.

That one as well:

"I really have to emphasize this: In no way, ever, can diffraction produce worse results on a higher resolution sensor than a lower resolution sensor. EVER."

[unfocused]

Where did you get this rumor from - CR :-> ? I think the decisive Canon people are in Japan and a small circle of execs (like the CEO in a recent interview) and maybe some tech people, and I would be very surprised if any information from places like this would reach them - most likely they rely on their internal marketing studies and some direct feedback from cps professionals and beta testers.

How harmful would it be for them to pay a small team of photo nerds to surf the web all day and find out what buyers want?
(Heck, I'd do it for free if they let me try some prototypes every now and then.)

Very harmful.

The data would not be simply useless, it would be misleading garbage.

Market research is a science. Canon has decades of hard data based on its sales figures. In addition, everyone who fills out a warranty card goes into a data base that Canon can mine for additional information. They scour competitors annual reports and sales figures, just as the competitors scour theirs.

The data is supplemented with surveys that adhere to strict standards to assure that they are polling a representative sample of customers. Those surveys are no doubt conducted on every continent to study trends in each region. They also likely use focus groups of key customer subsets to flesh out the data. They use field testers to try products out in the real world and give them feedback.

A basic rule of any polling is that you never, never use subjects that self-select. They are notoriously unreliable and not representative of the larger universe. Just one tiny example shows how fruitless it would be: there is zero ability to match up the data on this website with actual buying patterns. There is nothing that could assure Canon that the whining, trolling and fantasizing that goes on on internet forums has any relationship whatsoever to sales.

At most, they may employ robots to search the internet for key words and associations. The best you can hope for, is that if Canon finds a spike in the internet on some word string, like: "5DIII, Weak Focus, Low Light " they might investigate it further to see if there is a trend or problem developing.

But, it is crazy to think Canon or any other manufacturer is going to waste valuable resources trolling internet forums.

[Marsu42]

How harmful would it be for them to pay a small team of photo nerds to surf the web all day and find out what buyers want?

Because people posting in forums are not necessarily buyers, or at least there is hardly any connection between internet opinion and sales, the latest example is the underwhelming but popular 6d.

What happens if you ask buyers about what they want you can as well file for bankruptcy right away like when Homer's brother let him design a car: http://simpsons.wikia.com/wiki/The_Homer

[Plamen]

I really have to emphasize this: In no way, ever, can diffraction produce worse results on a higher resolution sensor than a lower resolution sensor. EVER.

That one as well:

"I really have to emphasize this: In no way, ever, can diffraction produce worse results on a higher resolution sensor than a lower resolution sensor. EVER."

Right. I do not know why this obsession with diffraction and not with lens aberrations, for example. The resolution of a system depends on several factors, and if you improve one of them, you improve the final image. At what point you get "diminished returns" is a philosophical question.

[Radiating]

Just an FYI regarding all this talk of too much resolution.

Tests have shown that even the best APS-C lenses are limited to around 20 megapixels of resolution, regardless though there are some benefits to more pixels, namely:

Better de-mosaicing, ideally each pixel in a photo should have 4 pixels behind it (1 red 2 green and 1 blue) as current photos are done with false color and the bayer pattern naturally limits resolution to around 90% of what it should be.

Less interference from anti-aliasing filters, this is actually a huge deal and can increase lens resolution by 20%.

Better color gradiation and tones.

The disadvantage to more resolution is bigger file sizes (less of an issue with S_RAW (which Nikon inexplicably doesn't have).

24 Megapixels is the perfect resolution for an APS-C camera and 28 megapixels is the perfect resolution for a full frame camera with current lens designs.

[Plamen]

you tell me that this forum  is the best place , but no one have answerered my question,  is this a wish list or is it a reality when it comes to Canons sensors? 24Mp at a APS size and what  is the signal noise ratio with current technology and  the dynamic range compared to others

See this.

[jrbdmb]

How harmful would it be for them to pay a small team of photo nerds to surf the web all day and find out what buyers want?

Because people posting in forums are not necessarily buyers, or at least there is hardly any connection between internet opinion and sales, the latest example is the underwhelming but popular 6d.

What happens if you ask buyers about what they want you can as well file for bankruptcy right away like when Homer's brother let him design a car: http://simpsons.wikia.com/wiki/The_Homer

And another example highlighted above.  Only a small number of people in Canon know whether the 6d is "popular" as far as meeting expected sales figures.  And a similar small number know whether the "underwhelming" view of the 6D is just from crazed internet forums or actually the view of the camera buying public at large.

For example, I think the 60D was under-speced, with the deletion of MFA a big mistake.  I also think the 6D should have had an AF system at least as good as the 60D.  But unless someone at Canon shows unusual candor, we'll never know if they agree with these points, or if they even acknowledge that these issues affected sales of the cameras.

[Lee Jay]

Tests have shown that even the best APS-C lenses are limited to around 20 megapixels of resolution,

Point me to those tests and I'll show you why they're wrong.

[Plamen]

what is it that I should look after?

The QE figures determine the shot noise per unit area. A sensor with 42% QE, for example will have 1/2 stop better noise performance than one with 30%, at base ISO, at least. (Ratio of 2 is one stop). This is valid for comparisons within the same format.

Then there is data for the read noise, this determines the deep shadow noise.

I know Bob,he is a friend of mine , who has this site.

I read some of his posts elsewhere, he is a smart and funny guy.

[Plamen]

yes I know what QE is , and ?

You asked about comparisons to others. They are a lot of them there. The newest Canon FF sensors are comparable to the Nikon ones in terms of QE. The current (a bit old already) crop sensors are at 40% or so, so they lag a bit behind. What I expect is to see a new crop sensor with QE around 55% or so. This would be about 1/2 stop better than the current 7D.

I have no idea what to expect about the read noise.

[Marsu42]

Only a small number of people in Canon know whether the 6d is "popular" as far as meeting expected sales figures.  And a similar small number know whether the "underwhelming" view of the 6D is just from crazed internet forums or actually the view of the camera buying public at large.

Since it is obvious we don't have that data (I guess even Canon doesn't have a representative overview on the worldwide opinion in different buyer segments on the 6d) I was talking from my personal perspective - and according to the initial reactions to the 6d (mine included) there shouldn't be anyone out there buying it, but obviously a lot of people are nevertheless (me included :-)).

For example, I think the 60D was under-speced, with the deletion of MFA a big mistake.

I also though so but stood corrected - removing afma from the 60d was a very smart move by Canon, lots of people bought the more expensive 7d.

[jrista]

Tests have shown that even the best APS-C lenses are limited to around 20 megapixels of resolution,

Point me to those tests and I'll show you why they're wrong.

I'd like to see these tests as well, Radiating. The Nikon D3200 is already a 24mp APS-C, and it produces some amazing quality, and its pixels are in the 3 micron range. Small form factor sensors have pixels under 2 microns now.

If one wants to talk about finite limits on pixel size, I'd offer around 800nm, or 0.8µm. Any smaller than that, and you are going to start clipping frequencies of visible light (starting in the deep reds)...but we are a LONG way off from sub-micron pixels in larger form factor sensors.

[DanielW]

Where did you get this rumor from - CR :-> ? I think the decisive Canon people are in Japan and a small circle of execs (like the CEO in a recent interview) and maybe some tech people, and I would be very surprised if any information from places like this would reach them - most likely they rely on their internal marketing studies and some direct feedback from cps professionals and beta testers.

How harmful would it be for them to pay a small team of photo nerds to surf the web all day and find out what buyers want?
(Heck, I'd do it for free if they let me try some prototypes every now and then.)

Very harmful.

The data would not be simply useless, it would be misleading garbage.

Market research is a science. Canon has decades of hard data based on its sales figures. In addition, everyone who fills out a warranty card goes into a data base that Canon can mine for additional information. They scour competitors annual reports and sales figures, just as the competitors scour theirs.

The data is supplemented with surveys that adhere to strict standards to assure that they are polling a representative sample of customers. Those surveys are no doubt conducted on every continent to study trends in each region. They also likely use focus groups of key customer subsets to flesh out the data. They use field testers to try products out in the real world and give them feedback.

A basic rule of any polling is that you never, never use subjects that self-select. They are notoriously unreliable and not representative of the larger universe. Just one tiny example shows how fruitless it would be: there is zero ability to match up the data on this website with actual buying patterns. There is nothing that could assure Canon that the whining, trolling and fantasizing that goes on on internet forums has any relationship whatsoever to sales.

At most, they may employ robots to search the internet for key words and associations. The best you can hope for, is that if Canon finds a spike in the internet on some word string, like: "5DIII, Weak Focus, Low Light " they might investigate it further to see if there is a trend or problem developing.

But, it is crazy to think Canon or any other manufacturer is going to waste valuable resources trolling internet forums.

It does make sense, indeed.
Thanks for the great answer!