Canon Confirms Development of High Megapixel Camera

[scyrene]

The original premise of the thread is the higher MP speculated on for a Canon EOS 135 format sensor, it was subsequently suggested that current APS-C sensors have similar, though lower, pixel densities to the rumours. This in turn led to the fact that even those crop camera pixel density 'resolution advantages' are practically impossible to realise in real world shooting except for a very few specific situations. As same generation crop and ff cameras use very similar technology and, very importantly, the same lenses, we can get real practical advance knowledge of what to expect from a next generation high MP sensor.

If one does actually compare same generation crop and cropped ff sensors when mitigating as many variables as possible, ie lenses, support, etc etc and use the optimal processing for both files I have never seen a set of comparison images that shows a marked crop camera resolution advantage (and I am far from alone). If somebody wants to point me to some then have at it, I am especially looking forward to a set of 5D MkIII and 7D MkII files.

That is the only relevant and practical comparison to make when trying to gauge what we might achieve with a 35-50MP 135 format sensor, especially considering it will be using those same EF lenses. Going off at tangents claiming all kinds of things that nobody disputes is just a smokescreen. Is an iPhone 'better' than a H5? Well it does resolve more per sensor area, nobody disputes that, but it doesn't use the same sensor tech or lenses and it can't take the same images, it is an irrelevant comparison.

Without wanting to wade into a row, and admitting I know very little about all this technical stuff, would a relevant study be to use the same lens on successively higher resolution sensors of the same size? The original 5D versus the 5D2, for instance. If you did it with enough resolutions, could you plot the increase in actually resolved detail, and make a prediction for as yet unreleased resolution sensors? I imagine the hypothesis is, the higher you go, the smaller the increase in actual resolution per extra megapixel?

 

[ajfotofilmagem]

The tests comparing different resolutions of sensors, but size, lenses and all else being equal, show diminishing returns sharpness by increasing the amount of megapixel.

Everyone here agrees that more megapixel is useful when you need crop to adjust the frame. But most of us agree that to enjoy dramatic increases of megapixel requires better lenses.

The great mistake of Nikon to launch the 36 megapixel sensor was the short supply of lenses capable of approaching the theoretical sharpness of the sensor.

Canon (and Sigma Art) are renewing their line of lenses to allow effective use 50 megapixel, but we do not know until limit will continue gaining real sharpness in full frame sensors. Maybe 100 megapixel?

 

[scyrene]

The tests comparing different resolutions of sensors, but size, lenses and all else being equal, show diminishing returns sharpness by increasing the amount of megapixel.

Everyone here agrees that more megapixel is useful when you need crop to adjust the frame. But most of us agree that to enjoy dramatic increases of megapixel requires better lenses.

The great mistake of Nikon to launch the 36 megapixel sensor was the short supply of lenses capable of approaching the theoretical sharpness of the sensor.

Canon (and Sigma Art) are renewing their line of lenses to allow effective use 50 megapixel, but we do not know until limit will continue gaining real sharpness in full frame sensors. Maybe 100 megapixel?

For bird photographers who want to crop more (like me!), the super telephoto primes are probably well placed to take advantage of extra resolution, right?

 

[Loswr]

The question is whether dividing the same sensor area into lots of smaller pixels or a smaller number of larger pixels will result in better resolution. Some people above erroneously claim that there will be little or no difference for various reasons (technique, lens sharpness, whatever). So, that one is about the same focal length and subject distance just with a different number of pixels (see the title of the thread).

There will be more pixels on the subject. Whether those additional pixels result in significant additional measurable spatial resolution depends on many factors. Practically speaking, the increased resolution generally falls well short of what one would extrapolate from the difference in pixel size, in many cases so far short as to result in no significant gain from the smaller pixels (in other words, the gain is small enough to not be readily observable in everyday shooting/viewing).

 

[Lee Jay]

The question is whether dividing the same sensor area into lots of smaller pixels or a smaller number of larger pixels will result in better resolution. Some people above erroneously claim that there will be little or no difference for various reasons (technique, lens sharpness, whatever). So, that one is about the same focal length and subject distance just with a different number of pixels (see the title of the thread).

There will be more pixels on the subject. Whether those additional pixels result in significant additional measurable spatial resolution depends on many factors. Practically speaking, the increased resolution generally falls well short of what one would extrapolate from the difference in pixel size, in many cases so far short as to result in no significant gain from the smaller pixels (in other words, the gain is small enough to not be readily observable in everyday shooting/viewing).

It should only fall far short or have "no significant gain" if one of three things is going on - horrible lens quality, lousy technique, or poor atmospheric seeing conditions.

It's simple enough to calculate the gain if you know the lens performance. Unfortunately no one site I know of does lens tests, they only do system tests so we don't have lens data.

To resolve that issue, one can stress test the lenses using teleconverters. Teleconverters effectively do the exact same thing as shrinking the pixels. If teleconverters produce increased spacial resolution, then so would smaller pixels. If not, not. Here's a test I did long ago on that topic. All images shot with the same lens from the same location. So, did the teleconverters add resolution? It looks to me like they did.

 

[Loswr]

The question is whether dividing the same sensor area into lots of smaller pixels or a smaller number of larger pixels will result in better resolution. Some people above erroneously claim that there will be little or no difference for various reasons (technique, lens sharpness, whatever). So, that one is about the same focal length and subject distance just with a different number of pixels (see the title of the thread).

There will be more pixels on the subject. Whether those additional pixels result in significant additional measurable spatial resolution depends on many factors. Practically speaking, the increased resolution generally falls well short of what one would extrapolate from the difference in pixel size, in many cases so far short as to result in no significant gain from the smaller pixels (in other words, the gain is small enough to not be readily observable in everyday shooting/viewing).

It should only fall far short or have "no significant gain" if one of three things is going on - horrible lens quality, lousy technique, or poor atmospheric seeing conditions.

It's simple enough to calculate the gain if you know the lens performance. Unfortunately no one site I know of does lens tests, they only do system tests so we don't have lens data.

To resolve that issue, one can stress test the lenses using teleconverters. Teleconverters effectively do the exact same thing as shrinking the pixels. If teleconverters produce increased spacial resolution, then so would smaller pixels. If not, not. Here's a test I did long ago on that topic. All images shot with the same lens from the same location. So, did the teleconverters add resolution? It looks to me like they did.

For me, 'everyday shooting/viewing' doesn't comprise tripod mounted static subjects cropped to 100%. If that's your usual method/subject, then bravo – your results have validity as far as comparing teleconverters vs. pixel interpolation for increased resolution, which is certainly not the topic at hand.

 

[erjlphoto]

Would be very surprised if Canon goes with a Sony sensor.
I appreciate that Canon makes their own rather than buying them from someone else.

 

[Lawliet]

For me, 'everyday shooting/viewing' doesn't comprise tripod mounted static subjects cropped to 100%. If that's your usual method/subject, then bravo – your results have validity as far as comparing teleconverters vs. pixel interpolation for increased resolution, which is certainly not the topic at hand.

Once you understand that this topic is about a high resolution camera and how flash photography works you'll realize why this is relevant for everyday usage.
That's even if we disregard the quality of the average column stand.

 

[Lee Jay]

For me, 'everyday shooting/viewing' doesn't comprise tripod mounted static subjects cropped to 100%. If that's your usual method/subject, then bravo – your results have validity as far as comparing teleconverters vs. pixel interpolation for increased resolution, which is certainly not the topic at hand.

You don't need a tripod and static subjects to get little or no relevant movement during an exposure. Good technique, fast shutter speeds, and IS can all combine to get pixel-level performance equal to the best you can get in the lab, and I do it regularly. I shoot a lot of airplanes and often have final images that are 1:1 pixel crops from a crop camera with a 2x teleconverter mounted.

 

[NancyP]

I'd be happy with a 6D-like camera with the current D810 sensor characteristics (MP and DR), to be used for landscapes - for that matter, I would be happy with a 6D, same 20MP, better DR.

Most local serious amateur bird photographers using supertele lenses use Canon bodies, with the expected combinations along the line of cost: FF with 400, 500, 600 mm f/4 primes; APS-C with 400 f/5.6L (that's me!), 100-400 f/variable L IS, various third party zooms reaching 400 mm or longer. Sure, APS-C is a compromise. Perhaps with the advent of the inexpensive Tamron zoom, more Nikon shooters will give bird photography a try.

A lot of MF users state that the color fidelity and subtlety is better with MF than FF, so MF users aren't just going after high numbers of pixels.

 

[Loswr]

For me, 'everyday shooting/viewing' doesn't comprise tripod mounted static subjects cropped to 100%. If that's your usual method/subject, then bravo – your results have validity as far as comparing teleconverters vs. pixel interpolation for increased resolution, which is certainly not the topic at hand.

Once you understand that this topic is about a high resolution camera and how flash photography works you'll realize why this is relevant for everyday usage.
That's even if we disregard the quality of the average column stand.

That will be of great benefit to the <4% of my shots which use flash. Woot. Granted, the majority of Rebel shooters use flash a lot more frequently. I love watching popup flashes firing when the subject far outside the effective flash range...

 

[Lawliet]

That will be of great benefit to the <4% of my shots which use flash.

Well, some consider shutter count a sensible metric - taping down the button will make you great really fast.
OTOH flash wins hands down in terms of income generated. Followed by the combination movie lighting and solid stands. To sad that getting things right the first time doesn't create impressive file numbers...can't have everything after all.

 

[sdsr]

The great mistake of Nikon to launch the 36 megapixel sensor was the short supply of lenses capable of approaching the theoretical sharpness of the sensor.

Why was that a "great mistake"? Regardless of whether lenses "capable of approaching the theoretical sharpness of the sensor" would make a significant difference, higher resolution sensors are also appealing if other "inferior" lenses perform better on such sensors than they do on lower resolution sensors. In my experience they do (though obviously I've not used a 50Mp ff sensor, only 12-36 Mp ff sensors), but my experiences are merely anecdotal and thus aren't likely to be of much interest to anyone else. But Roger Cicala has done some testing that's relevant here, including his conclusion, where among other things he points out that while the Canon 24-70 II outresolves its Tamron equivalent, the difference more-or-less vanishes when you compare Canon 24-70 on 5DIII vs Tamron on D800 (I misdescribed his conclusion in an earlier response to someone else):

http://www.lensrentals.com/blog/2013/01/a-24-70mm-system-comparison

The Canon on a higher Canon Mp body would presumably trounce Tamron on a D800. Of course, whether the performance of "lesser" lenses matters will depend on who you are, how you see, what your standards are, etc., etc.

 

[Sporgon]

For me, 'everyday shooting/viewing' doesn't comprise tripod mounted static subjects cropped to 100%. If that's your usual method/subject, then bravo – your results have validity as far as comparing teleconverters vs. pixel interpolation for increased resolution, which is certainly not the topic at hand.

You don't need a tripod and static subjects to get little or no relevant movement during an exposure. Good technique, fast shutter speeds, and IS can all combine to get pixel-level performance equal to the best you can get in the lab, and I do it regularly. I shoot a lot of airplanes and often have final images that are 1:1 pixel crops from a crop camera with a 2x teleconverter mounted.

You are so wrong there. Granted you may; or may not. For the bayer pattern of pixels to describe everything accurately any microscopic movement and your four three colour arrays will receive confused information. Frequently the data from a hand held shot can looked clogged up - if you're going to be really picky about it, and that is infantisimal movement. IS does not produce the same data as a genuinely stable shooting platform, and remember just because it's on a tripod doesn't mean it's totally stable.

So when you say "good technique, fast shutter + IS = as good as in the lab" ( by the 'lab' I presume you mean rock steady platform etc etc.) then that statement is both misleading and wrong.

 

[dgatwood]

For landscapes, if you really need a high-MP landscape photo, you can save a few thousand bucks by just stitching multiple shots together, because landscapes aren't moving.

Lol. Do you ever look at nature? Oceans, rivers, wind, lots of stuff moving in nature.

Yes, and nature photography also includes wildlife, which screams for a good AF system.

Like I said, I struggle to think of a category of photographer who would desperately want higher resolution, but who would settle for a subpar AF system.

Yes, if you believe that landscapes don't move, I see why you struggle. They move, but not in a way that requires anything more than manual or contrast-detect AF. Or perhaps you're suggesting that landscape photographers don't constitute a 'category of photographers'?

I don't believe that pure landscape photographers are the ones screaming for high megapixel count, for the most part. The photographers who want high megapixel count are mostly either people who shoot candids (where you want maximum ability to crop because Canon has no readily carried superzooms) or people who do two radically different kinds of shooting, e.g. landscapes and birds. In both of those categories, the reach-limited photographer is likely to want both high megapixel count and good AF.

I maintain the correctness of my original statement. Landscapes don't move. They may have some elements that move, like rivers, leaves, etc., but the scene as a whole does not move, and even the elements of the shot that do move usually do not move much. Thus, stitching is generally an acceptable way to get higher resolution if you really need it (unlike, for example, photographing people, animals, moving cars and so on). Is it perfect? No. Is it good enough? Usually.

 

[PureClassA]

Damn...I need to make more popcorn...

I haven't really shot on my 7D in a while, but all this debate is making me want to dust it off again. 6D vs 7D pixels. Looking at some of my 8x10 portrait prints from a few years ago on the 7D with an old 172E Tamron 90mm Macro produced some very lovely and well detailed results.

Looking forward to 2015 for Canon. I rent the 1DX for dance recitals anyway. If they make a great high MP camera with 11pt AF (a la 6D) I'd probably get it because most of what I do is portrait anyway.

 

[Loswr]

That will be of great benefit to the <4% of my shots which use flash.

Well, some consider shutter count a sensible metric - taping down the button will make you great really fast.
OTOH flash wins hands down in terms of income generated. Followed by the combination movie lighting and solid stands. To sad that getting things right the first time doesn't create impressive file numbers...can't have everything after all.

When did I mention numbers? Do you require remedial math education to learn the meaning of the percentage (%) symbol?

Also, since I don't generate income from my images, that's a useless metric as far as I'm concerned. Then, there's this thing called natural light, I like it and can certainly 'get it right the first time' using it...but I understand it's not for everyone.

 

[7enderbender]

That's all fine. Unless that's all they can come up with. I personally still don't get the high resolution thing other than for the ability to crop or to display things billboard size. Other than that printing technology is so far behind that I don't see very little value in anything over, say, 12MP. Still see no need to give up on my 5DII in fact.

 

[PureClassA]

That's all fine. Unless that's all they can come up with. I personally still don't get the high resolution thing other than for the ability to crop or to display things billboard size. Other than that printing technology is so far behind that I don't see very little value in anything over, say, 12MP. Still see no need to give up on my 5DII in fact.

If your 5D2 is giving you (and your clients, if for hire) everything you want, then no, you are correct sir. That said, like all things, something new comes out and you try it and then wonder how you ever got by with the old one ;-) Doesn't make the 5D2 a bad camera all the sudden though.

 

[erjlphoto]

That's all fine. Unless that's all they can come up with. I personally still don't get the high resolution thing other than for the ability to crop or to display things billboard size. Other than that printing technology is so far behind that I don't see very little value in anything over, say, 12MP. Still see no need to give up on my 5DII in fact.

To the "masses" a higher megapixel count means a better picture. They understand no other standard by which to judge a camera. If you are just showing your pics on an iPad or computer the extra pixels won't make a difference, but the casual user will feel gratified they are there none the less.

Basically, some of the appeal will be to increase sales, I doubt most photographers will find just extra pixels of much help.

Hopefully the new camera brings more to the table than increased number of pixels.