Jim Bob

Because your premise is false.

JPL

Hi, it is probably me that Chris is referring to in this topic. But he is not quoting me correctly. I just mentioned that according to DXOmark, very few lenses have more than 30 megapixel resolution and most do less then 20. If DXO are right or not is another story.

Here is a link to the info I am referring to https://www.dxomark.com/lenses/launched-between-1987-and-2017/mount_type-Canon_EF-Canon_EF_S/focal-from-1-to-1500/aperture_max-from-0.95-to-45/launch_price-from-0-to-13000-usd#hideAdvancedOptions=false&viewMode=list&yDataType=rankDxo

JPL

And here is the exact discussion about this between Chris and me in another topic. I hope this will save my reputation in this discussion


ChrisT wrote:
That might explain why EVERY current Nikon Entry-Level body is 24mp, but it doesn't explain why they backed it down to 20mp on the D500 and D7500, nor why the new D850 is 46mp ... or why Canon has embraced 50mp for their 5D S and 5D S R ...

It's all about marketing, I guess ...


Me:
I guess so.

But still an interesting subject. When we think about lenses there are not many that will do more than 30 megapixels according to DXO. (I have no idea if they are accurate or not) and most modern lenses do less then 20 megapixels, even on cameras with 30-40 megapixels sensors. And if it is true that old glass was worse then modern then we could assume that film resolution was limited by lenses with someting less than 20 megapixels and in many cases far less. However putting top end modern glass on film cameras and a high resolution film in the camera could give different results. So probably there is no fixed number here.

jmvaugh Loc: Albuquerque

Poor You! I hope you recover fully and can keep from becoming too bored reading and researching all your favorite topics and hobbies! Yes-getting old is hell but it beats the alternative.
Take care!

jmvaugh Loc: Albuquerque

I found this article from 2015 - not definitive but worth thinking about for those buying higher resolution full frame cameras https://fstoppers.com/originals/are-your-lenses-suddenly-obsolete-85888
And I’ll check out JPLs links.

mwsilvers Loc: Central New Jersey

That statement is inaccurate. Different lenses have different resolving power. Many older or less state-of-the-art lenses can't take full advantage of higher resolution sensors. For instance the resolving power of the Canon EFs 18-200 zoom lens is very significantly lower than the Canon EF 24-70mm L USM II. There is a reason that this shorter focal length zoom costs almost 3 times as much. For that 18-200 lens, anything over a 15mp to 18 mp sensor would be overkill. If you research lens resolution tests you will see how many lenses compare.

Chris T Loc: from England across the pond to New England

Bill, Terry just confirmed it was JPL ...

Chris T Loc: from England across the pond to New England

Thanks for confirming all that, JPL ... now, everyone can read your source for themselves ....

Chris T Loc: from England across the pond to New England

Thanks for your input, Mark ...

jcboy3

The response of "I guess so" to the statement that "It's all about marketing" did not help, that's for sure.

First, it is not the case that 35mm lenses are not capable of resolving greater than 20mp.

Discernable resolution is SIGNIFICANTLY greater, much greater than even the current mega-MP bodies. Resolution vs contrast is a continuum; it's what is represented by MTF curves. By convention, an MTF of 50% is used to define the limit of perceived image sharpness, but MTF levels as low as 2-5% are distinguishable, and an MTF of 9% corresponds with the Rayleigh diffraction limit. Note that film ratings usually went beyond 30% MTF, typically with twice the resolution of the 50% MTF level.

So, even if you just consider 50% MTF as defining your resolution limit, there are several lenses that exceed that limit. And, the 50% MTF is an arbitrary cutoff over a continuum that extends well beyond that limit, easily by as much as a factor of 2 in lines of resolution, and a factor of 4 in number of pixels.

Second, resolution is dependent on a number of factors, including aberrations, noise, dynamic range. Some of these are lens dependent, some sensor/body dependent. This is the basis for the DXOMark measurement of perceptual megapixels, P-Mpix.

Consider the following P-Mpix ratings for the Nikon AF-S NIKKOR 50mm f/1.8G on the following Nikon bodies:

Nikon D500 (20mp) 13
Nikon D7100 (24mp) 14
Nikon D5 (20mp) 15
Nikon D750 (24mp) 17
Nikon D810 (36mp) 22

P-Mpix increases with pixel density for same sensor, and with sensor size. The relationship is approximately proportional, but might be subject to deminishing returns with increasing pixel density. However, there is no indication in these results that a higher full frame pixel count won't result in an increase in perceived sharpness. The diffraction limit will eventually make increased pixel density of insignificant value, but the increase to 46 or 50 mp still produces an increase in image quality. If it's not proportional, then that indicates that we might be approaching a limit.

Note that the 50mm lens I chose because it's been tested on most Nikon bodies. But it is not the sharpest lens in the tool shed. So consider the Nikon D810. There are lenses that test up to 35 P-Mpix (the Carl Zeiss Apo Planar T* Otus 85mm F14 ZF.2 Nikon). That lens is going to rock on the D850. Nikon lenses that test high on the D810 are

Nikon AF-S Nikkor 200mm f/2G ED VR II - 33 P-Mpix
Nikon AF-S Nikkor 400mm F2.8G ED VR - 33 P-Mpix
Nikon AF-S Nikkor 300mm f/2.8G ED VR II - 32 P-Mpix
Nikon AF-S Nikkor 85mm f/1.4G - 30 P-Mpix
Nikon AF-S Nikkor 28mm f/1.8G - 28 P-Mpix

So, there are several lenses throught the focal length range that have resolutions close to the pixel count of the D810. These lenses are likely to produce significantly improved results with a higher pixel count body like the D850.

For full frame cameras, the increase in pixel count is likely to produce significant improvement with the best lenses, and even noticeable improvement with "simply good" lenses. If you are shooting with lower quality kit lenses, you might not see any improvement because you are at the limit on those lenses.

As far as Nikon backing down on the pixel density of their APS-C sensors; there are a number of factors to consider, such as noise, readout and processing speeds. The D5, D500 and D7500 are all ~21mp sensors, and feature fast readout speeds and burst modes. They might be using similar electronics; I don't know. But the ability to crank frames is going to be limited by pixel count.

Chris T Loc: from England across the pond to New England

Thanks for the link, joer ... bringing it up, now ...

Chris T Loc: from England across the pond to New England

John ... thanks very much for your very thorough contribution to this post. Now, then - your last point - has got me wondering. You seem to be indicating there's a direct correlation between stated res / MP count, and fps. Did I read you, right?

BHC Loc: Strawberry Valley, JF, USA

And that you took his information out of context.

E.L.. Shapiro Loc: Ottawa, Ontario Canada

This is kind of an old story- lens resolution, that is! Back in the film days there were lenses that boasted incredible resolution performance that was probably measured or evaluated with some sort of optical bench/electronic laboratory device. Problem was, most medium and high speed films could not resolve what some of theses lenses could put out. Many of theses lenses seemed to come into their own with slower speed films such as Panatomic-X and Adox KB-14. and of course, our old friend Kodachrome 25. Theses films were oftentimes just too slow for certain purposes and the mentioned black and white emulsions exhibited higher than desired contrast even with normal development. The black and white films also need to be CAREFULLY processed in fine grain developers. If there were temperature differentials between the baths with its accompanying emulsion swelling and accentuation of grain the apparent would suffer even more. When Kodak introduced it last generation of T-Grain films, I got to better appreciate my Hasselblad gear.

As a commercial photographer, I spent many years shooting large format, using crazy expensive lenses and films with the highest acuteness available and processed them in a very CLINICAL manner with precise temperature contrl, extrmrrly gentle adgitation and water filtration. All this in the name of SHARPNESS! Thing was, many of my images ended up on billboards, fast-food menu signage and displayed on the
sides of buses and transport trucks.

Nowadays, with digital photography, I just stopped worrying about pixel counts, resolution specifications and all of that kind thing. I have used both Nikon and Canon gear, full frame and cropped bodies and have never had serious issues with resolution. OK- admittedly, when I first made the transition into digital. I bit the bullet and digitized my Mamiya RZ67 and purchased a perspective control/tilt and swing rig for its front end. Unless a disproportionate degree of enlargement is required, nowadays, I do most of my work on my Canon DSLRs.

Too many photographers become preoccupied with some of theses specs which is kind of funny because so many of them are viewing most of their images on a small monitor or only making making small to medium sized prints.

daddybear Loc: Brunswick, NY

Welcome to the 70's club. I joined it back in July. Parts missing, parts replaced and parts swapped out. Can' t walk very far but I can still push a shutter when I get there

Regards,

DeanR.