Really “Is it the end of M road”?

[lct]

Advertisement (gone after registration)

Indeed relevance is a subjective concept. I don't need more than 12MP personally but this adds very little to the discussion. Fact is that many if not all digital photogs view their images on screen at 100% magnification. More pixels can only enhance, never reduce resolution. 

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[me111]

I don't understand why it's difficult to understand that blowing up images without any information, you don't get more information/resolution. 10000000 mp won't enable us to see microorganism. It's just going to give a lot of redundant (or corrupt) information because of the diffraction of the lens.

 

Sony promises that it will make lenses that can optimize its sensor. Doesn't it mean that Sony admits of its high-mp sensor's overkill?

[jaapv]

I fear that such a simplified argument won't cover the subject.

[01af]

I don't understand why it's difficult to understand that blowing up images without any information ...

That's not what is happening. There is information. I don't understand why it's so difficult to grasp the principle that both lens and sensor together limit the final image's resolution.

 

I guess in most laymen's heads there is the naïve notion of the image's resolution being equal to the minimum of the resolutions of the lens and the sensor. If the two don't happen to be equal then the extra resolution of the stronger component should be wasted on the weaker component. But that's not how things work.

 

To get a basic understanding how the resolutions of lens and sensor work together to form an image, go back to post #413 (and following) in this thread.

[adan]

Just to put some raw numbers into that equation for system resolution, a lens producing 80 lppm will produce an image resolution on a sensor, of

 

24 Mp (6 micron pixels) - 53.33 lppm

42 Mp (4.5 micron pixels) - 58.17 lppm

48 Mp (4.2 micron pixels) - 59.12 lppm

96 Mp (3 micron pixels) - 64 lppm

 

Whether that is noticeable or relevant will be up to the user.

 

Personally, I am seeing resolutions (fine details) from the M10 that outperform what the equation predicts. I can see hairs reproduced one pixel (1/167th mm) wide in some pictures. I won't theorize as to why that occurs - gets too deep into information theory. I don't doubt that it would extend to sensors with more pixels, up to a point (and I won't guess where that point is).

[01af]

... a lens producing 80 lp/mm ...

Our lenses are much better than that. If there's a single feature standing out from a featureless background with enough contrast, such as a single hair in a bright streak of light, then lenses can resolve detail that corresponds to 200 lp/mm and beyond. That doesn't necessarily mean the same lens could resolve a pattern of multiple parallel hairs, each a hair's width apart from the next. Anyway—specifying an absolute resolution limit for a lens in terms of one single scalar number is not so easy, to say the least. Maybe it's impossible, I'm not sure. When lens tests or data sheets specify a number then it always holds only under very specific conditions and prerequisites but mustn't be mistaken for an hard, unconditional, and absolute limit.

[mmradman]

Advertisement (gone after registration)

Just to put some raw numbers into that equation for system resolution, a lens producing 80 lppm will produce an image resolution on a sensor, of

 

24 Mp (6 micron pixels) - 53.33 lppm

42 Mp (4.5 micron pixels) - 58.17 lppm

48 Mp (4.2 micron pixels) - 59.12 lppm

96 Mp (3 micron pixels) - 64 lppm

 

Whether that is noticeable or relevant will be up to the user.

 

Personally, I am seeing resolutions (fine details) from the M10 that outperform what the equation predicts. I can see hairs reproduced one pixel (1/167th mm) wide in some pictures. I won't theorize as to why that occurs - gets too deep into information theory. I don't doubt that it would extend to sensors with more pixels, up to a point (and I won't guess where that point is).

Being pedantic here, please check sensor figures, doubling sensor pixel increases linear resolution by sqrt(2)=1.41.

[otto.f]

Isn’t this issue precisely analogue to the difference in sharpness between photo’s taken with the same lens on either Technical Pan or Tri-X (apart from the grain)?

[me111]

That's not what is happening. There is information. I don't understand why it's so difficult to grasp the principle that both lens and sensor together limit the final image's resolution.

 

I guess in most laymen's heads there is the naïve notion of the image's resolution being equal to the minimum of the resolutions of the lens and the sensor. If the two don't happen to be equal then the extra resolution of the stronger component should be wasted on the weaker component. But that's not how things work.

 

To get a basic understanding how the resolutions of lens and sensor work together to form an image, go back to post #413 (and following) in this thread.

 

Like you, I think the images from larger number of MP sensor will continue to be cleaner past the point that the sensor out-resolves the lens. However, like adan (I guess,) at the same point, the resolution will start to get deteriorated dramatically until one point that all you get is smeared information when we enlarge it further.

 

The comparison demonstrates the effect of lens diffraction, and diffraction (and aberration) limits lens resolution. Look closely at the edge of any circle, at some point, you cannot see the black background with any number of MP. Do you think MP count has a meaningful relevance to the IQ without limitation???

 

 

 

credit: https://www.shutterbug.com/content/here%E2%80%99s-everything-you-need-know-about-how-lens-diffraction-affects-your-photography-video (access 08/17/2008)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[me111]

If there's a single feature standing out from a featureless background with enough contrast, such as asingle hair in a bright streak of light, then lenses can resolve detail that corresponds to 200 lp/mm and beyond.

I believe that regardless of the contrast, if the hair is small enough and the diffraction is high enough, no. The light rays reflected on the hair will be focused elsewhere (not on the sensor.) All you can see is the "bright streak of light," with any number of MP.

[Chaemono]

Let's show some with a 1955 collapsible Summicron. The lens is in a good condition and is absolutely gorgeous. I took some pictures of distant street signs at f/4.0 as someone here suggested and, yes, the 42 MPx will show a bit more detail. I'll post those later. At f/2.0 this lens somehow performs better on the M10 even though the Sony may have a bit more detail in the center. I tried to match color and temperature a bit, adjusted the Sony for vignetting, applied equal amount of sharpening to both, tried to match exposure. But there is something about the look this lens creates on the M10 that appeals more to me than what I see in the Sony pictures. It could be that the Leica pictures look less digital, more organic. Anyway, these are just the first couple of test shots wide open. I'll do more over the next few of weeks.

 

Less compressed JPEGs here: https://www.smugmug.com/gallery/n-8pL8zB/

 

α7R III + Collapsible Leica Summicron 5cm f/2

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 200 f/2.0 @1/500 sec.

Edited August 17, 2018 by Chaemono

[Chaemono]

Less compressed JPEGs here: https://www.smugmug.com/gallery/n-8pL8zB/

 

M10 + Collapsible Leica Summicron 5cm f/2

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 200 f/2.0 @1/250 sec.

Edited August 17, 2018 by Chaemono

[Chaemono]

And now the crops.

 

Less compressed JPEGs here: https://www.smugmug.com/gallery/n-8pL8zB/

 

α7R III

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 200 f/2.0 @1/500 sec.

[Chaemono]

Less compressed JPEGs here: https://www.smugmug.com/gallery/n-8pL8zB/

 

M10

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 200 f/2.0 @1/250 sec.

[Chaemono]

Less compressed JPEGs here: https://www.smugmug....llery/n-8pL8zB/

 

α7R III + Collapsible Leica Summicron 5cm f/2

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 200 f/2.0 @1/640 sec.

[Chaemono]

Less compressed JPEGs here: https://www.smugmug....llery/n-8pL8zB/

 

M10 + Collapsible Leica Summicron 5cm f/2

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

ISO 200 f/2.0 @1/350 sec.

 

[Chaemono]

And now the crops.

 

Less compressed JPEGs here: https://www.smugmug....llery/n-8pL8zB/

 

α7R III

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Chaemono]

Less compressed JPEGs here: https://www.smugmug....llery/n-8pL8zB/

 

M10

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Advertisement]

Hallo,

Look here: Leica M

[Tailwagger]

Eventually, when Leica produces M camera with more than 24Mp full frame sensor all this will be forgotten and new "24+" will become new standard. 

In the meantime notion that more than 24Mp can be better for M photography than current 24Mp, or photography in general, seems to be fought like life depends on it.

 

I dont think there's much doubt in any quarter that Leica will up the pixel count out some point, assuming there are no inherent physical limitations imposed by the need for specialized microlensing.   The fight, if indeed there is one, is really around the gravity of need for this to happen sooner rather than later and therefore the value of the current offering.  For those in the business of producing gallery prints, I understand the commercial need to print large and be able to withstand a certain level of microscopic scrutiny.  But not being in that business and residing in a home with 53 windows in 2400 Sq Ft, I've neither the desire nor the wall space to print beyond 20x30. I try to remind myself occasionally that the finest landscapes I've ever held at arms length where 8x10s purchased by my father in the 1960's. Hence, outside of the amusement value of discovering elements within the scene that I could not actually see when the shutter was pressed, more MP is only occasionally valuable, at least in my case, for increased cropping flexibility. Not something that's high on my personal list of needs. 

 

It would be nice if a single device could provide perfect accommodation across all of its possible use cases.  But all engineering involves compromise and tradeoffs to serve a specific purpose. If that were not the case, panel vans would compete in F1 on the weekends.  While the M10 is by no means perfect, nor state of the art in any number of dimensions, it still represents the best set of compromises for my particular use case.  Likely for some, this iteration is the end of the road, or at least a 2-3 year obstruction  across the highway, given a strongly held belief that it lags too far behind in technically significant areas.  Yet, despite the occasional wandering eye when a shiny new bauble appears in an alternate manufacturer's stable, the M, given its other charms, has not yet fallen so far behind that its nearing the end of the road with me.  So no fight here, just a recognition that what is crucial for some, is not as gut wrenching for others.

Edited August 17, 2018 by Tailwagger

[IkarusJohn]

Interesting that the rumoured new Nikon Z6 is 24MP, as are the Sony A7 II & III (and the A7S is "only" 12.2).  Granted, both Sony and Nikon (apparently) offer higher MP versions of these cameras, but 24MP is still on offer.

 

I doubt more MP in an M camera is a foregone conclusion - or more accurately, I don't believe that Leica seems more MP as a goal in itself.