M10-R Lens Pairings & Real vs Perceived Benefits

[Rick in CO]

Advertisement (gone after registration)

On 11/25/2021 at 4:30 AM, pk851667 said:

To explain what kind of a range we're talking about - I have found some lenses like the 50 Cron ASPH, 90 Cron v2, and even some very cheap third-party lenses like the TTA 50 1.4 to be outstanding on the camera. But those are quite tight focal lengths, and if you want to maximize the use of the sensor you would want to go wider so you have room to crop. Well, the 28 and 24 Elmarit ASPH have both been pretty bland on the camera. Indeed, when cropping you are bordering on unusable. Any thoughts?

There was a previous discussion relating to Peter Karbe's suggestion that the higher resolution sensors would require lenses that can achieve at least 40 line pairs across the sensor.  If you look at the MTF graphs for the various lenses you can determine which of them, at what apertures, can meet this requirement.  I have not verified this assumption, nor how it relates to pictoral "quality", which as suggested above is in the eye of the beholder (Karbe vs Mandler).  You might also check out Erwin Put's Leica Compendium for his review of the Leica line-up.

[Jeff S]

1 hour ago, Rick in CO said:

There was a previous discussion relating to Peter Karbe's suggestion that the higher resolution sensors would require lenses that can achieve at least 40 line pairs across the sensor.  If you look at the MTF graphs for the various lenses you can determine which of them, at what apertures, can meet this requirement.  I have not verified this assumption, nor how it relates to pictoral "quality", which as suggested above is in the eye of the beholder (Karbe vs Mandler).  You might also check out Erwin Put's Leica Compendium for his review of the Leica line-up.

Karbe is interested in highest resolution/contrast (and, dare I say, promoting new lens sales); not everyone else is (post 20).   Roger Cicala’s commentary still applies, with all due respect to Karbe.

Jeff

[Advertisement]

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

[Rick in CO]

2 hours ago, Jeff S said:

Karbe is interested in highest resolution/contrast (and, dare I say, promoting new lens sales); not everyone else is (post 20).   Roger Cicala’s commentary still applies, with all due respect to Karbe.

Jeff

I agree, Jeff.  I just thought the comment was pertinent to this discussion.

[Jeff S]

44 minutes ago, Rick in CO said:

I agree, Jeff.  I just thought the comment was pertinent to this discussion.

No offense, just giving some context and perspective, as I see it.

Jeff

[pk851667]

7 hours ago, Jeff S said:

No offense, just giving some context and perspective, as I see it.

Jeff

I suppose to use this terminology - I'm looking for answer between the Karbe and Mandler theories. And even if someone is giving the straight Karbe answer - it still feels like a cop out to say "get the APO" and drop the mic. 

[a.noctilux]

34 minutes ago, pk851667 said:

I suppose to use this terminology - I'm looking for answer between the Karbe and Mandler theories. And even if someone is giving the straight Karbe answer - it still feels like a cop out to say "get the APO" and drop the mic. 

After some years of Mandler/Karbe lenses using, I don't see any theories in them.

They do the real things, and the customers 😉 do the rest.

To be precise, Mandler created those lenses to match (at best) his customers need ( limiting by the materials * of his time ).

Karbe  just does the same with new customers/materials of his time.

 

We are lucky to have choices of the two 'philosophies' and so many more to suit our taste.

 

* hardware/software/ideas/etc.

[frame-it]

Advertisement (gone after registration)

16 hours ago, bherman01545 said:

in general, the "higher-performing" lenses will perform better on a high resolution sensor.

interesting statement.. have you used old lenses on high rez 50MP sensors?

Edited November 30, 2021 by frame-it

[Ouroboros]

The idea of ‘matching’ a lens to the M10-R is absurd. 
 

[Jeff S]

4 hours ago, pk851667 said:

I suppose to use this terminology - I'm looking for answer between the Karbe and Mandler theories. And even if someone is giving the straight Karbe answer - it still feels like a cop out to say "get the APO" and drop the mic. 

I agree with post #26. Even using identical gear, no two (discerning) photographers produce the same looking pics and prints. Sensors don’t dictate lens choice; people do. Tastes and objectives vary.

Jeff

[Rick in CO]

To get back to the original post, it is interesting in Roger Cicala's article that apparently none of the wide angle lenses tested (none were Leica, however) met his criteria on the higher resolution sensors.  Isn't that what the original post described?

[Jeff S]

Cicala’s Appendix to the article provides the response I value.  And I don’t know any really ‘crappy’ Leica lenses.

Jeff

[pk851667]

14 minutes ago, Rick in CO said:

To get back to the original post, it is interesting in Roger Cicala's article that apparently none of the wide angle lenses tested (none were Leica, however) met his criteria on the higher resolution sensors.  Isn't that what the original post described?

This is an interesting point, as much of the marketing (both official manufacturer and among users) around high MP cameras is the ability to shoot wide and crop later. Indeed, this is a serious selling point for the Q2, M10M, SL2 and yes... M10-R. 

[Jeff S]

Resolution tests, marketing/sales and art are different things. Consider the source.

Jeff

[pk851667]

1 minute ago, Jeff S said:

Resolution tests, marketing/sales and art are different things. Consider the source.

Jeff

I'm not disputing that. I'm simply saying that if the resolution tests don't match up with clinical tests... it at least raises an eyebrow.

[Jeff S]

3 minutes ago, pk851667 said:

I'm not disputing that. I'm simply saying that if the resolution tests don't match up with clinical tests... it at least raises an eyebrow.

Clinical tests?  
 

Jeff

[pk851667]

9 minutes ago, Jeff S said:

Clinical tests?  
 

Jeff

I meant laboratory setting resolution bench tests.

[Jeff S]

9 minutes ago, pk851667 said:

I meant laboratory setting resolution bench tests.

Then post#34 makes no sense to me , if resolution tests don’t match resolution tests.

In any event, I feel we’re going in circles here and I’ve stated my views. Carry on.

Jeff

[aristotle]

1 hour ago, pk851667 said:

I meant laboratory setting resolution bench tests.

Can you post an image that you consider poor with one of your existing wider lenses?  I think it might help calibrate responses.  I, for example, can tell you that my 35 non APO cron is wonderful on the M10R, but you may be looking for something quite different than me.

[adan]

A few technical points here:

1. MTF charts do not display raw resolution.They display percentage of contrast at various fixed resolutions, the finest of which usually reported is 40 lpmm (at least from Leica and Zeiss).

2. 40 lpmm is really a pretty "low-rent" standard. Just "acceptable" for a larger-format lens like 6x6 or 4x5 - pretty weak for 24x36mm lenses.

The 40-lpmm came from Zeiss research into "apparent sharpness" as perceived by the average person - i.e. folks getting drugstore prints in the 4x6" to 8x10" (10x15 cm to 16/20x24 cm) range. High contrast (50%+) at 40 lpmm "appeared sharper" then lower contrast at 80 lpmm, to most people (i.e. "the herd"), in those smallish prints.

I go back far enough that the first lens tests I read (early 1970s) still separated "resolution" and "contrast" on separate tables.

Kinda like the tables on this site - absolute resolution: center, middle, corner. The number of pickets in the fence - period.

https://web.hevanet.com/cperez/MF_testing.html

And even in 1971, most decent lenses (e.g. the then-new Canon FD line) could resolve 70-80 lpmm at their peak - with "enough" contrast to be visible at bigger enlargements. And my experience is that most Leica lenses back to the 1970s also do at least that well, except possibly in the corners or at full aperture.

3. The M10-R has pixels about ~6.6 microns square. Which equates to 151.5 pixels per mm. Given the Nyquist-Shannon sampling theorem (we need to sample a signal (e.g. from a lens) at twice the signal resolution for accurate reproduction), that is just barely enough to resolve (without artifacts like moiré or jaggies) 75 line pairs per mm. Therefore, even if one accepts the "weakest link" idea, the M10-R sensor is not "better" than most Leica lenses from the past 40-50 years, except at the margins. It is, in fact, "well matched" to lenses putting out 70-80 lpmm. (And frankly, a bit weak for, perhaps, the 50mm APO-Summicron-M - rumored to hit around 120 lpmm at its peak.)

(Personally, I don't accept the weakest link idea (I did once, but live and learn) - I agree with Roger Cicala and Jeff S and others. All sensors can only "degrade" the output of all lenses (as measured by MTF), and all lenses can only "degrade" the output of all sensors. But - a better sensor or lens will degrade a lens's or sensor's performance less. The "system MTF" will be higher - but never perfect (1.00 - or a 1:1 correspondence to the subject) ).

[Advertisement]

Hallo,

Look here: Leica M

[adan]

Update to correct bad data on M10-R pixel pitch. New math in BOLD.

1 hour ago, adan said:

3. The M10-R has pixels about ~4.59 microns square. Which equates to 217 pixels per mm. Given the Nyquist-Shannon sampling theorem (we need to sample a signal (e.g. from a lens) at twice the signal resolution for accurate reproduction), that is just barely enough to resolve (without artifacts like moiré or jaggies) 108 line pairs per mm. Therefore, even if one accepts the "weakest link" idea, the M10-R sensor is only slightly "better" than most Leica lenses from the past 40-50 years, except at the margins. It is, in fact, "well matched" to lenses putting out 80-100 lpmm. (And frankly, a bit weak for, perhaps, the 50mm APO-Summicron-M - rumored to hit around 120 lpmm at its peak.)

Nevertheless, here is a sample on the plain M10, with the (oft-maligned) 90mm Tele-Elmarit-M of the 1970s.

It is producing detail fine enough to render at close to one single pixel width on the M10 (twigs, light filaments, glasses frame, typography) - or on the close order of 80 lpmm. With occasional artifacts (jaggies or moiré, or blurs from de-Bayering).

Therefore the M10-R would improve this image (if one is into pixel peeping) - fewer artifacts, and even tighter edges and details.

A better sensor lifts all lenses.

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!

Edited November 30, 2021 by adan