Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Sharpest Lenses
Post new topic   Reply to topic
Home   Manual Focus Lenses
View previous topic :: View next topic  


PostPosted: Mon Feb 07, 2022 9:53 pm    Post subject: Reply with quote
Ray Parkhurst wrote:
Fandyus wrote:

As I expected, the testing was completely inadequate to determine if the lens is actually sharp at f/0.95 because the test method is not capable of measuring sharpness properly at that aperture. All the method is testing is if the Z7 camera is capable of resolving the target, so basically this was a camera test, not a lens test.


Edited to add: I suppose one thing the method DID show is that lens aberrations were low enough that they did not noticeably degrade sharpness at the Z7's pixel pitch.

Further edit: Sharpness at f/0.95 requires that the test system be capable of >750lp/mm, or about 0.7mm pixel pitch. The Z7 has pixel pitch of 4.3um, implying a DLA of f6.7. Note that the DLA does not take into account the Bayer demosaicing problems that give only 50% pixel coverage to R/B pixels, so to have full color coverage it is more like f14. Ignoring the Bayer issue, it would still take a ~7x magnification to determine if the lens is "sharp" at f/0.95. Of course this is not difficult to do, but takes a bit more thought and work than was done in the YT video.

Before I would shell out $8k for such a lens, I'd expect some better testing.


PostPosted: Tue Feb 08, 2022 5:48 am    Post subject: Reply with quote
Ray Parkhurst wrote:
Fandyus wrote:

As I expected, the testing was completely inadequate to determine if the lens is actually sharp at f/0.95 because the test method is not capable of measuring sharpness properly at that aperture. All the method is testing is if the Z7 camera is capable of resolving the target, so basically this was a camera test, not a lens test.

Excuse me???
He is using a 40 something megapixel camera to see if the sensor can outresolve the lens at a given aperture. Any other f/0.95 lens he's ever tested was at least blurry in the corners wide open. This lens did not get outresolved by the sensor, however, at any part of the image. If that isn't sharpness to you then I don't what the hell you're on about.


PostPosted: Tue Feb 08, 2022 1:38 pm    Post subject: Reply with quote
Fandyus wrote:
Ray Parkhurst wrote:
Fandyus wrote:

As I expected, the testing was completely inadequate to determine if the lens is actually sharp at f/0.95 because the test method is not capable of measuring sharpness properly at that aperture. All the method is testing is if the Z7 camera is capable of resolving the target, so basically this was a camera test, not a lens test.

Excuse me???
He is using a 40 something megapixel camera to see if the sensor can outresolve the lens at a given aperture. Any other f/0.95 lens he's ever tested was at least blurry in the corners wide open. This lens did not get outresolved by the sensor, however, at any part of the image. If that isn't sharpness to you then I don't what the hell you're on about.


What I am "on about" is trying to help folks such as yourself understand what "sharpness" is. What the reviewer has shown is that the lens can produce a sharp image on a 40MP sensor camera. If next week Nikon comes out with a 160MP Z8, does this lens suddenly become less sharp because it looks blurry in the corners on a camera with half the pixel size?


PostPosted: Tue Feb 08, 2022 3:19 pm    Post subject: Reply with quote
Ray Parkhurst wrote:
Fandyus wrote:
Ray Parkhurst wrote:
Fandyus wrote:

As I expected, the testing was completely inadequate to determine if the lens is actually sharp at f/0.95 because the test method is not capable of measuring sharpness properly at that aperture. All the method is testing is if the Z7 camera is capable of resolving the target, so basically this was a camera test, not a lens test.

Excuse me???
He is using a 40 something megapixel camera to see if the sensor can outresolve the lens at a given aperture. Any other f/0.95 lens he's ever tested was at least blurry in the corners wide open. This lens did not get outresolved by the sensor, however, at any part of the image. If that isn't sharpness to you then I don't what the hell you're on about.


What I am "on about" is trying to help folks such as yourself understand what "sharpness" is. What the reviewer has shown is that the lens can produce a sharp image on a 40MP sensor camera. If next week Nikon comes out with a 160MP Z8, does this lens suddenly become less sharp because it looks blurry in the corners on a camera with half the pixel size?


I mean, every lens has a limit. There is no lens that will forever remain sharp as you increase the resolution. Sharpness is relative to the medium. If your camera is a 10$ webcam, you don't need a Zeiss lens for it, that would be useless.
The Noct can produce razor sharp images on a 40mp camera at f/0.95, no other lenses the guy has tested have been able to do this. If you think you know of a lens that could do even better feel free to suggest, but I really don't think you do.
And I can guarantee no vintage fast 50 is sharper than this, especially not wide open. Do you remember how poorly the Canon L 50mm f/1.0 performed?


PostPosted: Tue Feb 08, 2022 3:44 pm    Post subject: Reply with quote
Fandyus wrote:

I mean, every lens has a limit. There is no lens that will forever remain sharp as you increase the resolution. Sharpness is relative to the medium. If your camera is a 10$ webcam, you don't need a Zeiss lens for it, that would be useless.
The Noct can produce razor sharp images on a 40mp camera at f/0.95, no other lenses the guy has tested have been able to do this. If you think you know of a lens that could do even better feel free to suggest, but I really don't think you do.
And I can guarantee no vintage fast 50 is sharper than this, especially not wide open. Do you remember how poorly the Canon L 50mm f/1.0 performed?


The lens may be very sharp at f/0.95, but testing it on a Z7 simply won't tell you that. The Z7 is capable of testing sharpness of lenses at around f/7 or smaller, and that is being very generous. Anything larger may be sharper, but the camera can't tell you that, and the difference between f/7 and f/0.95 is huge.

Edited to add: Sharpness is a tradeoff between lens aberrations and diffraction. Usually a large aperture lens will have poor aberration control wide open, and will usually show this even at low pixel density. The Noct only shows a bit of this at the corners, so I'd conclude it is well-corrected and has a good chance of being very sharp indeed.

I only glanced at the video around the time marker you specified. Did the reviewer test the lens at smaller apertures? Did the corners improve?


PostPosted: Tue Feb 08, 2022 4:25 pm    Post subject: Reply with quote
Ray Parkhurst wrote:

Edited to add: Sharpness is a tradeoff between lens aberrations and diffraction. Usually a large aperture lens will have poor aberration control wide open, and will usually show this even at low pixel density. The Noct only shows a bit of this at the corners, so I'd conclude it is well-corrected and has a good chance of being very sharp indeed.


This was all people were trying to say at the beginning, though, and is going to be what's implied by 99% of people calling a photographic lens sharp.

If you were to average out people's expectations based on the gear used here, it would probably be a 24MP FF sensor / 24MP out of flat and fine grain film.

As you say the 46MP Z7 is only capable of testing f/7 lenses, well, adjust your expectations for the gear people use here- and the bar is even lower to clear. That's just how things are for photographic lenses.

Ray Parkhurst wrote:

The lens may be very sharp at f/0.95, but testing it on a Z7 simply won't tell you that. The Z7 is capable of testing sharpness of lenses at around f/7 or smaller, and that is being very generous. Anything larger may be sharper, but the camera can't tell you that, and the difference between f/7 and f/0.95 is huge.


...And the relevance goes down as quickly as the testing requirements go up Wink

Who's waiting for a mythical 160MP FF sensor/film to determine photographic lens sharpness?

Has this ever existed? No, probably not, because literally until this Nikon, f0.95 lenses were obviously not resolute.


PostPosted: Tue Feb 08, 2022 4:37 pm    Post subject: Reply with quote
eggplant wrote:

This was all people were trying to say at the beginning, though, and is going to be what's implied by 99% of people calling a photographic lens sharp.


I think the vast majority of folks don't know this, even on this forum. My posts were simply to point it out so perhaps next time they'll realize the limitations of the medium in determining "sharpness".


PostPosted: Tue Feb 08, 2022 5:49 pm    Post subject: Reply with quote
Ray Parkhurst wrote:
Fandyus wrote:

I mean, every lens has a limit. There is no lens that will forever remain sharp as you increase the resolution. Sharpness is relative to the medium. If your camera is a 10$ webcam, you don't need a Zeiss lens for it, that would be useless.
The Noct can produce razor sharp images on a 40mp camera at f/0.95, no other lenses the guy has tested have been able to do this. If you think you know of a lens that could do even better feel free to suggest, but I really don't think you do.
And I can guarantee no vintage fast 50 is sharper than this, especially not wide open. Do you remember how poorly the Canon L 50mm f/1.0 performed?


The lens may be very sharp at f/0.95, but testing it on a Z7 simply won't tell you that. The Z7 is capable of testing sharpness of lenses at around f/7 or smaller, and that is being very generous. Anything larger may be sharper, but the camera can't tell you that, and the difference between f/7 and f/0.95 is huge.

Edited to add: Sharpness is a tradeoff between lens aberrations and diffraction. Usually a large aperture lens will have poor aberration control wide open, and will usually show this even at low pixel density. The Noct only shows a bit of this at the corners, so I'd conclude it is well-corrected and has a good chance of being very sharp indeed.

I only glanced at the video around the time marker you specified. Did the reviewer test the lens at smaller apertures? Did the corners improve?


I honestly did not see any significant softness in the corners when watching the video, so I am not sure what you would expect to improve by stopping down.
Also, what even is sharpness to you? Do you genuinely expect me to have standards so high that I will not consider a lens sharp unless I'm using a 170mp sensor?
Like sure, I won't oppose you saying the Noct is overpriced. But it's sharpness at f/0.95 is literally unparalleled among consumer lenses today.
Perhaps NASA has something brighter and sharper, I dunno, sounds like you might wanna go ask them.


PostPosted: Tue Feb 08, 2022 6:45 pm    Post subject: Reply with quote
eggplant wrote:

Who's waiting for a mythical 160MP FF sensor/film to determine photographic lens sharpness?

My A7Rm4 generates 240MP images with its 60MP sensor in 16-pixel-shift mode. It is still nowhere near capable of determining (on its own) if a f/0.95 lens is sharp.


Fandyus wrote:

I honestly did not see any significant softness in the corners when watching the video, so I am not sure what you would expect to improve by stopping down.

It was mentioned by the reviewer. I'd presume he was able to view the actual image rather than something on a YT video.


Fandyus wrote:

Also, what even is sharpness to you? Do you genuinely expect me to have standards so high that I will not consider a lens sharp unless I'm using a 170mp sensor?

Sharpness has a lot of meanings in photography. But it is a bit more specific to say that "this lens is sharp at f/0.95", which means to me that the lens meets MTF50 at f/0.95 with a resolution target appropriate for that aperture.

I certainly don't expect such high standards as long as such specific statements are not made. If you say "this lens is sharp on my Z7", that pretty much says what I think you mean. But if you say "this lens is sharp at f/0.95", IMO it has a much higher standard expectation.


PostPosted: Tue Feb 08, 2022 7:04 pm    Post subject: Reply with quote
Testing equipment using high resolution target, portions of image circle viewed through microscope optics, is the laboratory method to determine sharpness.

Lens Testing Bench https://opg.optica.org/josa/abstract.cfm?uri=josa-43-1-44 and https://www.optikos.com/lens-testing-equipment/

Last edited by visualopsins on Wed Feb 09, 2022 5:59 pm; edited 2 times in total


PostPosted: Wed Feb 09, 2022 5:51 pm    Post subject: Reply with quote
Although I will add, for a demonstration of how much we have to go, inline with what Ray was saying, this is super interesting https://www.dyxum.com/dforum/the-joy-of-1-54m-pixels_topic108753.html

Pentax Q, 1.54µm pitch 12MP sensor, equivalent to 350MP ff.


PostPosted: Wed Feb 09, 2022 8:22 pm    Post subject: Reply with quote
eggplant wrote:
Although I will add, for a demonstration of how much we have to go, inline with what Ray was saying, this is super interesting https://www.dyxum.com/dforum/the-joy-of-1-54m-pixels_topic108753.html

Pentax Q, 1.54µm pitch 12MP sensor, equivalent to 350MP ff.


It is a good example.

I find the 100% crops of the tomatoes most interesting and relevant. Many folks looking at the crop from the A900 at f/4 would say that the lens is not sharp, yet the PQ image shows that it is the fault of the camera, not the lens. It's sort of the flipside of what I am saying, ie that if folks see a sharp image, they say that the lens must be sharp.

@visualopsins' links to the test benches show how testing pros view the aerial image under magnification to check sharpness. There are other methods as well, such as using a slanted edge (I forget whose software does this), but my preference is still to use teleconverters. It is similar to the test bench method, but can be done with equipment most folks already have, at least at the center. Corners require shift capability.


PostPosted: Sun Feb 13, 2022 2:12 pm    Post subject: Reply with quote
Sharpest of mine are
105mm f5.6 Apo El Nikkor
165mm f4 Ultra Micro Nikkor
85mm f1.0 Repro Nikkor

The 105mm Apo Lanthar is also no slouch.


PostPosted: Fri Feb 25, 2022 6:33 pm    Post subject: Reply with quote
You are taking this sharpness stuff too seriously. My understanding is that sharpness is not actually a real term used in testing but a subjective quality and it's based on perception and an image can look sharp by different combinations of factors. Low acutance and high resolution can look sharp. An image from one lens that has lower resolution may look sharper because it has higher contrast, etc.

The other problem is it seems there are very few definitions people agree on. Acutance. Resolution. Contrast. Microcontrast. Some appear to be well defined but others seem to vary in what they mean to different people. This of course doesn't even get into aberrations and shooting conditions.

It's nice just to have a thread where people say these are the lenses I own that look sharpest to me. Then people can decide if they want to try them for themselves.


PostPosted: Fri Feb 25, 2022 9:08 pm    Post subject: Reply with quote
As already mentioned, the criterion of sharpness is not sufficiently defined and is therefore a rather subjective one.

What I also consider important is that since digital photography, an additional variable has been added, which generally receives too little attention here on the forum:
Today, it is no longer possible to speak of the "sharpest" lens, but only of the combination of lens/camera or sensor, which can also be clearly understood at "dxomark": There are always different measured values when identical lenses are tested on different cameras.

Therefore, there will be no recommendation from my side, since there are different favorites for each of my cameras.


PostPosted: Fri Feb 25, 2022 9:12 pm    Post subject: Reply with quote
tb_a wrote:
...Therefore, there will be no recommendation from my side, since there are different favorites for each of my cameras.


Such a list by camera would be useful! Smile


PostPosted: Fri Feb 25, 2022 9:22 pm    Post subject: Reply with quote
visualopsins wrote:

Such a list by camera would be useful!


Sorry, I own 8 different digital system cameras with different sensors in MFT, APS-C and FF.

You don't seriously expect me to present a list of 80 lenses, do you?


PostPosted: Fri Feb 25, 2022 10:05 pm    Post subject: Reply with quote
tb_a wrote:
visualopsins wrote:

Such a list by camera would be useful!


Sorry, I own 8 different digital system cameras with different sensors in MFT, APS-C and FF.

You don't seriously expect me to present a list of 80 lenses, do you?


Aw come on; it'll be easy for you! Smile (And to repeat, useful for mflenses readers)


PostPosted: Fri Feb 25, 2022 10:27 pm    Post subject: Reply with quote
visualopsins wrote:
Aw come on; it'll be easy for you! Smile (And to repeat, useful for mflenses readers)


Seriously, I wouldn't even be able to name these lenses for a single camera, since it always depends on the intended use, e.g. landscape or anything else.
In the end, we still haven't defined what sharpness is at all.
Therefore my estimation would be useless for others anyway. I have experienced this here and in other forums many times already.
Example: Do you believe in the so-called "3d-pop"? I don't. I have already addressed this issue here in the forum a few years ago. It already starts here. ..

However, it was still a nice try on your part. Wink


PostPosted: Fri Feb 25, 2022 11:08 pm    Post subject: Reply with quote
Excuses, excuses.Laughing You're such a teaser Thomas... LOL

Maybe you could write it down each time; star another topic? Like Favorite Lenses: Lists by Camera


PostPosted: Fri Feb 25, 2022 11:41 pm    Post subject: Reply with quote
Good idea!

Another one: Subforums at least for different sensor formats. Wink


PostPosted: Sat Feb 26, 2022 12:32 am    Post subject: Reply with quote
tb_a wrote:
As already mentioned, the criterion of sharpness is not sufficiently defined and is therefore a rather subjective one.

What I also consider important is that since digital photography, an additional variable has been added, which generally receives too little attention here on the forum:
Today, it is no longer possible to speak of the "sharpest" lens, but only of the combination of lens/camera or sensor, which can also be clearly understood at "dxomark": There are always different measured values when identical lenses are tested on different cameras.

Therefore, there will be no recommendation from my side, since there are different favorites for each of my cameras.


You are probably trying to make your point in the wrong topic here, and so am I probably below...! Wink

But I'm totally with you re the impact of the camera; film levelled the playing field in a sense there. If you are serious about finding the sharpest lenses, you need a controlled test environment capable of resolving the theoretical diffraction-limitation of the imaging optics.

On a different note (and folks, before I get lambasted, I'm not suggesting for a minute that the quest for the sharpest lens is wrong or pointless), but I personally just don't get the appeal of this quest for sharpness. Sure, the best possible image sharpness can be required at times for specific image purposes, but generally, it is not what I really appreciate in a photograph. Some of the most revered images are anything but sharp. To me, composition and use of light, shadow, and colour are far more interesting.

E.g. take Autochrome A71243 by Fernand Cuville from the Albert Kahn archive (16 May 1918):

I can't show it here because of copyright, but Google for image "Albert Kahn A71243 Fernand Cuville Verona", should be the first image that comes up, probably from the Wiener Zeitung (second image in their slideshow, https://www.wienerzeitung.at/nachrichten/kultur/medien/565171-Fotoschaetze-aus-dem-Archiv-des-Albert-Kahn.html , click again to see it bigger). A girl sitting in front of an alabaster holy water font and pillar in the vault of the church of San Zeno de Maggiore in Verona.

An enchanting and beautifully executed image, but very soft all over upon inspection, not least because it is an autochrome and also the > 1minute exposure time (poor girl).

Personally, I would be far more interested to know which lenses are really pleasant to handle and focus, and have great haptics, but that is also a subjective experience. Wink


PostPosted: Sat Feb 26, 2022 12:41 am    Post subject: Reply with quote
@Mark

At least you are reading the oldest Austrian newspaper founded in my birth town 1703. Wink


PostPosted: Sat Feb 26, 2022 12:43 am    Post subject: Reply with quote
RokkorDoctor wrote:
tb_a wrote:
As already mentioned, the criterion of sharpness is not sufficiently defined and is therefore a rather subjective one.

What I also consider important is that since digital photography, an additional variable has been added, which generally receives too little attention here on the forum:
Today, it is no longer possible to speak of the "sharpest" lens, but only of the combination of lens/camera or sensor, which can also be clearly understood at "dxomark": There are always different measured values when identical lenses are tested on different cameras.

Therefore, there will be no recommendation from my side, since there are different favorites for each of my cameras.


You are probably trying to make your point in the wrong topic here, and so am I probably below...! Wink

But I'm totally with you re the impact of the camera; film levelled the playing field in a sense there. If you are serious about finding the sharpest lenses, you need a controlled test environment capable of resolving the theoretical diffraction-limitation of the imaging optics.

On a different note (and folks, before I get lambasted, I'm not suggesting for a minute that the quest for the sharpest lens is wrong or pointless), but I personally just don't get the appeal of this quest for sharpness. Sure, the best possible image sharpness can be required at times for specific image purposes, but generally, it is not what I really appreciate in a photograph. Some of the most revered images are anything but sharp. To me, composition and use of light, shadow, and colour are far more interesting.

E.g. take Autochrome A71243 by Fernand Cuville from the Albert Kahn archive (16 May 1918):

I can't show it here because of copyright, but Google for image "Albert Kahn A71243 Fernand Cuville Verona", should be the first image that comes up, probably from the Wiener Zeitung (second image in their slideshow, https://www.wienerzeitung.at/nachrichten/kultur/medien/565171-Fotoschaetze-aus-dem-Archiv-des-Albert-Kahn.html , click again to see it bigger). A girl sitting in front of an alabaster holy water font and pillar in the vault of the church of San Zeno de Maggiore in Verona.

An enchanting and beautifully executed image, but very soft all over upon inspection, not least because it is an autochrome and also the > 1minute exposure time (poor girl).

Personally, I would be far more interested to know which lenses are really pleasant to handle and focus, and have great haptics, but that is also a subjective experience. Wink


Like 1


PostPosted: Sat Feb 26, 2022 2:28 am    Post subject: Reply with quote
[quote="RokkorDoctor"]
tb_a wrote:

I can't show it here because of copyright,


There is no limitation on showing such an image under Fair Use doctrine.
Post new topic   Reply to topic
Home   Manual Focus Lenses
Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next
Page 6 of 7