Lenses and 40MP sensors

kodachromed

Active member
Messages
55
Reaction score
82
When Fuji first rolled out the 40MP sensor there were some questions about the performance of lenses not on Fuji's list of lenses that were optimized for the new sensor. The consensus seemed to be that though lenses not on the list may not be able to get the most out of what the sensor could do the lenses would certainly not perform worse. So relax if your 14mm f2.8 is not on the list; it will still be better than on a 26MP sensor.

On the other hand I noticed that there were some comments on this forum that described the 40MP sensor as being "very demanding" and that any lens imperfection was likely to be much more obvious than on a smaller MP sensor as a result. If this is true then it couldn't be said that all lenses are better on a higher MP sensor.

Obviously both positions can't be correct. i.e all lenses will benefit to a lesser or greater degree on a high MP sensor and at the same time say weaknesses in a lens will be more apparent. I wouldn't call a weakness being more apparent an improvement. Maybe some lenses are actually better off being paired with a lower MP sensor.
 
No lens will perform worse on the 40mp sensor. At a pixel level you might well see some of those “imperfections” on the newer sensor but when viewed at the same magnification across both 24/26 and 40 the images will be at least the same, if not better on the 40mp.
 
Think about it like this: you are sampling the same scene detail with a higher resolution "mesh" or "grid"; hence, you need to interpolate less to get the final image. So, in terms of resolution, you gain detail. But say your lens as a bad case of chromatic aberration, you will see it more.
 
When Fuji first rolled out the 40MP sensor there were some questions about the performance of lenses not on Fuji's list of lenses that were optimized for the new sensor. The consensus seemed to be that though lenses not on the list may not be able to get the most out of what the sensor could do the lenses would certainly not perform worse. So relax if your 14mm f2.8 is not on the list; it will still be better than on a 26MP sensor.
Not performing worse is not the same as necessarily performing better.
On the other hand I noticed that there were some comments on this forum that described the 40MP sensor as being "very demanding" and that any lens imperfection was likely to be much more obvious than on a smaller MP sensor as a result. If this is true then it couldn't be said that all lenses are better on a higher MP sensor.
Could you point to threads where these claims were made; specifically that ‘any lens imperfection’ will be much more obvious, or that ‘all lenses are better on the higher MP sensor”.
Obviously both positions can't be correct. i.e all lenses will benefit to a lesser or greater degree on a high MP sensor and at the same time say weaknesses in a lens will be more apparent. I wouldn't call a weakness being more apparent an improvement. Maybe some lenses are actually better off being paired with a lower MP sensor.
I think you’ve oversimplified and misinterpreted to the point of distortion, points which have been discussed at quite some length here. I don’t see any paradox in what’s been reported, and I’m not aware of anyone yet demonstrating that some lenses are universally ‘better off paired with a lower MP sensor”.

As always, context and detail are essential to a proper discussion and understanding of the facts. If you have relevant experience or specific evidence to share, that would be good. Which lens? What exposure? 100%? 200%?
You might be onto something here which I’ve missed in all the discussion to date but that needs demonstrating with more than some selective ‘summary’ of other people’s reported comments.
 
Last edited:
The root cause of this apparent inconcistency comes from the same source as the one about whether it affects noise level:

Whether the viewer is looking with a magnification that keeps pixel size fixed or image size fixed.

If the image size is fixed, as is the case when you look at the full image (or a region crop of it) on a fixed size print or monitor; there is no IQ downside to more pixel density. Assuming, of course, that the manufacturer didn't "cheat" their way to smaller pixels by making them a lower quality or different type.

If instead the pixel size is fixed, as I believe is the case when selecting a view mode like 100% or 200% in Adobe Lightroom, the same photo taken with more pixel density will be more zoomed in, which also zooms more in on the lens' imperfections.

When I talk from one perpective and another talks from the other perspective, it's impossible to agree. For any two people to have useful conversation about this, they must first agree on one perspective and on one agreed definition for every word in the photography dictionary :-D And for a whole forum to agree...
 
Last edited:
All lenses will give better IQ on a sensor with more pixels. Some impefections will be easier to see, like details are, but they won't be any more prominent in the finished image. If you read anything that implies otherwise write down the source and put it somewhere prominent so you can avoid reading anything else from that source. The internet is unedited.

--
Andrew Skinner
 
Last edited:
The root cause of this apparent inconcistency comes from the same source as the one about whether it affects noise level:

Whether the viewer is looking with a magnification that keeps pixel size fixed or image size fixed.

If the image size is fixed, as is the case when you look at the full image (or a region crop of it) on a fixed size print or monitor; there is no IQ downside to more pixel density. Assuming, of course, that the manufacturer didn't "cheat" their way to smaller pixels by making them a lower quality or different type.

If instead the pixel size is fixed, as I believe is the case when selecting a view mode like 100% or 200% in Adobe Lightroom, the same photo taken with more pixel density will be more zoomed in, which also zooms more in on the lens' imperfections.
However, that is comparing apples to persimmons. When you are viewing at 100% you are actually viewing the 40 mp sensor at an enlargement ratio that is Sqrt[20/13) greater than the 26. Viewed on the same size physical display, the high MP will result in a lower enlargement ratio with smoother tonal gradation and more detail.
 
Think about it like this: you are sampling the same scene detail with a higher resolution "mesh" or "grid"; hence, you need to interpolate less to get the final image. So, in terms of resolution, you gain detail. But say your lens as a bad case of chromatic aberration, you will see it more.
Why should I see it more and not just the same amount of CA?

Looking at two images made with different sensor resolutions at the same print size, it should be really the same amount of CA. I see no reason why it should be more.

Looking at the images at 100 % magnification, of course, everything will appear larger. Any subject, any CA.

BR,

Martin

--
SmugMug - https://martinlang.smugmug.com
500px - https://500px.com/martinlangphotography
Insta - https://www.instagram.com/martin.lang.photography
Co-author on https://frickelfarm.de/
 
Last edited:
The only reason a “weakness” would be any more apparent is, on the same screen, a 100% 40MP view will look closer at the image than a 100% 24/26MP view. With the worst lens in the world with resolution that doesn’t even challenge a 24MP sensor, at 24MP (especially with an X-Trans sensor) there are going to be visible demosaicing/interpolation errors present that degrade the image further. At 40MP there is a lot more sensor data for the processing to work with and far fewer false detail/moire errors to make a mess of things so, even with relatively low resolution lens, whatever detail it can deliver will be rendered more accurately both in terms of fine detail and color fidelity too - it’ll look better. Processed well and viewed at the same size, the higher resolution sensor should always produce a visibly superior result regardless of the lens used. Never worse.

Even with the sharpest lenses, you may indeed need to employ precise shot discipline and careful processing to fully reap the rewards that a 40MP sensor offers, but the potential is there for genuinely superior results relative to what you can get at 24/26MP.
 
Last edited:
When Fuji first rolled out the 40MP sensor there were some questions about the performance of lenses not on Fuji's list of lenses that were optimized for the new sensor. The consensus seemed to be that though lenses not on the list may not be able to get the most out of what the sensor could do the lenses would certainly not perform worse. So relax if your 14mm f2.8 is not on the list; it will still be better than on a 26MP sensor.
when the whole image is viewed as a whole image.
On the other hand I noticed that there were some comments on this forum that described the 40MP sensor as being "very demanding" and that any lens imperfection was likely to be much more obvious
when a part of the image is viewed at 100% zoom.
than on a smaller MP sensor as a result. If this is true then it couldn't be said that all lenses are better on a higher MP sensor.

Obviously both positions can't be correct.
They can, because they correspond to DIFFERENT viewing conditions.
i.e all lenses will benefit to a lesser or greater degree on a high MP sensor and at the same time say weaknesses in a lens will be more apparent. I wouldn't call a weakness being more apparent an improvement. Maybe some lenses are actually better off being paired with a lower MP sensor.
Nope, absolutely not.
 
Last edited:
The root cause of this apparent inconcistency comes from the same source as the one about whether it affects noise level:

Whether the viewer is looking with a magnification that keeps pixel size fixed or image size fixed.

If the image size is fixed, as is the case when you look at the full image (or a region crop of it) on a fixed size print or monitor; there is no IQ downside to more pixel density. Assuming, of course, that the manufacturer didn't "cheat" their way to smaller pixels by making them a lower quality or different type.

If instead the pixel size is fixed, as I believe is the case when selecting a view mode like 100% or 200% in Adobe Lightroom, the same photo taken with more pixel density will be more zoomed in, which also zooms more in on the lens' imperfections.
However, that is comparing apples to persimmons. When you are viewing at 100% you are actually viewing the 40 mp sensor at an enlargement ratio that is Sqrt[20/13) greater than the 26. Viewed on the same size physical display, the high MP will result in a lower enlargement ratio with smoother tonal gradation and more detail.
I believe this is what I was saying. If I'm missing something, please let me know.
 
I have 5 Fuji primes from the X-T1 era (14/2.8, 23/1.4, 35/1.4, 56/1.2, 60/2.4), all perform as well on the X-T5 as they did on the X-T1. I'm not seeing any glaring deficiencies in the lenses caused by the increase in sensor resolution.

The only thing I noted moving from the 16mp sensor to the 40mp sensor was how much tighter my technique needs to be. The 40mp sensor is very sensitive to any camera movement.

--
Bill.
Proud user of Pentax and Fuji camera gear.
 
Last edited:
When Fuji first rolled out the 40MP sensor there were some questions about the performance of lenses not on Fuji's list of lenses that were optimized for the new sensor. The consensus seemed to be that though lenses not on the list may not be able to get the most out of what the sensor could do the lenses would certainly not perform worse. So relax if your 14mm f2.8 is not on the list; it will still be better than on a 26MP sensor.

On the other hand I noticed that there were some comments on this forum that described the 40MP sensor as being "very demanding" and that any lens imperfection was likely to be much more obvious than on a smaller MP sensor as a result. If this is true then it couldn't be said that all lenses are better on a higher MP sensor.

Obviously both positions can't be correct. i.e all lenses will benefit to a lesser or greater degree on a high MP sensor and at the same time say weaknesses in a lens will be more apparent. I wouldn't call a weakness being more apparent an improvement. Maybe some lenses are actually better off being paired with a lower MP sensor.
Actually, I think both can be true.

Noticing imperfections more easily doesn't necessarily mean that the lens isn't giving you better overall results on a higher MP sensor.

Just about any lens will give you more detail on a 40 MP sensor than on, say, a 16 MP sensor. That may also mean that perhaps you see a loss in corner sharpness more easily, as an example, if you zoom in to 100%.
 
The only reason a “weakness” would be any more apparent is, on the same screen, a 100% 40MP view will look closer at the image than a 100% 24/26MP view. With the worst lens in the world with resolution that doesn’t even challenge a 24MP sensor, at 24MP (especially with an X-Trans sensor) there are going to be visible demosaicing/interpolation errors present that degrade the image further. At 40MP there is a lot more sensor data for the processing to work with and far fewer false detail/moire errors to make a mess of things so, even with relatively low resolution lens, whatever detail it can deliver will be rendered more accurately both in terms of fine detail and color fidelity too - it’ll look better. Processed well and viewed at the same size, the higher resolution sensor should always produce a visibly superior result regardless of the lens used. Never worse.

Even with the sharpest lenses, you may indeed need to employ precise shot discipline and careful processing to fully reap the rewards that a 40MP sensor offers, but the potential is there for genuinely superior results relative to what you can get at 24/26MP.
Of course 40mp is more than 24 mp mathematically. There really is a potential difference as you say. Something like 25-30% more resolution and so on. But where and how can you see the IQ superiority ( if we do not talk about pixel peeping in 200%) ?

Nowadays people show their images in internet and they are looked at a PC screen ... Big prints are quite rare and it is really very difficult to see any quality differences in big prints like 60cm+90cm if you compare 24 and 40... where can we really see the superiority ? Good lenses are of course better and there are visible differences. PP can produce differences ...

I actually mean that a 40MP sensor is not a problem and all lenses are OK , but are we just fooled commercially ... "MP war" just goes on and we are the victims (our wallet I mean)
 
The only reason a “weakness” would be any more apparent is, on the same screen, a 100% 40MP view will look closer at the image than a 100% 24/26MP view. With the worst lens in the world with resolution that doesn’t even challenge a 24MP sensor, at 24MP (especially with an X-Trans sensor) there are going to be visible demosaicing/interpolation errors present that degrade the image further. At 40MP there is a lot more sensor data for the processing to work with and far fewer false detail/moire errors to make a mess of things so, even with relatively low resolution lens, whatever detail it can deliver will be rendered more accurately both in terms of fine detail and color fidelity too - it’ll look better. Processed well and viewed at the same size, the higher resolution sensor should always produce a visibly superior result regardless of the lens used. Never worse.

Even with the sharpest lenses, you may indeed need to employ precise shot discipline and careful processing to fully reap the rewards that a 40MP sensor offers, but the potential is there for genuinely superior results relative to what you can get at 24/26MP.
Of course 40mp is more than 24 mp mathematically. There really is a potential difference as you say. Something like 25-30% more resolution and so on. But where and how can you see the IQ superiority ( if we do not talk about pixel peeping in 200%) ?
The difference is most obvious in the ability to crop significantly without loss of IQ.
Nowadays people show their images in internet and they are looked at a PC screen ... Big prints are quite rare and it is really very difficult to see any quality differences in big prints like 60cm+90cm if you compare 24 and 40... where can we really see the superiority ? Good lenses are of course better and there are visible differences. PP can produce differences ...

I actually mean that a 40MP sensor is not a problem and all lenses are OK , but are we just fooled commercially ... "MP war" just goes on and we are the victims (our wallet I mean)
If anyone bought a 40mp camera purely for the greater pixel density, then some might say they were fooled, or foolish. If that camera had enhanced features or capabilities which justify buying it, and it happens to have the 40mp sensor, that’s a different thing IMO. Also happens to be what I did.
 
Last edited:
The only reason a “weakness” would be any more apparent is, on the same screen, a 100% 40MP view will look closer at the image than a 100% 24/26MP view. With the worst lens in the world with resolution that doesn’t even challenge a 24MP sensor, at 24MP (especially with an X-Trans sensor) there are going to be visible demosaicing/interpolation errors present that degrade the image further. At 40MP there is a lot more sensor data for the processing to work with and far fewer false detail/moire errors to make a mess of things so, even with relatively low resolution lens, whatever detail it can deliver will be rendered more accurately both in terms of fine detail and color fidelity too - it’ll look better. Processed well and viewed at the same size, the higher resolution sensor should always produce a visibly superior result regardless of the lens used. Never worse.

Even with the sharpest lenses, you may indeed need to employ precise shot discipline and careful processing to fully reap the rewards that a 40MP sensor offers, but the potential is there for genuinely superior results relative to what you can get at 24/26MP.
Of course 40mp is more than 24 mp mathematically. There really is a potential difference as you say. Something like 25-30% more resolution and so on. But where and how can you see the IQ superiority ( if we do not talk about pixel peeping in 200%) ?

Nowadays people show their images in internet and they are looked at a PC screen ... Big prints are quite rare and it is really very difficult to see any quality differences in big prints like 60cm+90cm if you compare 24 and 40... where can we really see the superiority ? Good lenses are of course better and there are visible differences. PP can produce differences ...

I actually mean that a 40MP sensor is not a problem and all lenses are OK , but are we just fooled commercially ... "MP war" just goes on and we are the victims (our wallet I mean)
I agree that visible improvement at 40MP vs. 24/26MP may very well be minimal for some users. If you’re just shooting jpegs or aren’t carefully processing your RAW files, there may be very little difference to see at all. That said, to my eyes, a well processed 40MP RAW file looks very noticeably better than 24/26MP at 100%, even with an older lens. Is it a big enough difference that I must upgrade NOW? No is not. I’ll get there soon enough, but my 24/26MP cameras are still producing very satisfying results for the time being.
 
Last edited:
When Fuji first rolled out the 40MP sensor there were some questions about the performance of lenses not on Fuji's list of lenses that were optimized for the new sensor.
Optimized doesn't mean the ones not on the list do not work well.
The consensus seemed to be that though lenses not on the list may not be able to get the most out of what the sensor could do the lenses would certainly not perform worse. So relax if your 14mm f2.8 is not on the list; it will still be better than on a 26MP sensor.
The difference between 26mp and 40mp is about 3x5" printing size at 300dpi. It isn't that much.
On the other hand I noticed that there were some comments on this forum that described the 40MP sensor as being "very demanding" and that any lens imperfection was likely to be much more obvious than on a smaller MP sensor as a result.
People say this with every change in MPs.
If this is true then it couldn't be said that all lenses are better on a higher MP sensor.

Obviously both positions can't be correct. i.e all lenses will benefit to a lesser or greater degree on a high MP sensor and at the same time say weaknesses in a lens will be more apparent. I wouldn't call a weakness being more apparent an improvement. Maybe some lenses are actually better off being paired with a lower MP sensor.
Is the lens sharp enough not to be distracting? Then it is good enough. That is how I view it.
 
The only reason a “weakness” would be any more apparent is, on the same screen, a 100% 40MP view will look closer at the image than a 100% 24/26MP view. With the worst lens in the world with resolution that doesn’t even challenge a 24MP sensor, at 24MP (especially with an X-Trans sensor) there are going to be visible demosaicing/interpolation errors present that degrade the image further. At 40MP there is a lot more sensor data for the processing to work with and far fewer false detail/moire errors to make a mess of things so, even with relatively low resolution lens, whatever detail it can deliver will be rendered more accurately both in terms of fine detail and color fidelity too - it’ll look better. Processed well and viewed at the same size, the higher resolution sensor should always produce a visibly superior result regardless of the lens used. Never worse.

Even with the sharpest lenses, you may indeed need to employ precise shot discipline and careful processing to fully reap the rewards that a 40MP sensor offers, but the potential is there for genuinely superior results relative to what you can get at 24/26MP.
Of course 40mp is more than 24 mp mathematically. There really is a potential difference as you say. Something like 25-30% more resolution and so on. But where and how can you see the IQ superiority ( if we do not talk about pixel peeping in 200%) ?
The difference is most obvious in the ability to crop significantly without loss of IQ.
Nowadays people show their images in internet and they are looked at a PC screen ... Big prints are quite rare and it is really very difficult to see any quality differences in big prints like 60cm+90cm if you compare 24 and 40... where can we really see the superiority ? Good lenses are of course better and there are visible differences. PP can produce differences ...

I actually mean that a 40MP sensor is not a problem and all lenses are OK , but are we just fooled commercially ... "MP war" just goes on and we are the victims (our wallet I mean)
If anyone bought a 40mp camera purely for the greater pixel density, then some might say they were fooled, or foolish. If that camera had enhanced features or capabilities which justify buying it, and it happens to have the 40mp sensor, that’s a different thing IMO. Also happens to be what I did.
Next step is 100MP

But you did not tell how the difference is possible to see. Photographs are a visual thing - we look at them. On paper on on screen...

More MP to crop more - how much more can you crop in numbers ?
 
For me the test is 200%. At that magnification, my 18mm 1.4 still looks good, but I notice flaws and weakness in the other lenses. At 100 % they all look fine.
 
For me the test is 200%. At that magnification, my 18mm 1.4 still looks good, but I notice flaws and weakness in the other lenses. At 100 % they all look fine.
You do know that 200% is interpolated by the display driver. That is the display driver interpolated every other pixel that is displayed. How is that a test for a lens.
 

Keyboard shortcuts

Back
Top