Any downsides related to pixel binning/reduced mode?

[Jon Warwick]

Advertisement (gone after registration)

I like using the DNG-S for (say) 20x16” prints of headshots. To my eyes, the pixel binning reduces the acutance compared to a DNG-L being downsized in post, and for my taste that’s pleasing for subjects such as portraits where you don’t want a super precise rendering. 

[Jeff S]

13 hours ago, Stuart Richardson said:

The motion blur is a non-issue: any motion blur there at 60mp is also there at 36mp. That is a magnification issue, not a resolution issue.

This should be a sticky.

[pgh]

^ agreed.

If the same issue is addressed by image > image size > (downsize) then the camera isn't helping in any special way, which I think is sort of everyone's point about this. You're not magically using a 36 mp sensor with all the traits it would have. You're using a 60 mp, with information tossed out. The difference isn't so academic in some real world scenarios. 

[adan]

10 hours ago, SrMi said:

Usually, a sensor with smaller resolution would have more aliasing than a sensor with higher resolution. However, the source of all three M11's resolutions is the same (60 MP), so aliasing should be the same.

Exactly.

An M11 60Mpixel .DNG image resampled in the camera to an 18 Mpixel .DNG is less subject to moiré (a form of aliasing) than a 24 Mpixel M10 original (let alone an 18Mpixel original - e.g. M9).  See ribbed white suitcase below.

The oversampling at the sensor surface at the moment of exposure sets a higher Nyquist limit - regardless of whether the image is reduced thereafter.

https://en.wikipedia.org/wiki/Oversampling

For an 18Mpixel result, the M11 sensor oversamples by a factor of about 1.82 (linear resolving power, 9528 pixels vs. 5212 pixels) compared to a standard 18Mp camera.

A definite upside to the M11, whatever final resolution is chosen.

 

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!

[jaapv]

An interesting comparison. The increase of aliasing vs the loss of acuity. It is comparable with the effect of an anti-alias filter.  TANSTAAFL. 

[SrMi]

8 minutes ago, jaapv said:

An interesting comparison. The increase of aliasing vs the loss of acuity. It is comparable with the effect of an anti-alias filter.  TANSTAAFL. 

Increased aliasing reduces resolution. If you see a loss of acuity in high resolution, then it is a problem with the test. Here are two examples of 24 vs 60MP, reduced to the same size (DPR studio scene):

 

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!

[adan]

Advertisement (gone after registration)

18 minutes ago, jaapv said:

The increase of aliasing vs the loss of acuity.

I would not read too much into that. Those pictures are not intended to show anything except the moiré - or lack thereof.

The original test was to determine how much usable ISO range the M11 added (if any). So these happen to be made at different ISOs, and had varied noise reduction applied. (M11 was at 40000, M10 was at  20000).

The discovery of the anti-aliasing effect was serendipitous. And was visible before any added post-processing.

And, of course the resampled M11 image is simply 33% lower resolution in the first place.

[jaapv]

Well, yes, the aliasing at of lower resolution sensors is well-known. Anybody taking images of birds is aware of the fact. But even with different ISO the loss of detail in your second shot is quite extreme. 

[Jon Warwick]

I agree that higher megapixels reduce aliasing. But even my 100mp GFX was prone to it. Less than my M11, but it would still occur for some subjects.

IMHO, the main way to fully eliminate it is multishot on the color sensor cameras (on the SL2S, I see less aliasing in multishot mode than native resolution on either my GFX100 or M11). And perhaps the truest fidelity of all, with no aliasing, is a Monochrom sensor.

[Kiwimac]

I generally leave mine at 60MP. I’ve tried the others and in normal use, I don’t really see much visible difference aside from when wishing to crop in post. Obviously the more resolution you have to start with, the more you will have in the final crop. 
 

I vaguely recall reading something when the M11 first appeared that theoretically the lower resolution settings should have less noise in them. I’ve not actually tested that. 
 

It’s a shame that there isn’t a setting to save the image at different DNG resolutions at the time of shooting, which would allow easy comparison in real world rather than test shot scenarios. 

[Derbyshire Man]

On 7/11/2025 at 2:45 AM, Kiwimac said:

I generally leave mine at 60MP. I’ve tried the others and in normal use, I don’t really see much visible difference aside from when wishing to crop in post. Obviously the more resolution you have to start with, the more you will have in the final crop. 
 

I vaguely recall reading something when the M11 first appeared that theoretically the lower resolution settings should have less noise in them. I’ve not actually tested that. 
 

It’s a shame that there isn’t a setting to save the image at different DNG resolutions at the time of shooting, which would allow easy comparison in real world rather than test shot scenarios. 

I don’t think that’s the way it works re noise as aren’t the additional pixels simply ignored rather than summed? I’d imagine the best way to reduce noise would be to shoot at 60mp and then shrink the image?

From a monochrome standpoint using silver efex the standard grain settings are way out for 60mp files so I usually turn the dial up to 11 if I want a more convincing film look but when I’ve experimented the best thing to do for triX or similar is to shoot at 18mp in the first place and add grain then. Could of course just crop/shrink first. If I could add a quick function to LR it would be duplicate file with resize. 

[SrMi]

54 minutes ago, Derbyshire Man said:

I don’t think that’s the way it works re noise as aren’t the additional pixels simply ignored rather than summed? I’d imagine the best way to reduce noise would be to shoot at 60mp and then shrink the image?

No, the pixels are not ignored. The noise is the same as when reducing the image in the post.

[Kiwimac]

Most reviews say that the DR is 14 stops at 60MP and 15 stops at 36MP.  Also that noise and colour at 18MP provides better results at high ISO settings. 
 

I’ve not really had much need (nor inclination, frankly!) to carry out tests. 
 

For anyone familiar with Linn hifi equipment, Ivor Tiefenbrun who founded the company has always said “If it sounds better, it is better”. Essentially, the ‘how’ of the improvement isn’t how you define whether it’s an improvement. You listen. 
 

Likewise, the same with cameras. The ‘how’ doesn’t really matter at the end of the day. To paraphrase Ivor, if it looks better, it is better. 

[SrMi]

1 hour ago, Kiwimac said:

Most reviews say that the DR is 14 stops at 60MP and 15 stops at 36MP.

I have not seen serious/competent reviews that say that (always compare DR at the same output size).

1 hour ago, Kiwimac said:

Also that noise and colour at 18MP provides better results at high ISO settings. 

I have not seen that either, nor can I observe that on my Leica cameras.

Leica has said that reduced resolution is the same as when resizing in Photoshop.

[Kiwimac]

4 hours ago, SrMi said:

I have not seen serious/competent reviews that say that (always compare DR at the same output size).

I have not seen that either, nor can I observe that on my Leica cameras.

Leica has said that reduced resolution is the same as when resizing in Photoshop.

From Chris at Phoblograper in their M11 review:

“The Leica M11 gains more dynamic range with lower megapixels. At 60MP, you’ll get 14 stops of dynamic range. AT 36MP and 18MP, you’ll get 15 stops of dynamic range. That’s really cool.”

[Smudgerer]

Well I've done this before now on this subject but risking on going brassic I will just throw another two-pennies-worth into the mix.

I've the M11-D and the M11-M and despite my initial resistance to trading up from my M10-R and the M10-M I have been very content with the move, not only because of the 11's sensor but the other things it bought to the table as well, battery, the choice of resolutions, USB-charging, and yes the new base too that the traditionalist in me didn't think I'd like, but now I do.

This "resolution thing" has been beaten to death, it really comes down to personal choice based on the final images and the "look" of the images that each would prefer. It's as simple as that, everything else is just a lot of waffle.

Soon after getting the M11-D I experimented with the three resolutions and quickly decided that I liked the 18mp setting more than the other two, when the M11-M came along the choice was more pronounced for me because that camera's RAW B&W output was flatter / smoother than my previous M10-M's files and I for one like to have some texture in my prints, ( I always shoot to print not to, excepting the edit process, display on a screen in any form ), so that camera went to 18mps too and both M11's have stayed at that setting. For the images I make for my personal work I like the output at 18mp, and as I regularly print to A1 or +A1 size if the image deserves I have found that setting to more closely deliver the qualities that I prefer. That I am not using the higher resolutions that the 11's offer doesn't phase me and the technical this and that back and forth bantering is meaningless to me, if whatever any camera offers works for you, it works.....end of story.

[SrMi]

8 hours ago, Kiwimac said:

From Chris at Phoblograper in their M11 review:

“The Leica M11 gains more dynamic range with lower megapixels. At 60MP, you’ll get 14 stops of dynamic range. AT 36MP and 18MP, you’ll get 15 stops of dynamic range. That’s really cool.”

That is nonsense, unless not compared properly.

P.S.: the original source of that claim may have been Leica's initial PR blurb. They removed that claim later.

Edited July 14 by SrMi

[LBJ2]

The most interesting comparison I've seen so far claiming the 36/18MP resolution might have a slight DR edge over the 60MP file is a two year old video by the MathPhotographer in which he tested and compared the Q3's triple resolution. But at the same time, I don't think many on the forums agreed with his results. Personally, I have not found evidence in my own simple/basic comparisons with the M11 triple resolution files, but MathPhotographer certainly carries out an impressive approach to test and compare IMO:

 

 

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!

 

 

[01af]

Am 10.7.2025 um 17:38 schrieb Ne314satel:

The second is more serious—if you shoot with old lenses then reducing the resolution will give a slightly different rendering.

Yes—different. Which means: worse.

.

Am 10.7.2025 um 17:38 schrieb Ne314satel:

I use reduction for optics from the '40s.

Bad idea. Not recommended.

.

vor 17 Stunden schrieb Kiwimac:

Most reviews say that the exposure range is 14 stops at 60 MP and 15 stops at 36 MP.  Also that noise and colour at 18 MP provides better results at high ISO settings.

That's from 'reviewers' who didn't review the camera but Leica Camera's original press releases. Those who actually own, and use, the Leica M11 know that it is not true. Switching from 60 MP to 36 or 18 MP won't provide any advantages whatsoever except smaller files—at the expense of lower image quality. Seriously: very bad idea. Definitely not recommended.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[SrMi]

55 minutes ago, LBJ2 said:

The most interesting comparison I've seen so far claiming the 36/18MP resolution might have a slight DR edge over the 60MP file is a two year old video by the MathPhotographer in which he tested and compared the Q3's triple resolution. But at the same time, I don't think many on the forums agreed with his results. Personally, I have not found evidence in my own simple/basic comparisons with the M11 triple resolution files, but MathPhotographer certainly carries out an impressive approach to test and compare IMO:

 

 

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!

 

 

He obviously measured DR at the pixel level instead of at the (same-sized) output level. Also, Leica removed any initial claim that lower resolutions increase DR.