M10M & M10-R dot artifacts in the DNGs

[hdmesa]

Advertisement (gone after registration)

Anyone else here seen this or able to duplicate the issue – black dots with white rings around them? The Q2M sensor doesn't have this issue, and it may move me toward the M10 as a path to the M system instead of starting with the M10-R.

https://diglloyd.com/blog/2021/20210115_1132-LeicaM10M-BlackDotsOnWhitePimples-CookingRawData.html

Please keep the responses thoughtful and focused on the issue and not the reviewer. Thanks.

[t00l1024]

Was discussed here:

 

[Chips]

On 1/17/2021 at 4:15 PM, hdmesa said:

Anyone else here seen this or able to duplicate the issue – black dots with white rings around them? The Q2M sensor doesn't have this issue, and it may move me toward the M10 as a path to the M system instead of starting with the M10-R.

https://diglloyd.com/blog/2021/20210115_1132-LeicaM10M-BlackDotsOnWhitePimples-CookingRawData.html

Please keep the responses thoughtful and focused on the issue and not the reviewer. Thanks.

If it is a firmware issue could it be adjusted through a software update ?

[Mark II]

The Canon 5DII had a similar issue at launch and it was addressed with a firmware update. Maybe this can be too.

FWIW, you can see this in Reid Reviews comparison images between the M10M and other cameras. The burned-out specular highlights tend to have a small black dot. Not sure that it is likely to be significant in real world use.

[hdmesa]

9 hours ago, Mark II said:

The Canon 5DII had a similar issue at launch and it was addressed with a firmware update. Maybe this can be too.

FWIW, you can see this in Reid Reviews comparison images between the M10M and other cameras. The burned-out specular highlights tend to have a small black dot. Not sure that it is likely to be significant in real world use.

I don’t think it is an issue for street, portraits, and general shooting; however, for landscape where the photographer is looking to emulate high contrast film on detailed subjects, this could be a issue. For example, emulating TMAX 100 with a deep red filter printed to high contrast paper requires liberal use of the dehaze tool along with increasing contrast, compressing levels, etc., which increase the odds these artifacts may show up.

[dkmoore]

21 hours ago, hdmesa said:

I don’t think it is an issue for street, portraits, and general shooting; however, for landscape where the photographer is looking to emulate high contrast film on detailed subjects, this could be a issue. For example, emulating TMAX 100 with a deep red filter printed to high contrast paper requires liberal use of the dehaze tool along with increasing contrast, compressing levels, etc., which increase the odds these artifacts may show up.

Or use a red filter. 
 

also, if you are trying to mimic T100 film you won’t be able to tell the grain from this supposed artifact anyway. 

I have not noticed this issue with over a year of shooting and who knows how many thousands of images. I’m sure it exists but if you have to brutalize your file to show it I can live with that. 

[hdmesa]

Advertisement (gone after registration)

14 hours ago, dkmoore said:

Or use a red filter. 
 

also, if you are trying to mimic T100 film you won’t be able to tell the grain from this supposed artifact anyway. 

I have not noticed this issue with over a year of shooting and who knows how many thousands of images. I’m sure it exists but if you have to brutalize your file to show it I can live with that. 

Um, did you miss me saying deep red filter? So many people skim read and respond  Point is the monochrome sensors need more than just the red filter to mimic the dark skies and high contrast of some films – Dehaze, contrast, levels all required for that. If that's not a style you shoot for, then of course you'll probably never see it.

[dkmoore]

14 minutes ago, hdmesa said:

Um, did you miss me saying deep red filter? So many people skim read and respond  Point is the monochrome sensors need more than just the red filter to mimic the dark skies and high contrast of some films – Dehaze, contrast, levels all required for that. If that's not a style you shoot for, then of course you'll probably never see it.

Hello, I did not mean my post to be confrontational but I do think I read it correctly. You made two points in your original post:

1. The dots could be a bigger issue shooting landscape when trying to emulate film - I disagree because the grain will hide or cover up the dots most likely

2. You state that if you want to emulate Tmax film with a red filter on DIGITAL that you would need to use the dehaze tool, contrast slider, levels, etc. - If you use the red filter on digital you won't have to apply as much dehaze. 

 

I think you skim wrote 😂

But, anyway, I think in the end we are aligned. 

 

[hdmesa]

1 hour ago, dkmoore said:

Hello, I did not mean my post to be confrontational but I do think I read it correctly. You made two points in your original post:

1. The dots could be a bigger issue shooting landscape when trying to emulate film - I disagree because the grain will hide or cover up the dots most likely

2. You state that if you want to emulate Tmax film with a red filter on DIGITAL that you would need to use the dehaze tool, contrast slider, levels, etc. - If you use the red filter on digital you won't have to apply as much dehaze. 

 

I think you skim wrote 😂

But, anyway, I think in the end we are aligned. 

 

Lol. Oh well.

I brought up film and specifically TMAX not because I wanted to emulate all of the TMAX characteristics but only as a way to try and describe using a red filter to get black skies. Maybe the M-Mono is different, but on the Q2M, even the 091 deep red filter I can't get black skies without quite a bit of pushing the files in post. If it's easier to get black skies on one of the M-Mono sensors, that might be worth a look on my part. Maybe the issue for me is I've not had deep blue skies since summer due to wildfires further out west.

[mmradman]

For the desired dark sky effect in mono blue skies works better than hazy. On the colour wheel red and blue are opposite and haze obliterates deep blues.

[dante]

I'm sure you've seen my treatment of Leica's "color balance" for Mono, but the Monochrom cameras essentially have a light green filter all the time. The sensor has very little response past the orange part of the spectrum. I'm sure the same is true of the M10M.

https://themachineplanet.wordpress.com/2015/09/16/the-leica-monochrom-typ-246-and-filters/

Adding red will not kill blues. It will kill everything. That's why it doesn't work. @mmradman, "minus blue" is a yellow-orange, not red. Red and green are complementary colors.

I'd suggest using a polarizer to address the sky.

[t00l1024]

"Pimples in the sand." 🤪

I couldn't help but think of this discussion yesterday hiking around. The light rain did nothing for the gypsum sand, but the fatter raindrops did. M10M + APO 50mm + Orange filter.

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!

[mmradman]

On 1/26/2021 at 3:25 PM, dante said:

I'm sure you've seen my treatment of Leica's "color balance" for Mono, but the Monochrom cameras essentially have a light green filter all the time. The sensor has very little response past the orange part of the spectrum. I'm sure the same is true of the M10M.

https://themachineplanet.wordpress.com/2015/09/16/the-leica-monochrom-typ-246-and-filters/

Adding red will not kill blues. It will kill everything. That's why it doesn't work. @mmradman, "minus blue" is a yellow-orange, not red. Red and green are complementary colors.

I'd suggest using a polarizer to address the sky.

Interesting.

i never obsessed much, I did try classic B&W filters on M246 and never fell in love with results when using Red filter. Orange on the other hand  is pretty much my standard filter or no filter. 

As of recently I have added H&Y filter holder aiming to use soft graduate neutral, as it also comes with polariser will definitely start experimenting once Covid restriction get relaxed and weather improves.

[t00l1024]

8 minutes ago, mmradman said:

Interesting.

i never obsessed much, I did try classic B&W filters on M246 and never fell in love with results when using Red filter. Orange on the other hand  is pretty much my standard filter or no filter. 

As of recently I have added H&Y filter holder aiming to use soft graduate neutral, as it also comes with polariser will definitely start experimenting once Covid restriction get relaxed and weather improves.

My workflow/usage isn't too different.

Indoor - No filter

Everyday outdoor - Yellow

Need for "drama" in landscapes - Orange

Red - Stays at home in the photography storage pile.

[mmradman]

46 minutes ago, t00l1024 said:

My workflow/usage isn't too different.

Indoor - No filter

Everyday outdoor - Yellow

Need for "drama" in landscapes - Orange

Red - Stays at home in the photography storage pile.

Today was a good day. I have now dug out all Yellow filters in my stock, and put them together in large filter boxes to be handy. I happened to have them is same sizes as my Orange - 39, 46, 49,55 and 67mm. Years of amassing miscellaneous is paying off 😅.

As I stated I was never  impressed with Red filter on M246, it definitely behaves differently from film. As the whole thread is reaction to DigI Lloyd observations I have observation of my own. He is using deep red filters on M10M and pictures don’t look right to me (free blog only), but hey he is the expert so I leave him to his devices.

 

[davidbaddley]

Amusing; this thread began by asking about those nasty little artifacts and asked to stay on topic, then quickly went down the road of color filters.

Has anyone ever figured out what's going on with these artifacts (white spots with dark spots inside of them)? My standard sharpening protocol made them really stand out. I sharpen differently now so that they aren't as apparent, but they're still there if you know what you're looking for and it isn't very "natural."

I don't get them with softer lenses like a Jupiter 3, but they're all over the place with sharper lenses like the Summilux 50 1.4 ASPH, mostly in areas of repetitive texture but even along clean edges like a roofline against the sky.

I used an M9M for years without anything like this. I got this M10M because I bought into the idea that every pixel corresponds to a site on the sensor and is a true representation of the light that hit it and not the result of interpolation. Is this just a myth? How can these artifacts be anything but the result of interpolation (gone wrong)?

I'd love to hear whatever insights you might have. I've heard this topic mentioned several times, but only superficially. Has anyone figured out what causes these artifacts or if there is a fix? Even if there isn't a fix, I'd still like to understand what's happening. Thank you.

 

 

 

[jaapv]

I have no proven explanation but I have a suspicion. We need to look at this holistically. When the subject detail frequency and lens resolution approach the sensor Nyquist frequency you will get localized aliasing. On a Bayer sensor this will be negated by the demosaicing. 

[davidbaddley]

Interesting idea, best I've heard so far. Thanks. Since Leica prefers software solutions (rather than filters) for anti-aliasing, couldn't this be addressed with a firmware update?

 

[thighslapper]

Must be a really big problem if there are only 18 posts about this in 2 years.

I can't find it in any of my images

Even DGL only sees it at 2x magnification..... and with 'blown out' specular highlights. 

It would vanish in a print ..... even at fairly enormous size, given a 40mpx raw file. 

 

[Advertisement]

Hallo,

Look here: Leica M

[davidbaddley]

Glad to hear that you haven't experienced this frustrating issue. I suspect that it's lens-dependent, which might explain why it isn't experienced by more of us. It only happens to me when using the Summilux 50 1.4 ASPH, but that is my favorite lens which I use more than any of my others, and it is one of Leica's more popular lenses.

I think the "blown out specular highlights" thing is a mistake; although the spots themselves might look like tiny specular highlights with strange black dots in them, they occur as often, if not more often, in midtone areas of textural contrast.

This isn't enough of a problem that it keeps me from using this camera (if it was, I would have returned it), but for those of us who experience it, it isn't insignificant either. I'd rate it as on-par with issues like fringing, sensor corrosion, color-shifts in vignetting, etc. It is noticeable in prints, sometimes looking like tiny dust spots, or regions of unnatural texture, similar to but not exactly like over-sharpening or even reticulated film grain (but finer). Again, it is more noticeable with some sharpening protocols than others.

I initially assumed that this was some software strategy by Leica designers to add a small black dot to the center of small white spots to make them less noticeable while preserving sharpness, but this didn't explain where the white spots were coming from in the first place. I like Japp's theory better. I see how this could be the result of some sort of wave interference.

For those of us who experience this, it is a REAL thing. It seems too consistent to be an anomaly, but maybe my camera is just defective?

 

 

Edited February 20, 2023 by davidbaddley
typo