Jump to content

M10 files come out darker/underexposed than other cameras?


scottee320

Recommended Posts

Advertisement (gone after registration)

Executive summary

Now that camera makers make (or choose) their own "film" - the sensor and all of the processing that occurs between the sensor and the file actually saved to the SD card - there is no longer a rational expectation that "ISO xxx" will produce identical brightnesses with identical exposure (shutter/aperture) from two different camera models. At least not SOOTC (straight out of the camera).

My experiments with my M10s show that its "ISOs" comply reasonably well with the requirements of the ISO standards. Specifically, it will meter and reproduce a gray card at a brightness of 121 using "base" settings in Adobe Camera Raw (all sliders set to "0"). And providing quite considerable shadow detail.

That does not mean its pictures, especially in .DNG format, always look correct SOOTC.

What is digital ISO? Mostly, whatever the camera maker subjectively chooses.

The ISO (International Organization of Standards) provides at least five methods of determining ISO. That is in ISO specification 12232:2006.

See a little over halfway down the page for the digital discussion: https://en.wikipedia.org/wiki/Film_speed

Three of them apply only to jpegs, since they must be measured from a "finished image" - i.e. a viewable 8-bit jpeg created in the sRGB color space. They are meaningless when considering a 12/14/16-bit "raw" or .DNG image. Which is not a measurable image until the photographer has done his/her post-processing, and converted it into an 8/16-bit image in sRGB (most of us use Adobe1998, some use ProPhoto).

Two of them can be applied to "raw" or .DNG files, but they are rather esoteric measurements of signal-to-noise ratio. And the test images must be of a scene with "average illuminance" - (your guess is as good as mine what that means ;) ). And there is yet more wiggle-room - an "acceptable" image need only show an SNR of 10:1; an "excellent" image will produce an SNR of 40:1.

https://en.wikipedia.org/wiki/Signal-to-noise_ratio_(imaging)

Without know exactly how Nikon determined their "ISO 400" and Leica determined their "ISO 400" - and whether those determinations were according to the three "jpeg" techniques/standards for consumer cameras, or the two other techniques for raw/.DNG images - and how the voltage flowing off the CMOS chip is being converted to digital values, there is no way to say that "the sensor" is somehow less sensitive to light.

The general subject of studying and measuring how photographic materials and devices respond to light and produce an image (sensor or film; silicon, silver, or dichromate, etc) is called "sensitometry."

What about a medium-gray card? Isn't that a standard?

Well, sort of. A gray card can be used to determine ISO several ways - even with film.

The "official" ISO technique (roughly similar to Ansel Adams' Zone System technique) is to meter the gray card, underexpose it 3-4 stops, and find out which ISO on the meter will result in a film density just distinguishable from unexposed film (film base + fog). For those who like hard numbers, a density of 0.1 more than the fb+f density.

However Ilford, for example, specifically says they do not use that ISO standard - they use a subjective "practical evaluation" - i.e "looks good to us." ;)

Quote

It should be noted that the exposure index (EI) range recommended for HP5 Plus is based on a practical evaluation of film speed and is not based on foot speed, as is the ISO standard.

https://www.ilfordphoto.com/amfile/file/download/file/1903/product/693/

Photographers may also base their exposure on the approved technique - or also on "practical considerations," such as when push-processing film to get a higher Exposure Index (not ISO) in dim light. Where ideally a medium-gray still images as a medium gray in between the darkest and lightest tones, even if the highlights blow out or the shadows/"foot speed" are blank.

There is also a question of "what, if anything. is a medium gray?" There are at least six definitions of medium gray.

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

My M10 and its meter correctly (for a given value of correctly) reproduce a gray card as brightness 120-121, when metered for the gray card (and checked against an incident reading from a hand-held Sekonic). Right in between the L*a*b gray value (119) and the anticipated gray-card gray value (124). That will be darker than the value in sRGB, and substantially darker than a "patch" that is exactly half white pixels and half black pixels (absolute whiteness = 50%). But it is not incorrect.

As a practical example, here is a recent M10 picture (135mm Tele-Elmar, no exposure compensation set). The way it came out the camera, (Adobe sliders all at "0") and with appropriate post-processing.

It is clear from tests that the M10 does apply a rather strong default gamma curve to pictures, in an attempt to regain the punchy contrast and saturation of M9 CCD images. There is far more shadow detail than there appears to be, SOOTC. This of course can be removed by the photographer, if they want to get the fullest dynamic range the camera/sensor is capable of.

Since I have done my sensitometry of the M10, this was expected.

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 by adan
  • Like 3
  • Thanks 1
Link to post
Share on other sites

@adan and @Kwesi thank you both so much! It does make sense that ISO is to some degree a subjective value that varies from manufacturer to manufacturer, and it also makes sense that the Leica may be aware that the m10 sensor does much better in shadows than highlights, and so the metering nudges you in the direction of "underexpose and dodge afterwards."

 

It's also reasonable to think "well, ISO 640 on an m10 might look just as good on that sensor as ISO 200 does on the d850 sensor." (Although I won't allow any besmirching of that camera, it's my main workhorse and a real powerhouse!)

I do live just a 30 minute train ride from Leica SoHo so that'll be a fun little fact-finding mission to take up soon. And I'm still real curious to see how the same sort of "identical settings pointed at the same thing, shot within the space of a few seconds" test works out when A/Bing the m10 and m6 with Portra...

Link to post
Share on other sites

According to DXO mark M10 iso rating is much below the rated value, about half a stop below D850 (which is same as M240). On top of it the way Leica implement the default contrast curve will affect the rendering too. There were discussions about M10 having much deeper shadows than M240 straight out of camera when it first came out.

In the real world it may or may not be a concern, take ISO value lightly and you'll enjoy the camera more. You have to shoot it for a while to have a feel if the camera is capable to shoot comfortably in low light or not. I find the M240 shoots comfortably with a summicron lens except scenes that's so dark that's not possible to make out the content with the naked eyes (which is pointless to point your camera at anyway). The M10 should perform better and this shouldn't be a concern.

  • Like 1
Link to post
Share on other sites

some interesting info about ISO here, but the main reason is that the Leica is using effectively wide spot metering whereas your Nikon’s are using matrix metering.

if you point the Leica at a light source and all around is dark, it will correctly exposure the centre (basically around the focus patch) and the rest will be very dark

i always meter with the M using the centre patch for the areas I want mid, then hold the shutter half way down which locks the exposure, and recompose. It’s an easy technique once you get used to it. 
 

I do a lot of night shooting and the thing I really don’t like about matrix is that it makes dark scenes too light and also drops the shutter speed and raises the iso too much. 
 

some M10 examples :

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 by colonel
  • Like 4
Link to post
Share on other sites

Fujifilm also overrates their ISO relative some other manufacturers. In the Fujifilm forums/Facebook groups, posts pop up regularly from people who own both Fujifilm and some other brand that are confused when they compare them both at the same settings.

ISO for sensors is a bit like aperture numbers for lenses, but at least for lenses we have T-stops.

Link to post
Share on other sites

Advertisement (gone after registration)

On 3/2/2021 at 9:56 PM, colonel said:

some interesting info about ISO here, but the main reason is that the Leica is using effectively wide spot metering whereas your Nikon’s are using matrix metering.

Yes, but the OP is about two images made at identical manual settings... Metering was not involved.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...