M9 constantly underexposes

[zlatkob]

Advertisement (gone after registration)

I re-did my test with a white sheet of paper, camera set to A mode, and exposure compensation settings of -1, zero, and +1. I processed the files in Adobe Camera Raw with all ACR settings set to default. Here are the histograms below. It seems accurate, or at least close enough.

 

Maximus, your camera seems to be a slightly biased toward underexposure, but it does not appear to be too far off — at least not enough to explain the serious underexposure in the photo of the couple in your first post. If your camera were a full stop off, your test histogram would be even more to the left (something like my -1). I am guessing your meter is close to accurate (maybe -1/3 off). If you're testing with the aperture wide open, vignetting will likely move the histogram to the left a little.

 

Be sure your raw converter is not messing things up. It seems you still need to compare with another meter or camera that you trust, as Lars suggested, to try to isolate the problem.

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!

[250swb]

But if you expose "perfectly" you get even more "bags" of shadows. Usually when I try to photo a highly contrast scene, I expose to the limit of burning highlights. 1% of burns for example won't be ever noticed. but in turn this will show me detailed shadows. Now, as Maximus says, if this is happening in all our photos, then chances are that indeed it is a calibration issue.

 

But only if we know what Leica calibrate the meter for in the first place (what type of 'average scene' do they have in mind) can we know if any meter needs calibrating. Its an average meter, its not meant to be pointed at grey cards in particular, its just meant to be 'average'. In the UK the intensity of the light and contrast range is somewhat different to that of the South of France for instance, and exposure does change for your average scene. Thats why you need to learn what your meter is telling you, there isn't a definitive 'right or wrong' when using a crude meter, its always going to be about interpreting the meter readings.

 

Steve

[lars_bergquist]

But only if we know what Leica calibrate the meter for in the first place (what type of 'average scene' do they have in mind) can we know if any meter needs calibrating. Its an average meter, its not meant to be pointed at grey cards in particular, its just meant to be 'average'. In the UK the intensity of the light and contrast range is somewhat different to that of the South of France for instance, and exposure does change for your average scene. Thats why you need to learn what your meter is telling you, there isn't a definitive 'right or wrong' when using a crude meter, its always going to be about interpreting the meter readings.

 

Steve

Meters are not calibrated for this or that scenery, in this or that lighting. The industry standard is a neutral grey surface with a reflectance of 18% (a.k.a. Kodak Gray Card). So it does not matter if you do the test in Darkest Peru or anywhere: Whatever light there was from the beginning, the card will reflect eighteen percent of it. Fifty years ago, before electrical meters became common, I remember there was lots of superstition that denied this of course, mostly because people did not understand the difference between absolute illumination and contrast.

 

The general level of illumination changes all the time of course, not only between different places on the globe, but also depending on time of day (natch!) and weather. Main factors are:

 

a) Height of sun above horizon;

Condition of atmosphere;

c) Shadow.

 

The 'sunny sixteen' rule applies, to an approximation, if the sun is more than 20 degrees above the horizon, you are out in the open in direct sun directly behind you, and the sky is not hazy or cloudy. Not ideal light for a photo, and in fact conditions that apply only in the middle of the day, where I live, and if the weather is preternaturally fine. Hence the incident meter. An exposure at the speed of 1/ISO would be closer to 11 on a god day.

 

The basic scientific testing procedure is to nail down all variables, preferably quantitatively, and then change one variable at the time. This means you can never be certain without an independent outside control (hence, again, that Gossen or Seconic). What Jaap said about digital exposure being much like slide film, is correct: No exposure latitude! With chromes, the piece of film that you exposed was also the piece that you projected on the wall. With digital, the signal that you get out of the sensor is all the signal you will ever have. Negative film had latitude because the neg was a halfway house: Under the enlarger, you could use that long 'shoulder' of the emulsion to produce, by sleight of hand and different contrast grades of paper, identical or near identical prints from negs that were several f-stops of exposure from each other. Those days are gone. You nail the exposure, or you pay the price.

 

The old man who came out of the darkroom, reeking hypo

[jaapv]

18% gray is for film, Lars. According to LFI, for sensors this is no longer valid. fwiiw....

[250swb]

Yes Lars, of course meters are not calibrated for this or that scenery, but they are used differently depending on this or that scenery. As long as you know how your meter responds thats all you need to know. It doesn't need 'calibrating' unless it is wildly off.

 

That is the whole point of the Zone System, you test everything as a whole process, from your meter to the printing paper and developer, and its your datum point. Its a rigid system, but it doesn't matter if your meter reads half a stop under or over compared to somebody else's meter because its factored in to your results. Its not so different with digital, you find what works and its foolish to imagine somebody elses meter reading is going to work 'exactly' for you because of all the variables involved. You need to know what you have pointed the meter at, you need to know what RAW software you are going to use, and etc.

 

Steve

[AlanJW]

Digital may be like slides in the old days but I do not agree that if you don't nail the exposure, you are cooked. We didn't have post processing engines that allowed you to "develop" different parts of a frame in different ways. I may make fewer mistakes now than when I was younger (I am not as old as Lars but have been at this for 51 years) but I still make plenty. The good news is they are easier to fix.

[Jamie Roberts]

Advertisement (gone after registration)

18% gray is for film, Lars. According to LFI, for sensors this is no longer valid. fwiiw....

 

Do you recall how or why a middle gray reading would change for digital, Jaap? (again, I'm more interested in calibrating for highlight detail, personally).

[jaapv]

I'd have to look for the article - it was last month I think...

[diogenis]

But only if we know what Leica calibrate the meter for in the first place (what type of 'average scene' do they have in mind) can we know if any meter needs calibrating. Its an average meter, its not meant to be pointed at grey cards in particular, its just meant to be 'average'. In the UK the intensity of the light and contrast range is somewhat different to that of the South of France for instance, and exposure does change for your average scene. Thats why you need to learn what your meter is telling you, there isn't a definitive 'right or wrong' when using a crude meter, its always going to be about interpreting the meter readings.

 

Steve

 

Steve, the meter calibration is left for Leica to solve. Not us.

Average is also a grey card, like 18% grey. Anyway it doesnt really mater.

Lighting schemes may change when one moves from North of Europe to South, but this is also irrelevant. We are interested in big contrast differences here. And you bet, you can find those both in France and Britain. I dont get it what is it so difficult about meters... It is there to help you in exposure. You can still evaluate the scene and lean towards more shadows or highlights

[Jeff S]

I'd have to look for the article - it was last month I think...

 

Jaap, you might be referring to the article that appeared two issues ago (last month focused on color...colour...I think).

 

I thought that the point made in the prior issue was that you could trust using a mid-tone value (18%) in film to yield a proper overall exposure given the film curve. But, in digital, one can no longer trust this 18%...or midtone exposure...theory, i.e., one has to understand the sensor tendency to sharply cut off highlights, so that manual compensation is required (shifting the histogram appropriately) in order to maintain full dynamic range.

 

Perhaps my words aren't perfect, but this was the gist of the discussion as I recall. If so, then the point wasn't that sensors (or meters) are designed at something different than 18%; but rather, the very nature of digital requires us as photographers to understand sensor characteristics and adjust our practices accordingly.

 

Or, I might not know what I'm talking about.

 

Jeff

[Maximus]

So, do you think my meter would benefit from recalibration?

[Jamie Roberts]

So, do you think my meter would benefit from recalibration?

 

Who are you asking?

 

Go shoot a gray card, check it against an external meter as suggested (or check an incidence meter for highlights on a white subject).

 

If the discrepancy between your camera and the meter is a stop or less, I wouldn't bother having Leica look at it. If it's more, then it might be worth sending in.

[Jeff S]

So, do you think my meter would benefit from recalibration?

 

The question is whether you might benefit.

 

Seriously, though, the goal should be to consistently get well exposed photos, which for me means well exposed prints. The key is understanding your equipment and technique well enough to achieve this. If you've properly isolated the variables (including those further down the processing chain as Andy said earlier), then how you get there is less important than just getting there...consistently.

 

If you can achieve this by dialing in some compensation, or adjusting your technique, then I wouldn't get hung up on the meter (unless it's erratic). This is no different than in the film days when we tried to get the whole chain...from meter to film speed to development times to paper contrast, etc...in synch.

 

Jeff

[250swb]

Steve, the meter calibration is left for Leica to solve. Not us.

Average is also a grey card, like 18% grey. Anyway it doesnt really mater.

Lighting schemes may change when one moves from North of Europe to South, but this is also irrelevant. We are interested in big contrast differences here. And you bet, you can find those both in France and Britain. I dont get it what is it so difficult about meters... It is there to help you in exposure. You can still evaluate the scene and lean towards more shadows or highlights

 

But I never mentioned grey cards or 18% or test targets. I was talking about metering real life images, and those are determined by the size of the metering patch, distance from subject, and to a great extent the lens. A combination of those form the meters response to an average scene. The OP started off by showing a real life image, and thats the important point, practical metering not test cards. But without an idea what 'average' is its not possible to all sing from the same hymn sheet in anything other than photographing test cards, which teaches very little.

 

For instance, a 135mm lens used for a portrait and you may well get a reading directly off the face, with skin filling all the metering area, and this skews the true exposure. Put on a 21mm lens the person becomes a figure in the landscape and again the meter can read off an inappropriate tone, like sky. So if my meter 'apparently' reads half a stop less (left to its own devices) than yours it doesn't mean it needs calibrating for many reasons, but the first is that I may simply be using different lenses in different situations.

 

Steve

[adan]

I just have to ask this question of Maximus:

 

Has it occured to you that your M9 meter might be fine - and that it is the default settings in LightRoom that are in error and in need of calibration?

[Maximus]

I just have to ask this question of Maximus:

 

Has it occured to you that your M9 meter might be fine - and that it is the default settings in LightRoom that are in error and in need of calibration?

Yes it had. Are you saying the Lightroom M9 profile may be at fault?

[innerimager]

The M9 never underexposes. No camera does...or doesn't. The photographer gets the exposure right, or wrong. The photographer learns how the meter works, (in camera or not), and adjusts accordingly. With the feedback of the histogram, this is not that hard to do. But fortunately, also not so automatic as to rob the process of the need to develop skills and utilize them.

[wlaidlaw]

The M9 never underexposes. No camera does...or doesn't. The photographer gets the exposure right, or wrong. The photographer learns how the meter works, (in camera or not), and adjusts accordingly. With the feedback of the histogram, this is not that hard to do. But fortunately, also not so automatic as to rob the process of the need to develop skills and utilize them.

 

OK but the meter is a measuring tool like a ruler. If a ruler is graduated wrongly, it would for sure, make life difficult for a draughtsman. The histogram on the M9 is not too easy to use due to its small size. Steve also made a very good point about the effect of certain lenses. The 135 T-E for example, is well known for producing somewhat over-exposed images - why - no idea. Again this was worse on the M8 than the M9, on which they are about right, with the inherent tendency to underexpose offsetting the lens effect. This is just one more problem you may have to remember to compensate for mentally. The advantage of digital is that you can get some idea immediately by chimping/histogram peering but only rather crudely. This is nothing new - I well remember the horrible difficulty in getting IR film exposures anywhere near correct, whatever meter or compensation you used.

 

Wilson

[Jeff S]

I just have to ask this question of Maximus:

 

Has it occured to you that your M9 meter might be fine - and that it is the default settings in LightRoom that are in error and in need of calibration?

 

Precisely, Andy, why I referenced you twice in my responses that the OP needed to account for the entire processing chain.

 

Jeff

[Advertisement]

Hallo,

Look here: Leica M

[zlatkob]

Make sure Lightroom is set to defaults as shown here (outlined in red) for testing purposes. You can change the default any time, but the original default is good as a reference point. If you have it set this way and still have nearly all of your photos underexposed, then you should really compare your camera with another meter or camera that you trust. Any discrepancy between the two should be small.

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!