M9 constantly underexposes

[Maximus]

Advertisement (gone after registration)

Hi Maximus,

 

May I please ask which dSLR system you have been using prior to your M9?

 

Thanks,

 

Rick

Sure. I still have a Canon 1Ds MkIII and a 5D MkII.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[diogenis]

And it's not just the cameras.

We also need a good monitor correctly calibrated for proofing colors and tonal balance.

And then the same for whatever printer we use.

I could easily bet, his meter is ok. Leica's QC would catch a 1 stop difference easily.

[Jeff S]

I agree with Lars' point about using an incident meter. Then compare shots using the incident meter with another using your standard routine. If the results are consistent, then Andy's advice is also wise. There are many links in the chain, and one needs to isolate them in order.

 

The good news is that you seem to be consistent. Therefore, compensation is easy in several ways. But, this method will better get to the root cause.

 

Jeff

[Jamie Roberts]

Maximus...

 

According to the EXIF information in the shots you posted, you have -1/3 EV dialled in as exposure compensation. I know that's not a lot, but it will make your shots darker.

 

I recommend you don't dial in negative compensation, and I'd check to make sure, as Lars suggests, against an incidence meter how the camera actually checks against the meter first.

 

FWIW, my M9 seems a bit less sensitive to light than my M8--perhaps a third of stop. So add that level of sensitivity to another - third EV compensation and your shots will be dark, a lot of the time.

[jbgeach]

Maximus, if you pictures are consistently dark, just dial in +1/3 to +2/3 EC and forget about it. That is the reason it is a user selectable option. We all have different taste and this is a problem easy to remedy.

[zlatkob]

I don't think this is a question of personal taste or monitor calibration. A simple test is to set the camera to A mode and no exposure compensation, and photograph an even surface like a white or gray cabinet or wall. The resulting photo should have a histogram bunched up in the middle, something like this. If the histogram is significantly off-center, then something is wrong with the meter or camera.

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!

[lars_bergquist]

Advertisement (gone after registration)

Lars, I understand that. So why are ALL my images under exposed?

You have to do a controlled test. You can't just snap away and judge the results one by one. By now you know what you have to do to set up such a test. Do it. If there is a consistent, actual underexposure, i.e. the 'spike' from a neutral medium 18% grey surface is consistently displaced to the right on the histogram, then you have a consistently mis-exposing M9.

 

But that would be the first case in my knowledge.

 

The old man

[lars_bergquist]

I don't think this is a question of personal taste or monitor calibration. A simple test is to set the camera to A mode and no exposure compensation, and photograph an even surface like a white or gray cabinet or wall. The resulting photo should have a histogram bunched up in the middle, something like this. If the histogram is significantly off-center, then something is wrong with the meter or camera.

 

Didn't see Zlatkob's posting. He is right. Follow his advice.

 

He is also right that you an use ANY even surface. A white piece of cardboard e.g. will be exposed as if you had used a grey card and produce that sharp spike. In order to see it firsthand, look at the camera histogram. That will tell you directly what the camera does, and not leave any lingering doubt about Photoshop etc.

 

The old man

[Maximus]

Thank you for the advice.

 

Here's a shot just taken of plain white paper. I took 3 to be sure and the all have the same shaped histogram - right edge almost vertical and over the "1.00" scale with the left edge more gradual to the left.

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!

[Maximus]

Here's 2 more for good measure.

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!

[diogenis]

it seems you got a perma error. I suppose you did check that compensation is 0?

[Maximus]

it seems you got a perma error. I suppose you did check that compensation is 0?

Yes, there is no exposure compensation set.

[diogenis]

Well, add a +.5 ... +1 EV and have the camera checked at your next convenient opportunity. But before, better get sure that the surface you are testing is 18% reflective.

 

i meant Grey

[wlaidlaw]

Compared with the M8, I too find quite a lot of my images are technically a touch underexposed, maybe by half a stop. However there is so much detail in the darkness (what is the opposite of headroom - perhaps bottomroom!) that it is not a problem. I would much rather have to do a little post production tweak, which in C1 can be batched, than have to try and recover detail from blown highlights. The only downside is that it can result in slightly more noise in the dark areas, by the time you have corrected. That is where I sometimes wish I had a larger histogram to read on the LCD, when I am taking photos in low light. I quite often end up using my Polaris spot meter and manual on the M9.

 

Wilson

[jaapv]

I find the same, Wilson. Of course that is a difference to exposing for film. As digital has a sharp cutoff in the highlights and film more or less in the shadows, it is wise to expose a digital image tighter than a negative film like we used to do with slide film. I think Leica made a wise choice with this bias on the M9. Most users had their M8 set to underexpose by 1/3rd or 2/3rd for the same reason.

[Guest imported_torben]

..... Leica's QC would catch a 1 stop difference easily.

...

Leica's WHAT?

[250swb]

I've been looking through an average selection of images that I know I wouldn't have made any exposure changes to, and I'd say my M9 under exposes by maybe two thirds of a stop under normal conditions. The histogram is always biased off to the left if I just let the camera meter the scene with no compensation. Which is pretty good in my book. As Andy has already said, ACR deals with it easily (just create a profile for the M9), and there's bags of detail in the shadows. I just don't see the difference between what could be called perfect exposure (with spot meter and grey card) and what the camera reads as being a problem, just a characteristic of the meter that needs to be understood just like a lens's character.

 

Steve

[Maximus]

I've been looking through an average selection of images that I know I wouldn't have made any exposure changes to, and I'd say my M9 under exposes by maybe two thirds of a stop under normal conditions. The histogram is always biased off to the left if I just let the camera meter the scene with no compensation. Which is pretty good in my book. As Andy has already said, ACR deals with it easily (just create a profile for the M9), and there's bags of detail in the shadows. I just don't see the difference between what could be called perfect exposure (with spot meter and grey card) and what the camera reads as being a problem, just a characteristic of the meter that needs to be understood just like a lens's character.

 

Steve

Not sure I agree with this Steve. If a meter constantly underexposes that isn't character, it's calibration, surely?

[Advertisement]

Hallo,

Look here: Leica M

[Jamie Roberts]

Not sure I agree with this Steve. If a meter constantly underexposes that isn't character, it's calibration, surely?

 

Actually, I've always had to calibrate an internal meter w/ film and what an incident meter says, in part of course because different films respond differently in terms of density and colour.

 

With digital, I'm much more interested in holding highlights, but I also don't want to underexpose. So I check the camera against an incident reading designed to essentially give me RGB 242/242/242 for a highlight. Then I compensate... usually manually in my case, since I don't like auto exposure, but you can absolutely adjust with EC too.

 

My Canons were different than my Nikons, and the M8 was more sensitive than both of them (between the D3 and the M8 there's almost a third of a stop difference).

 

But they all vary. If you feel yours is really off by more than a stop, then you could have it adjusted; personally, I'd just dial in +1 and forget about it.

[diogenis]

I've been looking through an average selection of images that I know I wouldn't have made any exposure changes to, and I'd say my M9 under exposes by maybe two thirds of a stop under normal conditions. The histogram is always biased off to the left if I just let the camera meter the scene with no compensation. Which is pretty good in my book. As Andy has already said, ACR deals with it easily (just create a profile for the M9), and there's bags of detail in the shadows. I just don't see the difference between what could be called perfect exposure (with spot meter and grey card) and what the camera reads as being a problem, just a characteristic of the meter that needs to be understood just like a lens's character.

 

Steve

 

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.