5D mark III metering problem

I bought my 5dm3 months ago but wasn't really shooting much with it. I just noticed recently that my metering is off. I have to always offset my exposure to +1 when its metered at 0. I noticed it when I compared it to my trustee 5DM2 and used a light meter. Should I send it to Canon to be recalibrated? I don't think off setting it all the time will help specially on weddings. Any advice or suggestions?

It seems to be the way the metering algorithms are set on the 5DIII, others have mentioned it too.

I have the 5DIII myself though and I haven't noticed an issue, however I use the metering to get a starting point and correct from that as appropriate for the scene anyway, always have with any camera body. I then use either manual or Av / Tv with EC according to what I need. Sometimes I set +ve EC and sometimes -ve EC, it all depends on the scene and brightness of subject and background.

I suspect that I am using +ve EC more than I would on other bodies, but it really doesn't matter to me. I don't find any problem whatsoever in setting the correct exposure for the shot I am taking.

I don't think that sending it in will help, as I suspect it is "in spec" and this is just the way the camera is set up. Different models of camera do meter differently according to how they are setup to read the scene.

Its curious how we seem to be seeing a number of basic metering questions cropping up recently. Often trying to blame the camera for basic operator errors.

I remember being told that in the sailing magazine world, every three years the readership would recycle back to beginner level. So never ever throw the original articles out, just wait three years and republish!

I can see parallels here.

There are several threads here about an underexposure issue with the 5D3.

https://photography-on-the.net …/showthread.php​?t=1164735

Mine underexposes when there is a lot of white in the scene (beach, snow, clouds, etc.). The camera meters for middle gray, so an underexposure under those conditions is expected. I just adjust it myself.

Are you checking your histogram? And what metering mode are you using? Remember, Evaluative metering is biased towards the focus points you are using, while the other modes are not.

There's a reported light-leak around the top LCD, which may be affecting the metering. Here's just one source. But it's been discussed on the forum previously.
http://www.dpreview.co​m …nowledges-5dii-light-leak

That being said, I've alway had to go +1/3 to 2/3 on my 40D to get what I considered correct exposure. Now the same on my 5DIII.

See if the metering changes when you take it out on a bright day, cover and then uncover the top LCD.

LV Moose wrote in post #15778783
There's a reported light-leak around the top LCD, which may be affecting the metering. Here's just one source. But it's been discussed on the forum previously.
http://www.dpreview.co​m …nowledges-5dii-light-leak

That being said, I've alway had to go +1/3 to 2/3 on my 40D to get what I considered correct exposure. Now the same on my 5DIII.

See if the metering changes when you take it out on a bright day, cover and then uncover the top LCD.

Oh, FFS, don't lets start all that baloney up again.

The tiny amount of light getting past the top LCD only ever affected metering when the camera was in the dark, or had the lens cap on, and somebody shone a light at the LCD. The only way they could possibly see a metering change by covering / uncovering the LCD on a sunny day, is to keep the lens cap on and cover up the viewfinder. As soon as they take the lens cap OFF (you know, as you do when taking a picture for example) the vast amount of light flooding in through the lens will make the odd photon getting in at the top completely undetectable.

You will get far more light coming in through the viewfinder, past your eye, than will ever get past that LCD. Several other cameras also have the light leak past the top LCD (some other Canon models, Nikons etc) but nobody ever said anything because nobody happened to try taking shots with the lens cap on, whilst shining a torch at the camera.

There is absolutely no way that this could be causing the OPs camera to meter wrongly in normal shooting.

^ okie dokie then

So this is what I did in order to know if this is an operator error. I placed the 5dm3 in a tripod, shoot one subject in AV mode, then did it in manual with the same shutter, aperture, and iso setting. What the 5Dm3 on AV mode did was added 2 2/3 exposure on its setting. Thoughts anyone?

You're using EC in AV mode?

Anyway...a meter is a guide it is not a "put it in the middle" for a perfect exposure. It never has been. You must understand how metering works, and sometimes how the particular meter in your particular camera sees a scene and then adjusting to the scene to get the exposure that you want.

CoPhotoGuy wrote in post #15798887
You're using EC in AV mode?

Anyway...a meter is a guide it is not a "put it in the middle" for a perfect exposure. It never has been. You must understand how metering works, and sometimes how the particular meter in your particular camera sees a scene and then adjusting to the scene to get the exposure that you want.

I didn't replicate anything but the shutter,aperture and iso. In the histogram it shows that the 5dm3 compensated by 2 2/3 of a stop.

Yes sir I understand that about the meter. Here is an easier description I believe, the metering in my camera is 1/400 when my light meter is at 1/100 with both aperture at 2.8 and iso at 100. Isn't it a little bit off?

litratista wrote in post #15798846
So this is what I did in order to know if this is an operator error. I placed the 5dm3 in a tripod, shoot one subject in AV mode, then did it in manual with the same shutter, aperture, and iso setting. What the 5Dm3 on AV mode did was added 2 2/3 exposure on its setting. Thoughts anyone?

If you take a shot with the same aperture, shutter speed, and ISO, you get the same exposure whether the camera's onboard computer entered those values or you did it yourself. Your statement is thus confusing and leads me to believe you're doing something other than you say.

litratista wrote in post #15798946
I didn't replicate anything but the shutter,aperture and iso. In the histogram it shows that the 5dm3 compensated by 2 2/3 of a stop.

What does that last part mean, exactly? The histogram reflects the exposure taken, not how it was arrived at.

litratista wrote in post #15798946
Yes sir I understand that about the meter. Here is an easier description I believe, the metering in my camera is 1/400 when my light meter is at 1/100 with both aperture at 2.8 and iso at 100. Isn't it a little bit off?

A hand-held light meter measures the light hitting the subject. (incident metering) The camera's meter measures the light bouncing off the subject. (reflective metering) The proper exposure can be 1/100 but if your subject is a nearly 100% pure white wall, the camera will use a much faster shutter speed since it thinks that white wall should be middle grey. From 1/100 to 1/400 is two stops: Pretty much the difference between middle grey and a white wall.

Comparing your camera meter to a lightmeter isn't necessarily going to give you the same reading (or even close). As Mike says, an incident reading will often be very different to a reflected reading. Even assuming you used your lightmeter in reflective mode, there are still big differences possible due to angle of view. If your meter is set to a fairly wide angle, and your camera is reading from a fairly small portion of the scene, they are seeing two very different things. In addition, the camera then adds its own algorithms to give its idea of the "correct" exposure, a lightmeter won't do that, it leaves that to the photographer.

It is quite likely that the 5DIII algorithms do aim to underexpose to a degree, many DSLRs do, in order to minimise blown highlights. Again, a lightmeterwon't do that for you. Different cameras use different algorithms, so two different models of camera won't necessarily give the same suggested exposure, and they will both probably differ from a lightmeter most of the time.

I am often shooting with a little +ve EC on my 5DIII (or the meter to the right of centre, if in manual) but yesterday, I was in bright sunlight shooting racing cars and I was in -ve EC all day long (up to a stop and a third at times) to keep the whites from blowing out. So, this camera (which is "underexposing" byt your definition) was shooting over a stop MORE underexposed to get the correct exposure at times. I was generally shooting the white / light cars for a couple of laps, then shifting the exposure to shoot the darker cars. If I shot the darker cars at over a stop -ve EC, then yes, they were underexposed, I shot those at -1/3 to 0 EC get a good exposure, but at that exposure, the white cars blew out badly.

As has been said above, the meter is a guide to help you assess the correct exposure. The camera doesn't underexpose, the photographer does. Sometimes you need higher than the suggested meter reading, sometimes lower. The camera makes a suggestion, that is all, and probably errs on the side of underexposure deliberately to avoid blowing highlights. It is for you, the photographer, to interpret that suggestion and obtain the correct exposure which may be a couple of stops away from the meter reading, in either direction.

I don't understand the difference between Av and manual of over two stops. If as you say the shutter speed, aperture and ISO were all the same in both shots, then the exposure would be the same, so I don't see how that could happen. Metering wouldn't affect that, EC wouldn't affect it either (as you say you took the shot in Av (so the EC would have been applied) then used the same settings to shoot in manual.

I don't even understand what you were attempting to show wioth that test either. If your metering is "wrong" then it would affect the Av shot, yes, but then using the identical settings in manual ignores the meter completely, so should be identical.

I will post pictures on what I did and show you guys what Im talking about. I want it to be operator error alone so I can fix it with more knowledge. Let's hope Im not explaining it wrong also, I think my photography lingo is still in development. Thank you for all the replies!

The first Canon I ever bought had metering problems. After struggling between sending iit back and just making manual adjustments I finally sent it back to Canon. They kept it forever, but when I got it back it was much better. Still not on the money, but not way off as it had been before. Out of the experience I acquired a nice hand held meter that I still have and use from time to time today.

I'm interested in seeing your pics. I recently shot a gray card and the histogram shows an underexposure when metered to 0. When I go up in the plus 1 range then I get a middle spike in my histogram.