5D III Exposure issues

Need a reality check please. Sorry to be so long to explain...

Long time Canon user. Had a Rebel, 40D, 1D MK III, 5D II, G9, G12. Many pictures I am proud of.

I bought a 5D III and sold all my other cameras. Took the standard test shots, read the manual cover to cover (seriously) and set it up so that I was very comfortable with the camera. Perhaps the hardest setup issue was a picture style that represented the jpg close enough to the raw so I had a decent indication of what I was shooting.

I took this (now only DSLR) on a Caribbean cruise with family. Only looked at pictures on the LCD and saw some blown highlights which is not unusual and usually not as bad as on the LCD and fixable in Lightroom, but all looked ok. A few blue skies that went white, but did not raise a huge flag. When I got home I saw images on my calibrated monitor that amazed me. For example, bright sunny day, cloudy but blue sky, green water like the following:

Content warning: NOT AN IMAGE

Content warning: NOT AN IMAGE

Content warning: NOT AN IMAGE

Content warning: NOT AN IMAGE

...this thread is sinking like a stone: so I'm bumping for you, and I have a question.

I'm not entirely clear what the issue is. From what I can tell what you see on the LCD doesn't match the image you view later on?

You've shown us a couple of pictures and gone: "What happened?"

The simple answer is you took some photos. I don't have a clue what you were expecting to see: what settings were your camera set too? Can you explain what the problems are with the photos?

rebop wrote in post #16751678
Then tried high contrast and the darks are too dark and the highlights too blown out. For example:

I was giving you the benefit of the doubt until you wrote this -unless you did not really mean it. You can't have both.

Personally I find DSLR's generally do worse in evaluative vs. compacts that I have had (both underexpose complex scenes and be less consistent).

I am not surprised about your results if you did not use EC. For the shots posted I would EC at least +1/3 - +2/3 with all my cameras in evaluative.

My 5DII seems to be worse than the 5DIII. I wish they would put the S95-S120 metering in their DSLRs (I guess they meter off the sensor though).

Either way you have to check the histogram and if the DR of the scene is too high, then it is too high - you need to either get a Nikon or Sony or install Magic Lantern and use DualISO. All those scenes have a high DR so you need to push the highlights as close to blowing out as possible. Unfortunately apart from the histogram, non of the leading camera manufacturers seem to think making this easy is a worthwhile endeavor.

EDIT: I downloaded the second shot to look at the luminance values and the highest was around 248, so not a lot of room to increase exposure. So it's high DR scene and you need one of the above or take the shot in better light or use HDR. The first shot could be brightened a little without cliping (maybe 1/3-1/2 stop), but the shadows would still be dark.

My standard EC setting for my 6D is +1/3. Quite frequently I use +2/3, and rarely +1. I am well over 1,000 shots on it and don't think I've ever had a properly exposed 0 EC shot. This is my fourth Canon digital camera. I would call it... normal.

Thanks for the replies. Funny thing is I have so many shots from my 1D III and 5D II with no EC and mainly evaluative that are so easy to tweak to get dialed in. The shots with this camera seem to be all over the board. And regardless of whether set to evaluative or not.

I really wish I knew why. If these shots demand EC, then many shots I have taken with the other bodies should have as well. And the flatness of those first shots amaze me. Just boosting exposure doesn't come close to matching how the scenes looked when shot.

But really appreciate the thoughts.

rebop wrote in post #16752276
Thanks for the replies. Funny thing is I have so many shots from my 1D III and 5D II with no EC and mainly evaluative that are so easy to tweak to get dialed in. The shots with this camera seem to be all over the board. And regardless of whether set to evaluative or not.

I really wish I knew why. If these shots demand EC, then many shots I have taken with the other bodies should have as well. And the flatness of those first shots amaze me. Just boosting exposure doesn't come close to matching how the scenes looked when shot.

But really appreciate the thoughts.

...sorry, still confused as to what you expected the images to look like. What did you expect the images to look like?

Part of me says the pictures you posted are essentially straight raw to JPG conversion without performing any other editing. In this case, you WILL get rather drab colors. Looking at the raw shots from my 5D3 in Lightroom has the same drabness to them. That's why Lightroom has various controls to do what the JPG processing would do in the camera.

Another part of me says you have made some settings that affect the JPGs output with a less than ideal setting. In particular are the 'picture style' settings which can be adjusted that they may 'flatten' the vibrant colors.

And lastly...watch the histogram. That tells you all you need if you're over/under exposing, etc. FWIW, I frequently look at the LCD to see if the exposure looks right and I check the histogram as well. I'm also happy to be within a stop over or under as I can generally correct that in Lightroom.

Oh, I just bought a G15 and am utterly blown away at the JPG results with full Auto! Obviously, Canon's JPG algorithms for JPG processing are more slanted towards the point and shoot non-photographer crowd and therefore give brighter everything.

IMO there are a number of issues mixed in this, that the OP tries to homgenize into one:

a) LCD shows different than PC:

LCD shows the Picture Style (PS) you have chosen, not the RAW. What PS were you using?

The RAW always needs some PP, otherwise it looks bland. Can you calibrate your monitor? What software are you using to PP the RAW. Are you post-processing the RAW or just work with the jpeg SOOC?

b) The Evaluative metering does not produce optimum results.

It appears that there are differences between cameras and between models. The popular wisdom is that the Canon Evaluative needs a 1/3-2/3 EC boost. I shoot my 60D with +2/3 EC. However, I found that my 5DIII works better with 0 EC. I hear other people adding a bit. So there is variation.

c) Your last photo (you called it 'high contrast', did you change any settings?). It's more a high DR judging from going from very dark shadows to very strong highlights. Again, is this RAW? If it is jpeg SOOC, can you give us a RAW of the same or similar scene?

EDIT: If you want to see what can be done with the photos you posted, turn on the 'Image Editing OK' (see under my avatar), you will find it in the User CP upper left of the page.

Just getting back to this thread today. I did spend a good amount of time in Lightroom last night and today with the vacation pics. Some could not be adjusted in Lightroom. Sliders ran out of room. that has NEVER happened to me before with nay other camera I have owned and they have also never looked this unsaturated as someone suggested this being normal. They are always needing adjustment. but not like this. I have to believe something is odd with either the 5D III or my 5D III.

Banquet bear - how can I describe what I expected the images to look like. A whole lot closer to a better exposure and with the dynamic range contained in the image. It is more than one problem going on. And for example, 90% of my 5D II images were decent straight out of the camera. Virtually none are with the III. If I hd never taken a good picture with other cmras, then of course its me. But that is just not the case.

MakisM1 - I am not trying to make this one issue other than trying to uncover if it is me, the 5D II or MY 5D III. I think I said I fully understand that the LCD is the embedded jpg. And I have tried Neutral and Faithful. Not much difference between the two on camera. And more difference on the 5D III in Lightroom between RAW and JPG than with previous cameras. The LCD looked acceptable when I shot and the RAW's were not close in Lightroom

I already mentioned I know that Canon claims evaluative as not ideal and mentioned I have tried center weighted and partial with little improvement. Also that 90% of my lifetime keepers were done with evaluative.

ALL my pictures were shot RAW. .What I posted were converted with Instant JPEG from RAW which is a very good representation. And sized.

Anyway, not quite going where I had hoped with the discussion. I find that no matter how carefully I think I choose my words, the ideas , questions and thoughts just do not seem to translate well on forums. After 10 years shooting Canon and just tossing away 85% of the trip pictures I think it is clear that whether it is me or the camera, I will never be happy with this camera. Wish I had kept the 5D II which delivered for me consistently.

I appreciate everyone's thoughts.

Sorry to hear your take. If you are looking for validation rather than solutions, then you can stop here. Sell the damn thing and by a 5DII (or a 6D...)..

As a last ditch effort, can you post in a Dropbox a RAW that is reasonably exposed but bland?

Also the 'High Contrast' one?

We are trying to help. We don't know how much you know, what your skill level is in LR (I for one don't use LR). All we have is your (descriptive) word about it.

The jpegs without any PP but with Exif intact may help too... (I find that the camera jpeg provides motre detailed exif than what I get from Darktable. Maybe it's just my setup).

I would recommend a reset, but you reloaded the Operating System... You seem not to like the result, but to me it looks ok. If I took a look at the RAW, I could say something more.

The first two shots, considering evaluative metering, 0 EC, and over 50% overcast/cloudy sky look better than I think they should. The third shot, beginning with a very high contrast scene, and adding to that more contrast, then using evaluative metering, and the center AF point, looks exactly as I would expect. EC would have been irrelevant here. Your chosen settings drove this shot from bad to worse. However you look at it, that is no fault of the camera.

I don't see it anywhere but you are aware that what you see in your LCD is not the raw image correct? It is always the jpg that you are creating with all your settings. Light room ignores all those settings and shows you its representation of the raw applying some default filters. DPP will show you the raw that will match the jpg you see on the LCD.

A high contrast setting only affects the jpg you see, and the raw in lr won't have those settings applied, so they will be really off from each other.

You posted a chart comparing CR2 and jpg RGB values. If you have RGB values that is a converted image, not a CR2. The numbers are meaningless unless you tell us what Raw converter was used, what settings it had (there is no such thing as "no processing", only different default processings) and in what color space the converted image was rendered.

Kind of vague, doncha' think? What profile? Which sliders? I'm confused; are you trying to lift underexposed shadows or to recover blown highlights? Or both, because the camera ran out of DR? If you want to know what really is in the Raw files, how the camera is performing, and whether you maxed out its DR, you can't do it in LR. By the time it hits your monitor it is no longer Raw. Post the files and I will post a Raw histogram. Or buy Raw Digger and do it yourself.

SqueekyBoy wrote in post #16754864
The first two shots, considering evaluative metering, 0 EC, and over 50% overcast/cloudy sky look better than I think they should. The third shot, beginning with a very high contrast scene, and adding to that more contrast, then using evaluative metering, and the center AF point, looks exactly as I would expect. EC would have been irrelevant here. Your chosen settings drove this shot from bad to worse. However you look at it, that is no fault of the camera.

Bright, sunny day.

Thanks all. Really appreciate you are trying to help. But the simple fact remains the results I get from this camera are so different than I have ever experienced with a Rebel, 40D, 1D MK III or 5D II that they do not work for me. Regardless of whether it is my lack of knowledge, a camera so different than any other I have owned I cannot adjust or a bad copy. Doesn't matter. If no one else has issues like this, I think that;s fabulous. Wish it were me. But got a bit of circles going on. For example, of course RGB means I converted the RAW. I did it in camera shooting RAW + JPG to compare the results with the grey card test. I was set on camera neutral. But that is not going to get me to a place with this camera that will make any difference. And nothing NOTHING like this ever happened with my 5D II which I now wish I had kept.

So really, thanks, but I'll see if I can get this thread closed.

~Bob

The OP appears to be asking what is so different from the 5DIII and all the other cameras he's owned in the past, especially his 5dII? What is causing the difference in how the cameras behave? Why is he having issues with the 5DIII that he was not experiencing with his 5DII? Is the 5DIII that "fickle"? I think I understand his frustration.