Strategies to prevent M11 overexposure

[J S H]

Advertisement (gone after registration)

1 minute ago, lct said:

I will go on nailing exposure when at risk of digital noise and underexposing by default otherwise but i will try @jaapv's and @hdmesa's methods if mine does not work any more, thank you for your efforts guys 🙂

Actually, it may be that you didn't read his advice carefully enough: Do some exposure bracketing going higher in exposure by 1/3 stops from where you normally shoot.  Post-process all of them to taste, then pick the best one. It will not be the one you thought it was. And you have learned a new trick – how to expose properly.

He wasn't saying you needed to change your approach, or that your method suddenly won't work any longer. Rather, he suggested that if you conduct a simple experiment, you might discover that there is actually a better technique. Respectfully, in my experience, he is correct and the ideal exposure is probably +2/3 to +1 stop above your average exposure, based on your described approach. We have all been taking pictures for years, but it's never too late to learn!

It may only matter for one of 100 images, but there is a point where you will see a net benefit from proper exposure and processing will actually be easier. 

[Advertisement]

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

[lct]

1 minute ago, J S H said:

Actually, it may be that you didn't read his advice carefully enough: Do some exposure bracketing going higher in exposure by 1/3 stops from where you normally shoot.  Post-process all of them to taste, then pick the best one. It will not be the one you thought it was. And you have learned a new trick – how to expose properly.

My method works well since i got my first digital camera in 2002 or thereabout, but i will certainly try @jaapv's and @hdmesa's ones if mine does not work any more due to new softwares or for any other reason 😎

[hdmesa]

8 minutes ago, J S H said:

Actually, it may be that you didn't read his advice carefully enough: Do some exposure bracketing going higher in exposure by 1/3 stops from where you normally shoot.  Post-process all of them to taste, then pick the best one. It will not be the one you thought it was. And you have learned a new trick – how to expose properly.

He wasn't saying you needed to change your approach, or that your method suddenly won't work any longer. Rather, he suggested that if you conduct a simple experiment, you might discover that there is actually a better technique. Respectfully, in my experience, he is correct and the ideal exposure is probably +2/3 to +1 stop above your average exposure, based on your described approach. We have all been taking pictures for years, but it's never too late to learn!

It may only matter for one of 100 images, but there is a point where you will see a net benefit from proper exposure and processing will actually be easier. 

I try to mention this where I can because a few years back, Jim Kasson on DPReview’s medium format forum helped me in the same way. I had @lct’s approach to exposure, but Jim challenged me to take some of my RAW files I thought were well exposed and look at the histograms in RawDigger. I was surprised to find my method was underexposing by one to two full stops over what the RAW file could handle. Granted cameras like the M11 and GFX series handle underexposure very well, but I still found that taking an extra shot or two above where I was setting my exposure led to overall images with better tonality. I could see the results right there in front of my face in Capture One. The image that had more exposure than what I had been doing just looked better after editing it. It looked more “alive” for lack of a better word. Doesn’t mean my previous images were subpar, only that they could be even better if I chose to learn to meter for the RAW files instead of using just the EVF exposure preview.

[hdmesa]

7 minutes ago, lct said:

My method works well since i got my first digital camera in 2002 or thereabout, but i will certainly try @jaapv's and @hdmesa's ones if mine does not work any more due to new softwares or for any other reason 😎

I will light a candle for you that one day you might see the light. “Nailing exposure” is not synonymous with “underexposing”, and it never will be.

[lct]

1 minute ago, hdmesa said:

I will light a candle for you that one day you might see the light. “Nailing exposure” is not synonymous with “underexposing”, and it never will be.

My bad English again? My method is nailing exposure when at risk of digital noise, otherwise i underexpose by default and adjust in PP. Not trying to convince anybody here, i was just replying to a colleague asking how we deal with exposure in digital photography. Just an example below. 6400 iso with my M240 and no noise reduction in PP. No underexposure of course knowing that 6400 iso is the limit of this camera.

https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M240-ZM-502/i-HCLdBLC/0/a803c344/X4/M2401879_sips-X4.jpg

[hdmesa]

42 minutes ago, lct said:

My bad English again? My method is nailing exposure when at risk of digital noise, otherwise i underexpose by default and adjust in PP. Not trying to convince anybody here, i was just replying to a colleague asking how we deal with exposure in digital photography. Just an example below. 6400 iso with my M240 and no noise reduction in PP. No underexposure of course knowing that 6400 iso is the limit of this camera.

https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M240-ZM-502/i-HCLdBLC/0/a803c344/X4/M2401879_sips-X4.jpg

I see. I read it as you were saying you nailed exposure by underexposing. Probably not your English. It's that English speakers sometimes read nuance into things that aren't intended by the writer if the writer's first language isn't English. 

I do still recommend checking a few of your favorite DNG files in RawDigger – at least once – to see where the exposures are landing on the histogram. It's a different histogram than the one in camera or the one in Capture One or Lightroom. In camera is only looking at the dynamic range of the JPEGs, and the histogram in C1/LR is expanding the data because there has been a profile applied. Maybe you're already close to where you need to be, I have no way of knowing. But I can tell you from having checked my RAW files in RawDigger, if you don't push your highlights in the EVF right to the edge, you're probably underexposing the image.

If you don't want to bother with Raw Digger, you could post a link to a few DNGs you took, and I can post back some screenshots of the histogram in RawDigger for you.

Edited December 3, 2023 by hdmesa

[lct]

Advertisement (gone after registration)

I do use RawDigger but not for histograms. As i tried to explain above, i don't like navigating by instruments. My method is very simple and works fine for me. Either i am at risk of digital noise and i nail exposure, like in the M240 snap bove . Or i am not exposed at that risk and i underexpose by default before adjusting in PP like in the pic below (M11 + Rokkor 28/2.8). Just an example among thousands in 20+ years.

https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M11-Rokkor-2828/i-7HbWJsp/0/041847d8/X4/M1004048_sips-X4.jpg

[hdmesa]

30 minutes ago, lct said:

I do use RawDigger but not for histograms. As i tried to explain above, i don't like navigating by instruments. My method is very simple and works fine for me. Either i am at risk of digital noise and i nail exposure, like in the M240 snap bove . Or i am not exposed at that risk and i underexpose by default before adjusting in PP like in the pic below (M11 + Rokkor 28/2.8). Just an example among thousands in 20+ years.

https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M11-Rokkor-2828/i-7HbWJsp/0/041847d8/X4/M1004048_sips-X4.jpg

Looking at the histogram in RawDigger isn't navigating by instruments, it's just taking a bit of time to learn how your in-camera meter functions versus the available highlight room in the DNG as shown in RawDigger.

Here's an example of how following the in-camera meter and one's "gut" can lead to incorrect exposure. Here is the edited image (M11 + 28 Cron II) exported from C1 showing what looks to me like great detail in the shadows:

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!

 

But if I open RawDigger and turn on exposure warnings, it shows underexposed areas in blue:

 

Now I look at the image in the RawDigger histogram (from the "Window" menu, choose "Histogram"), which shows the true RAW histogram. I can see the exposure falls short of the available highlight headroom on the right by several stops (look at the EV stops at the top of chart toward the right to see by how much). This image is severely underexposed. Had I taken another one with at least two stops more exposure, I would have ended up with a final image that had much better tonality? I'll never know as I didn't bracket this shot. Was the image I took good enough? Probably. But don't I want to do better if I can? I do!

 

Edited December 3, 2023 by hdmesa

[lct]

I respect you much @hdmesa and i'm sure you are convinced that your method is the best for you but you will see me watching histograms when pigs can fly 😄 or at least if my method happens to stop working for me. Now i am always glad to exchange experiences especially with people thinking or feeling differently to me 😎

[hdmesa]

18 minutes ago, lct said:

I respect you much @hdmesa and i'm sure you are convinced that your method is the best for you but you will see me watching histograms when pigs can fly 😄 or at least if my method happens to stop working for me. Now i am always glad to exchange experiences especially with people thinking or feeling differently to me 😎

It's not a "method". It's learning: how much does my meter typically over or underexpose a scene by? That's all. Once you figure that out, you don't need to watch anything – you just always know for typical scene the meter is going to be one stop too dark. That's it.

[lct]

I'll gladly let you have the last word, then 🙂

[hdmesa]

4 minutes ago, lct said:

I'll gladly let you have the last word, then 🙂

But did you, though? 😂 If you put only a portion of this effort into looking at two or three RAW files in RawDigger...

🫣

[lct]

2 minutes ago, hdmesa said:

But did you, though? 😂 If you put only a portion of this effort into looking at two or three RAW files in RawDigger...

If i answer you, you won't have the last word any more 😉 so i prefer letting you discuss with other colleagues more receptive to histograms than i am 😎

[LanceR]

@hdmesa you mentioned that you cannot achieve the same histogram in CaptureOne that you can in RawDigger.  I have no experience with RawDigger, but my question is if you apply the Linear Response curve profile for the camera in CaptureOne and you have not applied any edits, does that not result in a similar histogram to RawDigger?

[hdmesa]

10 hours ago, LanceR said:

@hdmesa you mentioned that you cannot achieve the same histogram in CaptureOne that you can in RawDigger.  I have no experience with RawDigger, but my question is if you apply the Linear Response curve profile for the camera in CaptureOne and you have not applied any edits, does that not result in a similar histogram to RawDigger?

Capture One has an RGB + Luminance channel histogram while RawDigger has the actual RAW histogram made up of the RGGB channels from the sensor. I don't know if there are times when that will make a large difference, but I do know that the histogram in RawDigger is much easier to read since it has +/-EV step markings on the charts. You can get an approximation of how much you're underexposing in Capture One by using the Linear profile and then seeing how many stops you have to move the exposure slider to get the optimal histogram and image that doesn't appear to blow the highlights, but there's a bit of a "shotgun" approach compared to looking at the histogram in RawDigger. RawDigger also allows you to look at the histogram in Logarithmic terms, which is usually shows the exposure is a bit higher, and I find this truer to what I see in post when editing. In Log mode, I can also sometimes see tiny parts of the image that are spaced a bit to the right of the end of a given channel. This can be important if you want to capture all the highlights such as a sky shot with white clouds – or sometimes it needs to be ignored in cases with specular highlights that need to be pure white.

Here's an underexposed image (taken using the M11 meter/EVF as a guide) showing the RGB(+L) histogram in Capture One with the Linear profile selected. The top half shows the image as-is, the second below is me increasing the exposure slide by 2 stops.

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!

 

Here is the same image in RawDigger using both Linear and Log. I added the white line to show the exposure difference between them. I also put a red square around the "rouge" highlights that in this case I would want to save. If the shot had included specular highlights (chrome bumper with sun reflection, etc.), I would ignore those stray marks.

 

So bottom line, if you don't want to buy RawDigger, you can approximate the stops of underexposure in C1 with the Linear profile.

[SrMi]

On 11/30/2023 at 8:28 AM, setuporg said:

So compared to M10R, M11P consistently overexposes, to my liking.  Maybe I was underexposing all along...  So using EV compensation -1/3 or -2/3 is a tried and true way with Leica.  What is it with them?  Why do we have to do it year on year on all models?

What other approaches have you seen work to achieve less blown up exposure overall?  

Did you use the same exposure metering methods? M10R's multi-field (live view only) should give the same metering as M11's default metering (multi-field), but I have not tested it.

[SrMi]

On 11/30/2023 at 10:19 AM, setuporg said:

You’d think the whole point of AI metering is it could do that for you, but no…

M11 has three different metering modes, none using AI, AFAIK.

[SrMi]

On 11/30/2023 at 10:24 AM, jdlaing said:

I think it better to overexpose than underexpose as a default. As you said “to my liking”. 

Why? Clipped highlights often cannot be recovered, but noisy shadows can be denoised with AI NR.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[SrMi]

7 hours ago, hdmesa said:

Capture One has an RGB + Luminance channel histogram while RawDigger has the actual RAW histogram made up of the RGGB channels from the sensor. I don't know if there are times when that will make a large difference, but I do know that the histogram in RawDigger is much easier to read since it has +/-EV step markings on the charts. You can get an approximation of how much you're underexposing in Capture One by using the Linear profile and then seeing how many stops you have to move the exposure slider to get the optimal histogram and image that doesn't appear to blow the highlights, but there's a bit of a "shotgun" approach compared to looking at the histogram in RawDigger. RawDigger also allows you to look at the histogram in Logarithmic terms, which is usually shows the exposure is a bit higher, and I find this truer to what I see in post when editing. In Log mode, I can also sometimes see tiny parts of the image that are spaced a bit to the right of the end of a given channel. This can be important if you want to capture all the highlights such as a sky shot with white clouds – or sometimes it needs to be ignored in cases with specular highlights that need to be pure white.

Here's an underexposed image (taken using the M11 meter/EVF as a guide) showing the RGB(+L) histogram in Capture One with the Linear profile selected. The top half shows the image as-is, the second below is me increasing the exposure slide by 2 stops.

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!

 

Here is the same image in RawDigger using both Linear and Log. I added the white line to show the exposure difference between them. I also put a red square around the "rouge" highlights that in this case I would want to save. If the shot had included specular highlights (chrome bumper with sun reflection, etc.), I would ignore those stray marks.

 

So bottom line, if you don't want to buy RawDigger, you can approximate the stops of underexposure in C1 with the Linear profile.

There is no difference in exposure between Log and Linear in RawDigger. The difference is that with Log, you can see the data which is too small to be visible with Linear (if you stretch the histogram vertically, you will see a hint of data in linear mode). You can check the max values for R G B G2. They do not change whether you select Log or Linear histogram (sometimes those max values are stuck at max).

[LanceR]

On 12/5/2023 at 6:02 PM, hdmesa said:

Capture One has an RGB + Luminance channel histogram while RawDigger has the actual RAW histogram made up of the RGGB channels from the sensor. I don't know if there are times when that will make a large difference, but I do know that the histogram in RawDigger is much easier to read since it has +/-EV step markings on the charts. You can get an approximation of how much you're underexposing in Capture One by using the Linear profile and then seeing how many stops you have to move the exposure slider to get the optimal histogram and image that doesn't appear to blow the highlights, but there's a bit of a "shotgun" approach compared to looking at the histogram in RawDigger. RawDigger also allows you to look at the histogram in Logarithmic terms, which is usually shows the exposure is a bit higher, and I find this truer to what I see in post when editing. In Log mode, I can also sometimes see tiny parts of the image that are spaced a bit to the right of the end of a given channel. This can be important if you want to capture all the highlights such as a sky shot with white clouds – or sometimes it needs to be ignored in cases with specular highlights that need to be pure white.

Here's an underexposed image (taken using the M11 meter/EVF as a guide) showing the RGB(+L) histogram in Capture One with the Linear profile selected. The top half shows the image as-is, the second below is me increasing the exposure slide by 2 stops.

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!

 

Here is the same image in RawDigger using both Linear and Log. I added the white line to show the exposure difference between them. I also put a red square around the "rouge" highlights that in this case I would want to save. If the shot had included specular highlights (chrome bumper with sun reflection, etc.), I would ignore those stray marks.

 

So bottom line, if you don't want to buy RawDigger, you can approximate the stops of underexposure in C1 with the Linear profile.

Thank you @hdmesa for the thorough response.  Much appreciated.