What is the best exposure for your Olympus? (ETTR and other techniques)

Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
I have been working a few months on an article to explain the inner workings of the camera and what we can do to reduce noise. There are many good articles and videos out there but I could not find much written specific for Olympus cameras. My aim was to list real techniques the Olympus photographer can test, apply and use.

While in the process writing the article, it looks like I discovered something really interesting, something that might be of interest to Olympus photographers. I have a little more work to do before my article is done but I thought to show you a teaser and a link to a video you should go see. This video will set the stage to understand why and how my proposed settings will work with Olympus cameras....

The proposed settings and thinking is nothing complex....it's always like that, while you researching and testing it seems like mount Everest you crossing, when you on the other side and you find the answer it looks so simple.... :) It could also be I am late to the party and nothing is new, I trust you will forgive me if the case....

This technique has different results with different cameras, for example the Pen F surprised me a lot. When you take an image in your living room in the evening using say aperture mode and auto ISO, most of the times the detail on people's hair is smeared or with no details. Last night testing I managed to get a clear detailed image that was never possible in the past.....

I am hoping to have the article done by Monday next week.

Please see the video here (new link)

Siegfried
https://myolympusomd.blogspot.com
https://www.instagram.com/siegfried_seierlein/


Here are two series of the same 3 test images:-

ETTR-Plus.jpg
Subscribe to see EXIF info for this image (if available)



ETTR-Plus2.jpg
Subscribe to see EXIF info for this image (if available)
 
Last edited:

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,159
Well, this was interesting. :2thumbs:

Watching it, I wonder if this is what we are doing by using a 3-stop ND filter to make our weak speedlight-type of flash units to provide enough fill-light in the bright sunlight if we cannot go above our flash exposure of 1/250 sec?
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,159
This has been bugging me. Had some clouds appear as the guy responded with this comment on the Youtube video.

"The test would be to go out and shoot a picture of some clouds on a bright day. Have the camera Auto expose the first image at ISO 100. Then throw the camera into manual mode and decrease the aperture by four stops and shoot the scene again at ISO 1600. Then look at the highlight details in the cloud."

So the claim is what appears to be the same 4-stop exposure setting results in a different cloud highlight with more detail even though the ISO is HIGHER.

My results are below on the E-M1X shot at ISO 200 (12-100mm f/4 at 16mm) and then ISO 3200 which is 4 stops. I knew the ISO 3200 would show some noise so I used DxO 3 Prime to hep a bit there. RAW to JPEG below.

1230080_DxO.jpg
Subscribe to see EXIF info for this image (if available)

1230082_DxO.jpg
Subscribe to see EXIF info for this image (if available)


Is there more detail in the cloud highlights in the ISO 3200 one? Is it just he is seeing highlight noise and claiming more detail? Whites appear brighter. I dunno.

:confused-53:
 
Last edited:
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
This has been bugging me. Had some clouds appear as the guy responded with this comment on the Youtube video.

"The test would be to go out and shoot a picture of some clouds on a bright day. Have the camera Auto expose the first image at ISO 100. Then throw the camera into manual mode and decrease the aperture by four stops and shoot the scene again at ISO 1600. Then look at the highlight details in the cloud."

So the claim is what appears to be the same 4-stop exposure setting results in a different cloud highlight with more detail even though the ISO is HIGHER.

My results are below on the E-M1X shot at ISO 100 (12-100mm f/4 at 16mm) and then ISO 3200 which is 4 stops. I knew the ISO 3200 would show some noise so I used DxO 3 Prime to hep a bit there. RAW to JPEG below.

View attachment 798270
View attachment 798271

Is there more detail in the cloud highlights in the ISO 3200 one? Is it just he is seeing highlight noise and claiming more detail? Whites appear brighter. I dunno.

:confused-53:
Hello Mack
I think what you do is great. Doing a few tests to understand your camera and the best settings you will use is awesome. What was always difficult for me was to know where to start. This will now be easier. Just a quick comment - in your text you write ISO100 (5 stops) and below the image ISO200 (4 stops).
The upping and down of ISO John Hess talks about is difficult to test and apply. Two reasons, each group Olympus cameras using the same sensor, ISO response differ from the next group. Some use the same 16MP sensor for example, the Pen F does not use the same sensor as the EM1.2 and the EM1X or EM5 III. The second reason I guess is, at what ISO do I need to start to set my 50% gray point. (Optimum exposure) Will it be ISO200 or will it be ISO1000? Olympus does not tell us. From that point you will up and down the ISO scale, right? DXO gives us the so called sports (high - light ISO) value, the EM1.2 is 1312 - is this the 50% gray point? In other words set the ISO to 1200 or 2500 and then measure the exposure in Auto Mode. Try using a gray card to set the WB. Start your test from this point onwards.
As per DXO the EM1.2 / EM1X and EM5 III sensor is the best M43 sensor ever used. From the tests I did it surely is different from the other cameras I tested.
If you feel like doing a few more tests let me know pls.
Siegfried
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,159
Hello Mack
I think what you do is great. Doing a few tests to understand your camera and the best settings you will use is awesome. What was always difficult for me was to know where to start. This will now be easier. Just a quick comment - in your text you write ISO100 (5 stops) and below the image ISO200 (4 stops).
The upping and down of ISO John Hess talks about is difficult to test and apply. Two reasons, each group Olympus cameras using the same sensor, ISO response differ from the next group. Some use the same 16MP sensor for example, the Pen F does not use the same sensor as the EM1.2 and the EM1X or EM5 III. The second reason I guess is, at what ISO do I need to start to set my 50% gray point. (Optimum exposure) Will it be ISO200 or will it be ISO1000? Olympus does not tell us. From that point you will up and down the ISO scale, right? DXO gives us the so called sports (high - light ISO) value, the EM1.2 is 1312 - is this the 50% gray point? In other words set the ISO to 1200 or 2500 and then measure the exposure in Auto Mode. Try using a gray card to set the WB. Start your test from this point onwards.
As per DXO the EM1.2 / EM1X and EM5 III sensor is the best M43 sensor ever used. From the tests I did it surely is different from the other cameras I tested.
If you feel like doing a few more tests let me know pls.
Siegfried
Opps! :dash2:

I see where I entered the ISO 100 as the test part for me. I just edited it to ISO 200 which is what I did since that is the recommended Olympus ISO setting. The info at the bottom of the images is auto-entered from the EXIF data so it is correct (I used a free FrameShop script plug-in in Photoshop CS6 does that and the drop borders and framing as well as the EXIF data from the image.). So it was a four-stop range per John's recommendation.

Still, the noise at ISO 3200 was apparent even in the white clouds. I don't know if trying it on the E-M1X at the lowest ISO (ISO 64 or what they call L64 in the menu.) and trying it four stops up at ISO 1000 would demonstrate the protected highlight detail better along with less noise.

I was also wondering if John's idea works better with full-frame cameras with a higher ISO before the noise floor enters since he started with ISO 100 and ended with ISO 1600? Maybe this will not apply to the current E-M5 III, E-M1 II, and E-M1X?

Staring at the two cloud images today with different eyes, maybe there is more contrast in the first one shot at ISO 200. Perhaps the flatter look to the ISO 3200 is what he is driving at by saying "the higher ISO protects the highlights?"

More testing I guess. Jury still in deliberation here.
 
Last edited:

exakta

Mu-43 Top Veteran
Joined
Jun 2, 2015
Messages
616
I wonder if this is what we are doing by using a 3-stop ND filter to make our weak speedlight-type of flash units to provide enough fill-light in the bright sunlight if we cannot go above our flash exposure of 1/250 sec?
Using ND filters for fill flash is not about controlling dynamic range. It's about allowing a wider aperture when the shutter speed cannot be increased further. There are two usage cases:

1. The lens aperture cannot be closed down enough for proper exposure of the ambient light.

2. The photographer wants to use a wider aperture to control DOF or diffraction.

Using ND filters is not overexposure, which is what ETTR is.
 
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
Opps! :dash2:

I see where I entered the ISO 100 as the test part for me. I just edited it to ISO 200 which is what I did since that is the recommended Olympus ISO setting. The info at the bottom of the images is auto-entered from the EXIF data so it is correct (I used a free FrameShop script plug-in in Photoshop CS6 does that and the drop borders and framing as well as the EXIF data from the image.). So it was a four-stop range per John's recommendation.

Still, the noise at ISO 3200 was apparent even in the white clouds. I don't know if trying it on the E-M1X at the lowest ISO (ISO 64 or what they call L64 in the menu.) and trying it four stops up at ISO 1000 would demonstrate the protected highlight detail better along with less noise.

I was also wondering if John's idea works better with full-frame cameras with a higher ISO before the noise floor enters since he started with ISO 100 and ended with ISO 1600? Maybe this will not apply to the current E-M5 III, E-M1 II, and E-M1X?

Staring at the two cloud images today with different eyes, maybe there is more contrast in the first one shot at ISO 200. Perhaps the flatter look to the ISO 3200 is what he is driving at by saying "the higher ISO protects the highlights?"

More testing I guess. Jury still in deliberation here.
Hello Mack

I think you doing the right thing, test, test, test is the only way to find the best answer for your own camera. That said I think you at the wrong end, this is why:-
Step 1 - The conditions is present to result in noise
Step 2 - We see noise
Step 3 - We try and clean noise (that's were you are?)

If you go back to step 1 the weak link is the sensor. Are you familiar with ETTR? (exposing to the right)

See this video

So in terms of priority, first optimize the sensor and then focus on ISO. You could fix the ISO at ISO200 and then find the max ETTR levels for your camera (shutter speed & aperture) for different scenarios (clouds, shadows or interior and more). Maybe you find +0.5 or +1 stop is enough and good for all cases.

Best
Siegfried
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,159
Hello Mack

I think you doing the right thing, test, test, test is the only way to find the best answer for your own camera. That said I think you at the wrong end, this is why:-
Step 1 - The conditions is present to result in noise
Step 2 - We see noise
Step 3 - We try and clean noise (that's were you are?)

If you go back to step 1 the weak link is the sensor. Are you familiar with ETTR? (exposing to the right)

See this video

So in terms of priority, first optimize the sensor and then focus on ISO. You could fix the ISO at ISO200 and then find the max ETTR levels for your camera (shutter speed & aperture) for different scenarios (clouds, shadows or interior and more). Maybe you find +0.5 or +1 stop is enough and good for all cases.

Best
Siegfried
Yes, I am aware of ETTR.

I often use a spotmeter, ether an old Pentax IV-F or a newer Sekonic L-858D-U since it does HSS readings for flash. I've been through all the Sekonic calibration with their Exposure II calibration card as well and know the Olympus curves it made from the Pen-F, E-M1 Mark II, and the E-M1X. Often I use a large white Styrofoam ball and use the camera's spot meter on it and set it to read +2.7 stops which just about tags the maximum in FastRawViewer or RAWDigger before the highlights are blown out. Doesn't always work given the lightness or darkness of a scene where it's at (e.g. A black bird in front of a dark wall.), but it's a start.

I think the video is focusing on video and not so much for still work. He goes into the video Log files a lot as well as the REC.709. Logs are really flat and less contrasty/colorful and require a lot of color grading in post. Those have a lot of latitude being flat. Could be - since it appears he is more video oriented - the claim of highlight protection by raising the ISO is more for video than still.

Imho, of course.
 
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
Yes, I am aware of ETTR.

I often use a spotmeter, ether an old Pentax IV-F or a newer Sekonic L-858D-U since it does HSS readings for flash. I've been through all the Sekonic calibration with their Exposure II calibration card as well and know the Olympus curves it made from the Pen-F, E-M1 Mark II, and the E-M1X. Often I use a large white Styrofoam ball and use the camera's spot meter on it and set it to read +2.7 stops which just about tags the maximum in FastRawViewer or RAWDigger before the highlights are blown out. Doesn't always work given the lightness or darkness of a scene where it's at (e.g. A black bird in front of a dark wall.), but it's a start.

I think the video is focusing on video and not so much for still work. He goes into the video Log files a lot as well as the REC.709. Logs are really flat and less contrasty/colorful and require a lot of color grading in post. Those have a lot of latitude being flat. Could be - since it appears he is more video oriented - the claim of highlight protection by raising the ISO is more for video than still.

Imho, of course.
Mack the goal should always be simplicity. That is the reason I introduced someone from the video community (John Hess) to talk about ISO. The general sensor size debate unfortunately confused the photographic community. Example:- The "exposure triangle" was labelled as something completely false and replaced by the SNR "base noise" theory. Perfectly OK but no connection to reality. Now that we know that camera ISO amplify SNR base noise, how does that help us? At least the exposure triangle tried to link ISO to the process setting up your camera for taking an image. The SNR theory left photographers generally afraid to mention the word ISO because we dare not say it's linked to gain? So what is the role of the camera ISO setting? If you have 10 photographers supporting the SNR base noise theory and you ask them.....what is the role of the camera ISO setting? How many will really know the answer? 1 or 2?

Again videographers simplified life - they not afraid to talk about ISO or noise and interestingly have a much better understanding and application know how....
 
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
So friends here the answer....

I read several comments of photographers saying they tried ETTR and it did not work. Have you tried ETTR? Do yourself a favor and type in ETTR in You-tube and watch a few videos. See how many does it completely wrong.... Did you see the mistake most of them are making?

Try this "exposure profile" in your Pen F
  • Menu D - Info Settings - LV Info - select C1 - select "Highlights & Shadows" and Histogram
  • Menu D - Histogram Settings - Highlights 253 and Shadow 2
  • Menu D - Art LV Mode - Mode 2
  • Menu E - Noise Reduct. - Auto
  • Menu E - Noise Filter - Low
  • Menu E - ISO - Auto
  • Menu E - ISO - Auto Set - Upper Limit/Default - High Limit = 800 - Low Limit = 200
  • Menu E - Lowest S/S Setting - 1/50 (1/60 is also OK)
  • Menu E - ISO-Auto - P/A/S
  • Menu E - Metering - Center Weighted Avg
  • Menu E - AEL Metering - Center Weighted Avg
  • Menu E - (...)Spot Metering - Mark "Spot"
  • Menu K - Exposure Shift - Select Center Weighted Avg and enter +3/6 (+6/6 also OK)
  • Select the creative lever to select the Curves function, press info, Mids = 4 clicks down
How to use:
  • This is NOT a one for all solution - you might have to adapt for your style
  • For general daylight, street, blue hour, landscape this profile is OK
  • This is a 80% "auto" solution - as always you need to think
  • Works well in Aperture Mode
  • Press info to see histogram - use histogram & blinkies to check exposure
  • These are the general goals for this profile:
    • Limit ISO role to up shutter speed - have IBIS take the load
    • Have the sensor "optimally" exposed at all times
    • Auto apply ETTR (Shutter/Aperture) when camera reach ISO800
    • Curves will reduce jpeg brightness to "normal" (RAW file w curves in WS)
    • In critical cases use SCP and select ISO200 for ETTR @ ISO200 up (Sky)
    • You will see a real improvement in IQ with low or no noise
For details on how I developed this profile see the article I am hoping to upload Monday evening the 27th.

- Other Olympus owners can adapt the profile to their cameras.....
- Its important to remember - only shutter speed and aperture expose the sensor - when applying ETTR its important to use shutter speed and aperture and NOT the camera ISO....
 
Last edited:
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
Thanks...read them.
- Keep in mind its only the analog signal that has SNR. From the ADC to the SD card we have a digital signal measuring bit error rate (BER).
- Those SNR components we like to keep an eye on is the sensor, analog gain and the digital to analog converter (ADC)
- We can influence the sensor (exposure) and the ISO (the sensor is the most critical one)
- Its easy to test the magnitude of photons - change the lens and check noise - teeeeny tiny impact
Study and try the exposure profile I gave in my other post - its simple, practical (no theories) and it works....
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,159
.....
Study and try the exposure profile I gave in my other post - its simple, practical (no theories) and it works....
Thanks. I'll try this and see how it works out.

I did have to set this up in A mode as M mode would not allow "Auto" (In menu E, I think?) to be selected as it was grayed out.
 

mfturner

Mu-43 Regular
Joined
Aug 6, 2019
Messages
106
I don't have a Pen F, and am trying to understand your recipe. Is your menu K setting plus the creative function (mid=4 clicks down) setting a way of slightly over exposing and bringing it back in post ("post" being the jpeg rendering in camera)? Essentially performing a bit of an ETTR? I sort of do this today on my m10.3, but only over exposing by 1/3rd stop and setting mids to 3 clicks down.

ETTR worked very well in my old Canon camera, to the extent I would use unity WB if lighting was dim and I cared about the outcome. I think the idea still works on my Olympus cameras, but I also think it's easier to blow out highlights on them, especially the PM1 with its ancient 12 mpx sensor. I haven't tried to create a unity WB setup on either Olympus camera yet, partly because it's only useful if you work with raw, and I'm content to simplify my flow right now and work with jpeg out of the camera.
 
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
Thanks. I'll try this and see how it works out.

I did have to set this up in A mode as M mode would not allow "Auto" (In menu E, I think?) to be selected as it was grayed out.
It's simple.... goto Menu E and select All
Now you can use M Mode. I do not really recommend this because I want an easy way out of the profile and that is Manual Mode where I will do everything manual....
 
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
I don't have a Pen F, and am trying to understand your recipe. Is your menu K setting plus the creative function (mid=4 clicks down) setting a way of slightly over exposing and bringing it back in post ("post" being the jpeg rendering in camera)? Essentially performing a bit of an ETTR? I sort of do this today on my m10.3, but only over exposing by 1/3rd stop and setting mids to 3 clicks down.

ETTR worked very well in my old Canon camera, to the extent I would use unity WB if lighting was dim and I cared about the outcome. I think the idea still works on my Olympus cameras, but I also think it's easier to blow out highlights on them, especially the PM1 with its ancient 12 mpx sensor. I haven't tried to create a unity WB setup on either Olympus camera yet, partly because it's only useful if you work with raw, and I'm content to simplify my flow right now and work with jpeg out of the camera.
You can apply this "profile" with any Olympus camera - some will benefit a lot from ETTR and other less. The EM1,2 sensor is very good and one see less of an improvement but its still there and enough to make a difference getting a really good image....
Cameras like the EM10 III and older needs a little change. These cameras does not have the "extended" ISO settings in Menu E - ISO Auto Set. All you do here is also set the max to 800 or 1000 and the low to 200.
I tried the Sport ISO setting on the DXOMark website as these are the max recommended ISO still giving great results for each camera....
Then the half stop - When you use your screen to show you the highlight and lowlight clipping then its easy to decide by how much you can over expose - plus one is seldom an issue... I have been going up +2,5 in some cases Another great feature Olympus has is config one of your camera buttons as "test image", set up the camera, take a test image and check if all OK. The test image is not saved.
Taking down the midpoint - Thats exactly what I do - in other words the jpeg is good - in fact you can use the camera as a post processing tool once the raw file is at max quality. Interesting the curves adjustments does show on the raw file when developed in WorkSpace.....
 
Joined
Mar 31, 2015
Messages
260
Location
Switzerland
Real Name
Siegfried
Folks my article is up....pls go see it here. I love to have your feedback.
I do believe the techniques and solutions I explain in this article will be a great value to MFT and Olympus owners.... You probably will have to study it a few times, but you will see once you understand the concept and master the application its not difficult.....
 
Last edited:
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Copyright © 2009-2019 Amin Forums, LLC
Top Bottom