Different processing RAW files for H or V shot

[Stephen D]

Hi Guys

As a stock photographer I often take the same shot in horizontal or vertical mode.

Sometimes shooting on a tripod, with a prime lens and in manual exposure mode I have to process the two shots differently in order to get them to match.

Any ideas why?

Steve

[wiskerke]

If there's no mechanical fault, something in your workflow must be on automatic.

Have you tried putting both H and V on 2 layers and aligning them to see if it's not an optical illusion?

Do the images look the same when viewed on the camera display? Are they still the same when viewed on the display of your computer?

Do you use a raw converter that allows you to apply the exact same settings as the previous conversion? (Like ACR.)

 

wim

 

edit: In 2J38WAM and 2J38WA6 it's light fall off (or vignetting) of the lens, that influences a different part of your image. Plus a different camera angle left/right.

Try taking 2 images using only the exact square center in H and V and rotated around the exact center point.

If you suspect mechanical failure of some sorts, try the same camera positions in upside down position also.

Edited April 8, 2022 by wiskerke

[Allan Bell]

You say "In manual mode" but when you shoot one image in horizontal the camera generally sees a different overall brightness than when you shoot in vertical (Portrait mode).

Could this be the problem?

 

Allan

 

[shearwater]

Maybe a polarizing filter? 

[MDM]

There is not enough info about how the OP is metering the scenes and what the actual differences are before processing. Saying manual (metering - same thing as manual exposure) alone is insufficient. Without knowing the metering mode  (spot, matrix etc) it is impossible to make an informed guess. If using the same exposures for both horizontal and vertical (as it would be with spot metering) then different white balances may be the issue. This is very common and not at all strange. If this is the case, the solution is simple - use the same white balance for raw processing. Even better use a grey card in the scene and use spot metering on the card.

Edited April 9, 2022 by MDM

[wiskerke]

8 hours ago, MDM said:

There is not enough info about how the OP is metering the scenes and what the actual differences are before processing. Saying manual (metering - same thing as manual exposure) alone is insufficient. Without knowing the metering mode  (spot, matrix etc) it is impossible to make an informed guess. If using the same exposures for both horizontal and vertical (as it would be with spot metering) then different white balances may be the issue. This is very common and not at all strange. If this is the case, the solution is simple - use the same white balance for raw processing. Even better use a grey card in the scene and use spot metering on the card.

 

I understand manual = the M setting on a camera = the exact same aperture and shutter speed for both H and V.

However with verticals the vignetting darkens the sky a lot more, which is in itself simple to see and measure, but it will still also fool the eye.

 

wim

[MDM]

17 hours ago, wiskerke said:

 

I understand manual = the M setting on a camera = the exact same aperture and shutter speed for both H and V.

 

 

I disagree with this. Depending on how you actually meter in manual mode, you could get two entirely different readings (light levels = aperture/shutter speed combinations) for horizontal and vertical framing. If using matrix or some other method that uses the entire frame for metering, the readings could be different for horizontal and vertical as there can be more sky in one (more likely in vertical) which will result in more light hitting the metering sensor). If one used spot metering to measure an approximate midtone in the scene then both exposures would be identical. Stephen has not explained what he means here.

 

17 hours ago, wiskerke said:

 

 

However with verticals the vignetting darkens the sky a lot more, which is in itself simple to see and measure, but it will still also fool the eye.

 

 

 

I guess vignetting could also be a factor which would act in the opposite direction to my scenario.

 

It would be great if Stephen could return and give more detail. It is not possible to tell anything looking at his processed images or what he is indeed asking.

 

Edited April 10, 2022 by MDM

[MDM]

If the person who posted the dummy face on my above post doesn't understand the simple technical point I am making and would like to tell me exactly what it is that they don't understand, then I will take the time to explain it to them if they wish. It might be futile to attempt to do so but I am willing to have a go. In any case, my reasoning there is entirely rational.

Edited April 10, 2022 by MDM

[wiskerke]

48 minutes ago, MDM said:

If the person who posted the dummy face on my above post doesn't understand the simple technical point I am making and would like to tell me exactly what it is that they don't understand, then I will take the time to explain it to them if they wish. It might be futile to attempt to do so but I am willing to have a go. In any case, my reasoning there is entirely rational.

 

No dummy face from me, but I do not understand your light metering method.

The idea is that one measures the subject or the light (=2 different metering methods). Then you choose a combination of shutter speed and aperture (and ISO), giving priority to one or more factors.

Now in the case of a manual setting on your camera, you transfer the found or computed settings to the camera and with that setting you shoot both horizontal and vertical image.

 

In your case you yourself would be the automated factor in the process, changing the setting on the camera between horizontal and vertical.

 

wim

[MDM]

1 hour ago, wiskerke said:

 

No dummy face from me, but I do not understand your light metering method.

The idea is that one measures the subject or the light (=2 different metering methods). Then you choose a combination of shutter speed and aperture (and ISO), giving priority to one or more factors.

Now in the case of a manual setting on your camera, you transfer the found or computed settings to the camera and with that setting you shoot both horizontal and vertical image.

 

In your case you yourself would be the automated factor in the process, changing the setting on the camera between horizontal and vertical.

 

wim

 

I didn't for a moment think it was you wim. I rate your intelligence very highly. Any confusion here between us is more a case of crossed wires than any real difference so let me have another go with some actual numbers. I am not talking about using an incident light meter, just a normal TTL meter with the camera set to manual mode. For simplicity let's say I am going to keep shutter speed and ISO constant at 1/125 and ISO100.

 

1. I meter a scene in manual matrix mode (whole frame) as it is called on Nikon cameras with some sky in horizontal position and l get a reading of f8 (1/125 and ISO100).

 

2. I turn the camera to vertical position where there is more sky in the scene and l get a reading of f11 (1/125 and ISO100) using the same method (manual matrix mode).

 

If I use those settings then the vertical image will be a stop darker than the horizontal one. I am not saying this is the correct thing to do - it's not but that may be the root of the problem here. In fact that is more or less what Allan was saying above. There may also be differences in white balance which will be evident if one examines the settings used by the camera.

 

3. I meter something in the scene that approximates to a midtone or a grey card in manual spot metering mode (avoiding sky completely) and I get a reading of say f5.6 (1/125 and ISO100). I use that reading for both horizontal and vertical shots which should then be identical in terms of brightness (your vignetting idea aside). This is effectively what you are saying but it only applies if you don't meter the whole scene as many people do who don't understand the concept of spot metering, metering the subject or using incident light meters.

 

However, if the camera is on auto white balance, there may well be differences between the white balance as shot even if the exposures are the same. I've experimented with this in the past and found it to be true.

 

The problem here is that Stephen has not told us how he is metering (apart from saying manual) and what the differences are that he is seeing in the post processing (differences in brightness and/or differences in white balance).

 

 

 

Edited April 10, 2022 by MDM

[Stephen D]

Hi Guys

Many thanks for your answers, especially wiskerke.

What I meant by manual was that I was in M mode so both H and V images had exactly the same aperture and shutter speed.

(There was no polarising filter).

However, having revisited the two images that I was particularly thinking about, I see that the camera was set to Auto White Balance.

I am pretty sure that this is the answer to my problem.

I have reset the camera to Daylight WB so hopefully I will not have the same problem again.

All the best

Steve

[spacecadet]

22 minutes ago, Stephen D said:

Hi Guys

Many thanks for your answers, especially wiskerke.

What I meant by manual was that I was in M mode so both H and V images had exactly the same aperture and shutter speed.

(There was no polarising filter).

However, having revisited the two images that I was particularly thinking about, I see that the camera was set to Auto White Balance.

I am pretty sure that this is the answer to my problem.

I have reset the camera to Daylight WB so hopefully I will not have the same problem again.

All the best

Steve

I'm sure you know this but in case everyone reading this doesn't, in RAW it's much easier to correct the WB in processing. I generally use AWB out side the studio and I find that images taken in the same light around the same time all tend to be within about 200K, such that I don't often find it necessary to correct them.

I suppose images with more or less skylight might vary a bit more, but I'm now looking at two images taken minutes apart of the same scene with a cloudy and a blue sky and they're within 50K- of course the exposures are different .

Wim and I would agree with you on what "manual mode" means.

[MDM]

1 hour ago, Stephen D said:

Hi Guys

Many thanks for your answers, especially wiskerke.

What I meant by manual was that I was in M mode so both H and V images had exactly the same aperture and shutter speed.

(There was no polarising filter).

However, having revisited the two images that I was particularly thinking about, I see that the camera was set to Auto White Balance.

I am pretty sure that this is the answer to my problem.

I have reset the camera to Daylight WB so hopefully I will not have the same problem again.

All the best

Steve

Different WB for horizontal and vertical was what I guessed in my first post. It doesn’t matter if you set WB in camera or indeed what WB you use when shooting raw but ideally photograph a grey card in the same light and use the eye dropper in post for accurate WB. 

[MDM]

50 minutes ago, spacecadet said:

 

Wim and I would agree with you on what "manual mode" means.

Don’t you shoot in programme mode? This is academic now but for clarity it may be but is not necessarily correct as I explained in detail above and what I said is correct. Of that I am absolutely certain. So once again you could get different readings metering in manual mode vertically and horizontally depending on how you meter as detailed above. 

[CAROL SAUNDERS]

I was out yesterday with the camera, I was shooting in A priority, took horizontal and vertical images of the same scene, no difference in settings or processing, can't detect any difference....I may have missed the point of course, wouldn't be the first time😁

 

Carol

[geogphotos]

3 hours ago, CAROL SAUNDERS said:

I was out yesterday with the camera, I was shooting in A priority, took horizontal and vertical images of the same scene, no difference in settings or processing, can't detect any difference....I may have missed the point of course, wouldn't be the first time😁

 

Carol

 

I process every shot as it comes so didn't really understand the OP's question. 

 

Like you I have never noticed any significant difference between H and V just by flipping the camera.  After all one is focusing on the same subject which most likely is in the middle of the frame.

 

 

Edited April 14, 2022 by geogphotos

[MDM]

13 hours ago, CAROL SAUNDERS said:

I was out yesterday with the camera, I was shooting in A priority, took horizontal and vertical images of the same scene, no difference in settings or processing, can't detect any difference....I may have missed the point of course, wouldn't be the first time😁

 

Carol

 

The main point that I am making is that when you meter a scene using matrix metering (you are a Nikon user I recall) or any other type of whole frame metering  such as the out-dated centre weighted metering rather than spot metering, the exposure readings you get may be different but are not necessarily different for horizontal and vertical orientations. The reason for this is that different amount of lights may reach the metering sensor if one orientation has more sky than the other for example (typically this will be the vertical orientation). This will apply when the camera is set to manual exposure mode as well as other exposure modes. In other words you can get different readings in manual mode for horizontal and vertical orientations but they do not necessarily have to be different - it depends on the actual scene as framed in the camera. 

 

The same applies to white balance readings if the camera is set to auto white balance which is the reason Stephen was getting different readings. White balance readings may not be as easy to see on the camera but are obvious in post.

 

To test this out, set your camera to manual exposure mode, turn off auto ISO, use matrix metering, turn on auto white balance and shoot raw. Frame an outdoor shot in a garden or a park say in horizontal orientation with the sky forming say 1/4 approx of the top of the frame. Take your readings and your shot. Then turn the camera into vertical orientation with the sky forming 1/2 the frame approx. Take your readings and your shot. Chances are the readings will be different as more light should be reaching the metering sensor. Take the images to your computer and check the white balance readings as well as the exposure readings. In LR, have WB set to As Shot. Chances are they will be different for the horizontal and vertical shots although they do not necessarily have to be different. It depends on the exact scenes as framed in the camera.

Edited April 15, 2022 by MDM

[M.Chapman]

51 minutes ago, MDM said:

The same applies to white balance readings if the camera is set to auto white balance which is the reason Stephen was getting different readings. White balance readings may not be as easy to see on the camera but are obvious in post.

 

That's the key point in this thread. Auto white balance (in camera or in post processing) can give different results when processing horizontal, vertical or square shots of the same scene, taken at the same time.

 

Mark 

[MDM]

4 minutes ago, M.Chapman said:

 

That's the key point in this thread. Auto white balance (in camera or in post processing) can give different results when processing horizontal, vertical or square shots of the same scene, taken at the same time.

 

Mark 

Exactly. The conversation about manual exposure was a diversion due mainly to lack of information about what differences the OP was seeing  in the first place

Edited April 15, 2022 by MDM

[CAROL SAUNDERS]

2 hours ago, MDM said:

 

The main point that I am making is that when you meter a scene using matrix metering (you are a Nikon user I recall) or any other type of whole frame metering  such as the out-dated centre weighted metering rather than spot metering, the exposure readings you get may be different but are not necessarily different for horizontal and vertical orientations. The reason for this is that different amount of lights may reach the metering sensor if one orientation has more sky than the other for example (typically this will be the vertical orientation). This will apply when the camera is set to manual exposure mode as well as other exposure modes. In other words you can get different readings in manual mode for horizontal and vertical orientations but they do not necessarily have to be different - it depends on the actual scene as framed in the camera. 

 

The same applies to white balance readings if the camera is set to auto white balance which is the reason Stephen was getting different readings. White balance readings may not be as easy to see on the camera but are obvious in post.

 

To test this out, set your camera to manual exposure mode, turn off auto ISO, use matrix metering, turn on auto white balance and shoot raw. Frame an outdoor shot in a garden or a park say in horizontal orientation with the sky forming say 1/4 approx of the top of the frame. Take your readings and your shot. Then turn the camera into vertical orientation with the sky forming 1/2 the frame approx. Take your readings and your shot. Chances are the readings will be different as more light should be reaching the metering sensor. Take the images to your computer and check the white balance readings as well as the exposure readings. In LR, have WB set to As Shot. Chances are they will be different for the horizontal and vertical shots although they do not necessarily have to be different. It depends on the exact scenes as framed in the camera.

Very interesting and apart from not using M mode I do all of the above, I rarely use auto ISO, use matrix and auto WB and raw.  Next time out will use M mode and your other suggestions, a good experiment to do.....have a few images awaiting keywording from the other day H & V so will attempt a reshoot using M mode.....looks like Stephen D is happy with all the info interesting topic yet again.....

 

Carol

 

 

[wiskerke]

5 hours ago, MDM said:

 

The main point that I am making is that when you meter a scene using matrix metering (you are a Nikon user I recall) or any other type of whole frame metering  such as the out-dated centre weighted metering rather than spot metering, the exposure readings you get may be different but are not necessarily different for horizontal and vertical orientations. The reason for this is that different amount of lights may reach the metering sensor if one orientation has more sky than the other for example (typically this will be the vertical orientation). This will apply when the camera is set to manual exposure mode as well as other exposure modes. In other words you can get different readings in manual mode for horizontal and vertical orientations but they do not necessarily have to be different - it depends on the actual scene as framed in the camera. 

 

The same applies to white balance readings if the camera is set to auto white balance which is the reason Stephen was getting different readings. White balance readings may not be as easy to see on the camera but are obvious in post.

 

To test this out, set your camera to manual exposure mode, turn off auto ISO, use matrix metering, turn on auto white balance and shoot raw. Frame an outdoor shot in a garden or a park say in horizontal orientation with the sky forming say 1/4 approx of the top of the frame. Take your readings and your shot. Then turn the camera into vertical orientation with the sky forming 1/2 the frame approx. Take your readings and your shot. Chances are the readings will be different as more light should be reaching the metering sensor. Take the images to your computer and check the white balance readings as well as the exposure readings. In LR, have WB set to As Shot. Chances are they will be different for the horizontal and vertical shots although they do not necessarily have to be different. It depends on the exact scenes as framed in the camera.

 

Metering is not the same as exposure.

 

wim

[MDM]

20 minutes ago, wiskerke said:

 

Metering is not the same as exposure.

 

wim

I agree. You meter and then expose according to the readings. Not sure what point you are making though. 

[Betty LaRue]

6 hours ago, MDM said:

 

The main point that I am making is that when you meter a scene using matrix metering (you are a Nikon user I recall) or any other type of whole frame metering  such as the out-dated centre weighted metering rather than spot metering, the exposure readings you get may be different but are not necessarily different for horizontal and vertical orientations. The reason for this is that different amount of lights may reach the metering sensor if one orientation has more sky than the other for example (typically this will be the vertical orientation). This will apply when the camera is set to manual exposure mode as well as other exposure modes. In other words you can get different readings in manual mode for horizontal and vertical orientations but they do not necessarily have to be different - it depends on the actual scene as framed in the camera. 

 

The same applies to white balance readings if the camera is set to auto white balance which is the reason Stephen was getting different readings. White balance readings may not be as easy to see on the camera but are obvious in post.

 

To test this out, set your camera to manual exposure mode, turn off auto ISO, use matrix metering, turn on auto white balance and shoot raw. Frame an outdoor shot in a garden or a park say in horizontal orientation with the sky forming say 1/4 approx of the top of the frame. Take your readings and your shot. Then turn the camera into vertical orientation with the sky forming 1/2 the frame approx. Take your readings and your shot. Chances are the readings will be different as more light should be reaching the metering sensor. Take the images to your computer and check the white balance readings as well as the exposure readings. In LR, have WB set to As Shot. Chances are they will be different for the horizontal and vertical shots although they do not necessarily have to be different. It depends on the exact scenes as framed in the camera.

I agree and have experienced it personally, re: more sky in vertical and have had to process differently. Not always, but often enough.

[wiskerke]

Just now, MDM said:

I agree. You meter and then expose according to the readings. Not sure what point you are making though. 

 

You are taking a reading between changing from horizontal to vertical. Which means that if you get a different outcome, you are changing the settings on your camera, or are letting your camera change the settings. So yes then you do get a different exposure.

Taking one reading, setting your camera manually, and not changing the setting between horizontal and vertical would mean you get the complete same result on film and in RAW.

(Oh and not only metering and exposure are two different things, also light metering and object metering are.)

 

Now the OP transferring this result to Photoshop or other Raw developer finds he has to use different settings for the horizontal and vertical image to get the exact same result. That is very counterintuitive and not really possible without any automation going on along the way. So the first suspect would be a setting on the camera. WB does nothing to change RAW. Auto ISO however does: it's auto, so there's a calculation going on. But if the result ISO value is the same, the outcome should be the same. A quick glance at the metadata will tell you all that has been going on.

Now suppose everything actually has been set to Manual and the ISO was fixed. Then what could have been the culprit?

First suspect: the Raw developer. With ACR that is not too difficult to see. In any case the settings are mostly visible. However the lens corrections are not and while you can turn them off or on in ACR, the result may be the same when the corrections are built-in: the Raw then is actually cooked. No idea if this is the case with Nikon lenses and their profiles. But it is a route that camera makers are more and more taking and it's difficult to find out.

Now suppose all this is not the case and both Raws are in fact identical and the Raw developer is treating them the exact same way.

Then what could be the culprit? Human perception? Would be my first suspect.

Especially density is very difficult to judge by eye.

We all remember the checker shadow illusion I think:

(wiki)

Which btw also lies at the heart of why HDR works.

I checked an image pair that struck me as a good example:

 

Aligning them In Photoshop the resulting stack suggested that the lens had some vignetting. It may have been added in post, but I doubt it. The vignetting being present means no baked-in profile btw.

Also the angle was different on both. The camera had not just been rotated on the lens axis, but both tilted a bit and swung to the left. That in itself would have been enough to fool the eye. But the vignetting adds a lot more density in the corners and even a bit on the sides. Was it a wide angle lens? A wide angle zoom perhaps?

Anyway, my guess: the checker board shadow illusion again.

 

(Youtube)

(Remember #thedress? Oh no! wiki)

 

wim

[M.Chapman]

1 hour ago, wiskerke said:

So the first suspect would be a setting on the camera. WB does nothing to change RAW.

 

I'm going to be pedantic here. Whilst I agree that the RAW image (pixel) data isn't affected by the camera's WB setting, the WB setting in use (and the camera's computed "Auto WB" of the scene) are recorded in the RAW file's meta-data. This is of no consequence, unless the RAW developer software is set to use the "as shot" WB, in which case the values stored in the RAW file's meta-data will alter the developed result.

 

Mark

 

 

Edited April 15, 2022 by M.Chapman