Dealing With Bright Midday Sun

[jprzybyla]

Looking at the image and considering the time of day it was captured my guess would be that the scene was beyond the dynamic range of the sensor/camera. In those instances you have a choice... expose for the sky (as the camera has done in the posted image) or blow out the sky and expose for the people. If exposing for the people in the scene add one or two stops of exposure metering off the sky.

[Mike Buckley]

There is no clipping to either side of the histogram. This is an indication that the exposure was indeed correct.

This histogram of the first image in the thread indicates otherwise about the clipping. It doesn't indicate anything about the exposure of the captured image, as the histogram is of the post-processed image.

[Brev00]

So far members have discussed solutions in terms of exposure which has been educational. I have been giving this same subject some thought having just returned from Gettysburg, Pa on a very sunny day in an open, preserved battlefield. Capturing a lot of blue sky can work in an image but it is important to get the blue right. I am typically less worried about shadows which can usually be lifted without too much trouble in post. So, I spot meter the brightest section of the sky and add 2 ev. This will lighten the blues but decrease the noise so when I bring down the exposure in post I can lift the shadows without much noise penalty. That is my exposure tip. Otherwise, I found more of the solutions involved composition. Essentially, I try to raise the horizon and thereby minimize the amount of sky in a shot. In this shot, you could have framed more water and much less sky giving you a more even-toned scene to capture. You can crop out some of the water later. Try more closeups, too. You can control the amount of bright sun in a longer focal length shot by moving about your subjects and changing the angle of the lens. As others have noted, your exposure here was pretty good even though your settings could be improved. Sometimes, if you get the shot, it can just be a matter of having the right software to really reach your vision.

[William W]

Andrew,

Look at the histogram of your original image. There is no clipping to either side of the histogram. This is an indication that the exposure was indeed correct. You will see a long blue spike in the histogram that is an indication that the Auto WB was not “correct”.. . . etc.

Totally irrelevant to the discussion about the exposure of the original scene.


The histogram of the image posted, is the histogram of a post produced image.

***

. . .If for any reason you cannot trust the TTL metering of your camera you might as well dump it and get a Nikon. . .

You quoted my words to get a direct response . . .

Andre – please re-read the original post. Andrew used spot metering.

The conversation which I am having with the OP is about the fact this particular the image is underexposed.

All the indications of underexposure are present in:
> the image;
> the EXIF;
> the scene itself.

It is not all that difficult to understand the most likely reason why the shot is underexposed and Dan K articulated that precise reason, in his post #11.

***

As you want to involve the aspect of "trusting" the TTL metering system . . .

Well of course one CANNOT "trust the metering system".

The TTL metering system is dumb - and also blind.

The TTL meter especially when in SPOT METERING MODE has absolutely no idea at what it is pointed.

So one CANNOT just blindly follow any SPOT METER reading to give the "correct" exposure for any scene.

One has to interpret the SPOT METER reading (or readings) and then modify the exposure as and when necessary. This is just a standard procedure that most knowledgeable and experienced photographers follow when using a Spot Meter, be it either a TTL Spot Meter or an Hand Held Spot Meter.

WW

[William W]

Therefore my suggestion would be that anyone with a modern camera that has an interesting and potentially useful feature, should investigate it thoroughly - how it works, whether it will work for you, and if the product can be edited relatively easily to give the image result that you visualise.

Cheers.
Philip

Sage.

And to be crystal clear, Philip, my previous was NOT arguing against your suggestion for the OP to try the HRD function in his camera: I was merely saying that it might not work, and providing a reason - i.e. because of the (many different) Subject Movements in the scene.

WW

[William W]

Looking at the image and considering the time of day it was captured my guess would be that the scene was beyond the dynamic range of the sensor/camera.

Extremely unlikely that the scene is beyond the DR of the camera/sensor, but yes, beyond the DR range of being able to reproduce the image here. The two are different and often confused.

FYI and for more details REF this conversation:

SOOC - Disappointing Exposure Issues - Please Help

***

In those instances you have a choice... expose for the sky (as the camera has done in the posted image) or blow out the sky and expose for the people. If exposing for the people in the scene add one or two stops of exposure metering off the sky.

No. We don't know what the TTL meter exposed for -

> firstly the posted image is post produced (so the sky could have been enhanced / manipulated)
> secondly Andrew has not disclosed what he Spot Metered upon.

*

I agree, there must be exposure choices made in regard to the vision of the final image.


Those exposure choices must also be made with due consideration to the Post Production Techniques that will be undertaken on the file to produce the final image.

WW

[William W]

Andrew -

I added this link to another reply to explain about the camera's DR.

You might find much of the information contained in that link useful as it touches on many of the (similar) aspects that are relevant to shooting in midday sun.

Some of the issues touched upon are, but not limited to:

> The DR of Modern DSLR Cameras
> Making exposure choices with a clear vision of the Post Production that will be applied
> The uses and values of HDR
> Capture and Post Production Techniquies for wide DR Scenes

This is the link to that previous conversation:SOOC - Disappointing Exposure Issues - Please Help

WW

[AB26]

This histogram of the first image in the thread indicates otherwise about the clipping. It doesn't indicate anything about the exposure of the captured image, as the histogram is of the post-processed image.

Mike,

You mean you assume it is a post-processed image.

I doubt Andrew would post an image that has been post-processed and ask for the assistance he has asked for. Would it not render the image useless with regard to assistance if it has been post-processed?

The slight touch of the histogram to the left will not render the dark areas in the image unrecoverable. It is so slight that it does not really matter. I believe the edit I have done pretty much proves my point.

How much detail can be recovered in black material with so much white in and image where the white was not “blown”?
The choice in this image is to either blow out the white clouds and get detail in the shadows or get detail in the clouds and don’t worry about the little lost detail in shadow areas.

Personally I think Andrew has done a pretty darn good job of getting the exposure right to the extent where the image is very much usable to solve Andrew’s unfounded “concern” with shots like this.

There is no single simple solution to shooting in harsh sunlight with shadow areas and get all of it perfectly exposed, without adding additional lighting.

[MrB]

...to be crystal clear, Philip, my previous was NOT arguing against your suggestion for the OP to try the HRD function in his camera: I was merely saying that it might not work, and providing a reason - i.e. because of the (many different) Subject Movements in the scene...

Thank you for your clarification, Bill - I did realise that, and I should have acknowledged in my next post (#14) that your point about movement was certainly worth making and very relevant regarding the OP's scenes.

As I stated in Post #18, my views there are intended to be general comments. To clarify my point, there are some photographers around - I have met some and come across many on the Web - who convey an attitude of such disdain for JPEGs, for anyone who shoots them, and therefore for any camera process that creates them, that they would never demean themselves by trying an in-camera HDR feature that results in a JPEG. In my view, by having such an attitude, they remain ignorant of the capabilities of modern technology, and they are missing out on potentially useful features.

Cheers.
Philip

[William W]

Philip,

Thank you for your detailed response.

There is no need to clarify your posts on my account. All your posts in this thread have been quite clear to me.

I think that what can sometimes happen is that two respondents, in the endeavour to assist the OP, can get caught in a tangential conversation: and it is worth while articulating that, that conversation is not an argument: I think that is precisely what we both have done.

Regards,

Bill

***

Post Script:

Ever since I cut over to Digital from Film, I capture 'raw + JPEG (L)' and I sometimes use the JPEG file and also I set Picture Style to suit the Shooting scenario as the JPEG has an influence on both the Histogram and also the Highlight Blinkies.

I use Canon DSLR and Fuji Film Mirrorless.

WW

[William W]

Mike, You mean you assume it is a post-processed image.

Andre,

That statement is a nonsense.

The OP posted the image on a website - it MUST be post processed.

And Andrew might respond and tell us exactly to what extent, that it was post processed . . .

please?

WW

[Mike Buckley]

Andre,

You mentioned that the histogram doesn't display clipping so I simply pointed out for the sake of the beginning photographers who might assume that to be true that in fact it isn't. I agree with you that the dark areas are recoverable but that's a separate issue from your clear-cut, incorrect statement about the histogram.

You're correct that I assume Andrew's image posted in the thread has been post-processed. My reasoning is that at the very least it appears reasonably sharp despite that it has probably been downsized and probably been converted to a JPEG for display on the Internet. (I seem to remember that Andrew has stated in the past that he always shoots RAW files.) Again for the beginners following along, sharpening an image does affect the histogram if only a little.

Let's assume though that the histogram of the image that is straight out of the camera resembles the histogram of the image posted in the thread. If I or my wife had reviewed that histogram immediately after capturing the image, we would have retaken it using more exposure in an attempt to get an ideal exposure. Clearly, this histogram indicates to me that though it is a good exposure, a better one can be achieved.

You have a history here of explaining the importance of trying so diligently to get everything "right" in the camera. It is especially in that context that I am surprised to see you stressing that the histogram, if it is of the image straight out of the camera, indicates that that had been achieved when in my opinion the histogram indicates that that had not been achieved.

'Nuff said!

[AB26]

[B][U]The TTL metering system is dumb - and also blind.

The TTL meter especially when in SPOT METERING MODE has absolutely no idea at what it is pointed.

So one CANNOT just blindly follow any SPOT METER reading to give the "correct" exposure for any scene.

One has to interpret the SPOT METER reading (or readings) and then modify the exposure as and when necessary. This is just a standard procedure that most knowledgeable and experienced photographers follow when using a Spot Meter, be it either a TTL Spot Meter or an Hand Held Spot Meter.

WW

Well I DO follow Spot Metering blindly, and I trust it as well! I am not a knowledgeable experienced Photographer, only an amateur trying his best. I have to trust the in camera electronic systems or else I would be completely lost.
I also trust the ABS and EBD systems on my car as much as I trust the steering and suspension.

One can only "interpret" the spot meter system and “modify exposure” if one knows how to interpret the histogram.
(I do not know how to "interpret" a metering system. I only know how to use it.) I think the metering system interprets data, or something like that?




This is a typical example of Andrew’s “dilemma”. I also use Spot Metering often.
This image was shot in RAW, cropped a tad converted to Jpeg and resized to post. Looking at the EXIF you will see I used SPOT METERING. EC = 0. You will notice that blue spike, in the histogram, as it was shot in Auto WB against a blue sky. In post I will most likely bring down the blue channel a little.

How far is this shot UNDEREXPOSED? You think the metering system had any idea where I was aiming?
Should I trust the SPOT METERING system on my 9 year old Nikon camera?

[Mike Buckley]

The OP posted the image on a website - it MUST be post processed.

My Nikon D7000 produces JPEGs that can indeed be displayed here with absolutely no post-processing. I don't know about Andrew's camera or how it was configured, though as mentioned in my previous post, I think he always shoots RAW files.

[William W]

My Nikon D7000 produces JPEGs that can indeed be displayed here with absolutely no post-processing. I don't know about Andrew's camera or how it was configuredm though as mentioned in my previous post, I think he always shoots RAW files.

Mike, Andrew shoots raw. The image to be a JPEG, is at the least post processed in camera: that's the point I was making.

The next step is for Andrew to disclose is any further processing of the image file.

WW

[William W]

Well I DO follow Spot Metering blindly, and I trust it as well! I am not a knowledgeable experienced Photographer, only an amateur trying his best.

OK.

We all make choices.

And that's your choice to use Spot Metering and your choice also to trust that the TTL Meter in Spot Metering Mode gives you the correct exposure all the time.

WW

[William W]

. . .One can only "interpret" the spot meter system and “modify exposure” if one knows how to interpret the histogram.

That is nonsense.

Spot Metering was used long before Digital Cameras & Histograms were invented.

WW

[DanK]

To clarify my point, there are some photographers around - I have met some and come across many on the Web - who convey an attitude of such disdain for JPEGs, for anyone who shoots them, and therefore for any camera process that creates them, that they would never demean themselves by trying an in-camera HDR feature that results in a JPEG. In my view, by having such an attitude, they remain ignorant of the capabilities of modern technology, and they are missing out on potentially useful features.

Philip,

I half agree with you. I think many of the 'you must do this' posts are nonsense. People should do what works for them. However, I think you are inadvertently verging on that yourself, insisting that people should explore all the options their cameras offer. For some people, that is good advice. For others, not. My newer body, a Canon 5D3, has a manual that is over 400 pages long. It can do all manner of things that I have no interest in learning how to do. The fact that Canon engineers found ways to load this stuff into my camera doesn't mean that it is good use of my time to learn how to use it. I'd much rather put my time into mastering the aspects of the system that will help me in shooting the way I want to shoot, e.g., mastering its extremely powerful but complicated AF system. While others may find it useful to shoot jpegs (more power to them!), I don't, and I don't want to spend my time learning how to shoot jpegs better.

Dan

[William W]

Philip,

I half agree with you. . . etc

Balance to the conversation.

Bravo.

WW

[DanK]

Well I DO follow Spot Metering blindly, and I trust it as well! ...I have to trust the in camera electronic systems or else I would be completely lost.
I also trust the ABS and EBD systems on my car as much as I trust the steering and suspension.

One can only "interpret" the spot meter system and “modify exposure” if one knows how to interpret the histogram.

Andre,

Keep in mind that people productively used spot metering for decades before there were any histograms. I started using it when I bought a Canon FTb in 1971 or 72. To use it well, you can't use it blindly. It's more like steering than a suspension.

Assuming you wanted the fur properly exposed, I would expect spot metering to underexpose unless you were certain that the entire spot metering area was in shadow. A quick work-around: hold the palm of your hand so that it is lit like the area you want neutrally exposed--in this case, shaded like the fur--spot meter off it, and open one stop. That will usually get you pretty close.

People were using reflective metering systems long before there was anything automatic in cameras. In fact, the early cameras with built-in meters often had only a single metering mode. Mine had only spot metering. If I recall correctly, the competing Nikon had only center-weighted averaging. (This was one reason I preferred Canon.) Film was expensive enough that you often didn't take multiple exposures, and you got to see if you screwed up only later, when you developed the film. The solution was not to trust the metering, but rather to understand its strengths and weaknesses and use it accordingly. I can't speak for him, but I think that is part of what WW was suggesting.

There are a lot of people who insist that the way to learn photography is to limit your use of modern technology. E.g., there are lots of people who say you learn faster using only primes. In general, I disagree. However, metering is one area where I do agree. I think the best way to learn metering is to take advantage of the fact that digital shots are free. Use manual mode, and experiment with different metering modes to see what works. Once you know what exposure you want, you can more easily figure out when you can make use of the camera's automatic features to make your life easier. Just my two cents.

Dan