Exposure Diagnostic help please

[pnodrog]

Andre,

Correct, I mixed up my pluses and minuses but the procedure is the same. Here's an example, one image taken at 0.0EV and the other taken with a -4.7EV dialed in, aimed at a darker area, exposure locked, re-composed and hey presto an image with very confusing metadata.

'P' mode, 1/500. f11, ISO400, 0.0EV (result as would be expected)


'P' mode, 1/500, f11. ISO400, -4.7EV set and indicated.


The very slight difference in exposure is due to changing light conditions due to the clouds or framing being slightly different.

I also managed it no problem indoors at these settings so it does demonstrate that this is something that can happen either inadvertently or deliberately. It possibly also ties in with comments from Catalina that she has been moving the camera around all over the place to get focus as AF is not working?

Grahame

Just did the same test on a D800 and the second shot at EC -4.7 was very dark and way under exposed. Also the first was f11 at 1/400 and the second went to f22 at 1/2000. Your exposure stayed the same?

P.S. Just noted you aimed at a darker area that's cheating.....

[Stagecoach]

Paul, you have the procedure wrong.

The first image with the camera set to P and 0.0EV, framed, and shot gives a good exposure as we would expect. It just happened the camera chose 1/500 @ f11.

For the second shot, the camera still in P mode had a -4.7EC entered. I moved the camera and pointed it around a darker area watching the speed/aperture figures move and locked exposure when it showed 1/500 f11. Then moved again to frame the area I had in the first shot and of course we know that 1/500 f11 is what the original needs.

Because of the -4.7EC entered whilst it is metering the darker area it is going to underexpose giving a higher speed wider aperture.

Grahame

[Stagecoach]

P.S. Just noted you aimed at a darker area that's cheating.....

It's not cheating it's demonstrating how unexplained things could happen quite easily

[pnodrog]

It's not cheating it's demonstrating how unexplained things could happen quite easily

Well by not cheating I know my camera does exactly what I and everyone else would expect.....

You by cheating also know your camera does exactly what you and everyone else would expect.....

[AB26]

presto an image with very confusing metadata.

Thank you Grahame,

Eureka – you found it: a way of “cheating” the Metadata.
Interesting, very interesting! I will try it as soon as we have sunshine again (raining at the moment).

Mystery solved, and something learned.

With regard to my experiment: the result was achieved by using the thumb wheel, yes. No BS and no cheating, in “P” mode, no other mode. Must be the way I have set up my camera in the Menu. The D300 is more advanced and you should be able to do the same thing. (Swop the wheels and see what happens.)

On my camera the Aperture and Shutter speed will do as you described until the aperture reaches max, thereafter the shutter speed will change until the correct exposure can be attained.

These are the two images in my experiment. See the Metadata.
#1 was 0 EC and #2 was shot with -3.0 EC dialled in using ± button. The -3.0 EC was countered by using the thumb wheel only. In the viewfinder I could see the image will be under exposed so I simply spun the thumb wheel to the left until the exposure meter indicated the correct exposure.

#1


#2

[AB26]

It's not cheating it's demonstrating how unexplained things could happen quite easily

Damn, I was under the impression the object of the thread was to find a way of "cheating" the Exif recorded.

[pnodrog]

Well at least it brings up the possibility of an amazing way of metering and then composing a photograph. With matrix metering it takes a fair bit of "cheating" to get the shift in exposure demonstrated. Hard to see how it would happen in normal operation. Easier to fool if spot metering.

I assume the lens cap is removed before metering.......... no I am kidding.

[pnodrog]

No chance they have tried to use a grey card held up close to the camera against the strong lighting from the sky? It would be in shadow and give a totally incorrect exposure.

[Stagecoach]

Damn, I was under the impression the object of the thread was to find a way of "cheating" the Exif recorded.

Andre,

Note I did not use such terms as Ureka because whilst I have shown what 'could' have happened having now got all of Catalinas image in ViewNX I also found that I was able to sort these by 'created date', something I have never bothered about before because my own images are sorted by file No.

What this has now shown is that out of all the images there are three consecutively by date order taken over a period of a week that each show them taken in 'P' mode and indicating the -4.7EV

Grahame

[Stagecoach]

Well at least it brings up the possibility of an amazing way of metering and then composing a photograph. With matrix metering it takes a fair bit of "cheating" to get the shift in exposure demonstrated. Hard to see how it would happen in normal operation. Easier to fool if spot metering.

I assume the lens cap is removed before metering.......... no I am kidding.

Paul,

It's actually extremely easy which surprised me and I have undertaken it in a sunny and cloudy scene and indoors twice.

Whilst I demonstrated it with 'exactly' the same exposure 1/500 @ f11 exposures somewhere near that would have also produced results within the novice tolerances we are looking at.

Grahame

[Stagecoach]

Andre,

With regard to my experiment: the result was achieved by using the thumb wheel, yes. No BS and no cheating, in “P” mode, no other mode. Must be the way I have set up my camera in the Menu. The D300 is more advanced and you should be able to do the same thing. (Swop the wheels and see what happens.)

On my camera the Aperture and Shutter speed will do as you described until the aperture reaches max, thereafter the shutter speed will change until the correct exposure can be attained.

Ok then !

On my camera this is certainly not the case, just confirmed it and no way of reconfiguring anything. In P adjusting the command dial puts the Aperture and Speed (changing balance) to their max settings in either direction and they lock there.

Now, If your D200 once getting to a limit of either ap or speed in the 'flexible' P mode THEN starts reducing one or the other while you continue spinning the command dial the exposure WILL change and if you can get the meter centred you will get a good exposure.

Now what if this is the case for the D40 and Catalina is operating things this way which is entirely wrong?

We have had questions like should I buy a light meter because I just can't get these reading right.

Hmmmmmmmmmmmmmmmm

EDIT: Forget that theory because we would be seeing limits eg very high or low settings of either aperture or speed that we are not.

And looking at the Exif of both of your images in Opanda and View NX they both show 0.0EV compensation?

Arghhhhhhhhhhhhhhhhhhhhhh

Grahame

[pnodrog]

Paul,

It's actually extremely easy which surprised me and I have undertaken it in a sunny and cloudy scene and indoors twice.

Whilst I demonstrated it with 'exactly' the same exposure 1/500 @ f11 exposures somewhere near that would have also produced results within the novice tolerances we are looking at.

Grahame

In the COW photograph that triggered this discussion apart from the sky the lighting is very flat and possibly a little under exposed. There does not seem to be a sufficiently large or dark enough area to fool the metering to the extent we are observing. With spot metering a chance but it looks like a scene where it would be hard work to achieve your results.

Even many of the most basic PP software have an auto brightness/contrast adjustment feature. I suspect a correction has been applied. The scene lighting suggests due to the bright area in the centre of the sky any EC should have been on the + side not -4.7. If we get to critique a new one perhaps we need to ask for a straight out of camera version before we comment

[Stagecoach]

In the COW photograph that triggered this discussion apart from the sky the lighting is very flat and possibly a little under exposed. There does not seem to be a sufficiently large or dark enough area to fool the metering to the extent we are observing. With spot metering a chance but it looks like a scene where it would be hard work to achieve your results.

Even many of the most basic PP software have an auto brightness/contrast adjustment feature. I suspect a correction has been applied. The scene lighting suggests due to the bright area in the centre of the sky any EC should have been on the + side not -4.7. If we get to critique a new one perhaps we need to ask for a straight out of camera version before we comment

Paul,

Pointing the camera at the ground there and locking the exposure I suspect would give you a good chance of having one that is going to counter a fairly large EC with that framed shot.

As for PP, this image blown up to 400% shows no appreciable noise which you would expect if PPd heavily. Andre PPd a similar shot I did at IS0400 -4.7EV and noise was extremely noticeable. I think it's pretty safe to say this image was taken at a good exposure 1/500, f11, ISO400 that would not have needed any significant exposure PP.

Grahame

[AB26]

What this has now shown is that out of all the images there are three consecutively by date order taken over a period of a week that each show them taken in 'P' mode and indicating the -4.7EV

Grahame

OOPS! -Eureka. Back to the drawing board.

[pnodrog]

Paul,

Pointing the camera at the ground there and locking the exposure I suspect would give you a good chance of having one that is going to counter a fairly large EC with that framed shot.

As for PP, this image blown up to 400% shows no appreciable noise which you would expect if PPd heavily. Andre PPd a similar shot I did at IS0400 -4.7EV and noise was extremely noticeable. I think it's pretty safe to say this image was taken at a good exposure 1/500, f11, ISO400 that would not have needed any significant exposure PP.

Grahame

I remain highly sceptical. The focal length was 18mm giving a reasonably broad evaluation area and measurements I have quickly done indicate that unless it was zoomed or put close to one of the dark patches on the nearby ground it would be unlikely to have corrected by much more than +2. Bearing mind the scene excluding the sky is about a stop underexposed I get to a about 3 stops of explanation unless some deliberate or misguided technique has been employed.

All this does not matter - it just confirms the conclusion that the camera is being used incorrectly. How is still a bit of a puzzle.

[AB26]

I remain highly sceptical.

L.Paul,

I can get the same result with the D200. It is not as easy for me (in Matrix) as Grahame says but it is attainable.

I must add that the test I did was in poor light (it is still overcast and raining here) in bright light it will be easier to get the result ( provided the DR is wide ).

I suggest you dial in -4.0 EC, hold the camera like you would when you review an image (point it towards your body), press the AE lock button (“by accident”) point and shoot. See the result.

[Stagecoach]

Andre,

Just to let you know I PMd Catalina and she has replied to me.

Firstly, she has a new lens so can now use AF
Secondly, she can not understand the 'two button reset' meaning but she has confirmed she has a camera manual.

I am going to assist with the two button reset as I believe this is a must in this situation. The next move I believe would be to take an image in P mode, no EC, of a scene and post it and we can evaluate if the exposure is ok for the lighting condition and mode and look at the Exif carefully for any anomaly. I'll produce instruction and reference to manual page No's.

I think in this case 'P' is the mode to start with although generally I would suggest Aperture priority.

Grahame

[AB26]

Firstly, she has a new lens so can now use AF

Secondly, she can not understand the 'two button reset' meaning but she has confirmed she has a camera manual.

Hi Grahame,

Great, good news.

Grahame, I would suggest you give her this link: http://www.dpreview.com/reviews/nikond40 and ask her to read every word in the review. These reviews are sometimes better, than the manual, to understand all the functions your camera offers. It will help her better understand her camera and understand the limitations of the camera.

Catalina will be in good hands, with you, guiding her in her endeavour.

I wish to thank you for this thread as I have definitely learned something new.