different exposure outcome with the same settings in the camera. pic after pic

R

rocketman122

Senior Member
I usually shoot on CH and shoot 2 pics in a row. this isnt the first time this has happened and its happened when my friend uses his D4, My D3 and my newly received (that was his) D3s.

didnt touch anything but place it in PS to resize.

Anyone experience this?
NKT_6735 copy.JPGNKT_6736 copy.JPG

weird how the difference is subtle online but on my computer, definetely a 1/2 stop difference in brightness between them
 
E

Eyelight

Senior Member
It is very subtle, but you can see it if you have room to see both images on the screen.

I know that was two shots within less than a second, but light travels at the speed of light, which means gad-zillions of photons hit the sensor during each exposure. So, I can imagine the light changing enough in less than a second to be noticeable.

if this is the case, it should be more pronounced at higher ISO and less at lower. Have you noticed either?
 
R

rocketman122

Senior Member
Eyelight said:
It is very subtle, but you can see it if you have room to see both images on the screen.

I know that was two shots within less than a second, but light travels at the speed of light, which means gad-zillions of photons hit the sensor during each exposure. So, I can imagine the light changing enough in less than a second to be noticeable.

if this is the case, it should be more pronounced at higher ISO and less at lower. Have you noticed either?
Click to expand...

I tell you what, because now I was editing an event, my eyes get tired after about an hour of editing, but when I start after a break, its very noticeable to me.

but something weird. when I was writing the above post I wrote "Anyone experience this?" and the picture was next to the text and the other under it. you could clearly see the difference. I then went in, pressed ENTER after the tex so it would be side by side, saved it and it wasnt so noticeable. I kid you not. and I worked in a printing lab so my eyes are very sensitive to 1/3 stops like crazy. very weird. on my screen in windows photo viewer, I press the back and forth button and its very noticeable. about half a stop.

I work very fast and precise and thats how my friend works as well. we hardly fire one pic at a time.
no havent noticed more at higher, less at lower. basically at lower iso.
this is very weird.
 
E

Englischdude

Senior Member
were you shooting in A mode? An subtle change in metering of the available light condition, shadows in the background etc. will influence the way the camera interprets the light available and adjusts accordingly. Safest would be to shoot in M with manual control over all settings. Were you using flash in tll mode? I always try to use flash in M mode also so the lighting setting remains constant.
 
R

rocketman122

Senior Member
no need for flash. he was in the shade. I think it was driving me crazy in my first few shots in A (not consistent) and I simply moved it to A. this was very normal. manual. one picture one exposure, and the 2nd a different. its happened before.
 
W

WayneF

Senior Member
rocketman122 said:
weird how the difference is subtle online but on my computer, definetely a 1/2 stop difference in brightness between them
Click to expand...

It is really subtle, and nowhere near 1/2 stop. If it were near 1/3 stop, you'd think settings would change to correct it.
If you don't have Raw, use the PS menu Image - Adjustments - Exposure to see what 1/2 stop looks like.

Maybe a passing cloud? They give me fits. :)
 
E

Eyelight

Senior Member
rocketman122 said:
no need for flash. he was in the shade. I think it was driving me crazy in my first few shots in A (not consistent) and I simply moved it to A. this was very normal. manual. one picture one exposure, and the 2nd a different. its happened before.
Click to expand...

So, if i read this right, A was not working so you used A. :)
 
R

rocketman122

Senior Member
whoops I meant M. duh.

Wayne- as I said, its not as drastic as it is the original file. very easy to notice by using the left right buttons and seeing it. yes, 1/2 stop. no it was rapid continuos shots so no time for a cloud to make a difference
 
Last edited:
E

Eyelight

Senior Member
Either pic without the other, something like this would never be noticed. It seems plausible that a very slight shift in light was detectable by the sensor that would not be noticed by the eye while shooting.

Shutter speed is relatively slow compared to light speed. Perhaps a small solar flare hit in between the exposures. :confused:
 
C

crycocyon

Senior Member
So it's a longer exposure on the second shot. Is that what you notice as being consistent?

I've not yet seen this on my D4, that is to say variation in exposures running at 10 fps and I have tested it quite a number of times on people (both stationary and moving). However I've not looked specifically at the first two frames. What happens after the second frame, does it stay brighter than the first frame?

I don't think it is changing light. The quality of the light in the images looks identical and clouds don't move that fast. Also the matter of the speed of light is not important because so many other things would affect the light other than how fast the light is traveling itself (like clouds).

I would run the tests on a white background....pure white and put the camera on a tripod with spot metering, then with matrix metering and compare them over a long cycle of maybe 20 frames using either A mode or manual mode. My guess is that if you are using matrix metering, the way that the computer is accessing the library of possible metering outcomes varies slightly as the frame is being sampled, similar to what you see with how the autofocus behaves in terms of sampling areas sometimes before reaching the face. At high frame rates the CPU has to keep up with both autofocus and metering. Also with an electromagnetic shutter, there are intermediate shutter speeds apart from the usual ones we are familiar with so the meter might be making slight adjustments based on this white person with bright white shirt against a white background. The meter is struggling to achieve a balance approaching 18% grey and with so much white it will be trying to balance the variables around a hovering ideal setpoint for that particular set of intensity values across the frame.
 
R

rocketman122

Senior Member
crycocyon said:
So it's a longer exposure on the second shot. Is that what you notice as being consistent?

I've not yet seen this on my D4, that is to say variation in exposures running at 10 fps and I have tested it quite a number of times on people (both stationary and moving). However I've not looked specifically at the first two frames. What happens after the second frame, does it stay brighter than the first frame?

I don't think it is changing light. The quality of the light in the images looks identical and clouds don't move that fast. Also the matter of the speed of light is not important because so many other things would affect the light other than how fast the light is traveling itself (like clouds).

I would run the tests on a white background....pure white and put the camera on a tripod with spot metering, then with matrix metering and compare them over a long cycle of maybe 20 frames using either A mode or manual mode. My guess is that if you are using matrix metering, the way that the computer is accessing the library of possible metering outcomes varies slightly as the frame is being sampled, similar to what you see with how the autofocus behaves in terms of sampling areas sometimes before reaching the face. At high frame rates the CPU has to keep up with both autofocus and metering. Also with an electromagnetic shutter, there are intermediate shutter speeds apart from the usual ones we are familiar with so the meter might be making slight adjustments based on this white person with bright white shirt against a white background. The meter is struggling to achieve a balance approaching 18% grey and with so much white it will be trying to balance the variables around a hovering ideal setpoint for that particular set of intensity values across the frame.
Click to expand...

there isnt anything the meter is doing as I shot in M mode. I set the exposure. thats what gets me confused. maybe the shutter isnt accurate. thats my only explanation.
 
C

crycocyon

Senior Member
Ok, well then again a test of a white background in static conditions would be in order, even of the cameras against each other. It would also be good to use an ambient light meter to set the shutter speed/aperture as the same on each camera. Then I would compare the histograms.
 
S

STM

Senior Member
Both images are overexposed, one just a little more than the other

Were you in A or M? I suspect you would not have that problem in manual. Also, at high frame rates and small apertures, the aperture may not close down to precisely the same setting with each shot. I think that is a lot more probable than an inaccurate shutter
 
Last edited:
R

rocketman122

Senior Member
STM said:
Both images are overexposed, one just a little more than the other

Were you in A or M? I suspect you would not have that problem in manual. Also, at high frame rates and small apertures, the aperture may not close down to precisely the same setting with each shot. I think that is a lot more probable than an inaccurate shutter
Click to expand...

I always overexpose 1/3 to 1/2 stop extra. its better to burn out some of the wrinkles. great for women. I then bring it back in acdsee if I have to. these were straight out of the camera and not touched. the exif is there. hover over the image. shot at f/4 in M. the lens matters not. 85/105 70-200. all the same inconsistencies. my best bud says his D4 and even D4s do it. I usually shoot 2-3 frames in rapid fire. click click, refocus again and another click click.
 
hark

hark

Administrator
Staff member
Super Mod
I wonder if it has anything to do with the speed of the burst rate...the faster the burst in continuous, the faster the aperture blades have to respond. Perhaps they aren't always able to keep up with some of the faster cameras? It looks like the second photo is overexposed slightly. Maybe the aperture wasn't quite stopped down when the shutter clicked.
 
E

Eyelight

Senior Member
If you correct the exposure of #2 to match #1 does the depth of field look a little shallower in #2?

It almost looks that way in the posted shots, but the more exposure of #2 would produce a little softness in the background anyway.

If it does not in these two images, you could take some shots with a more distant/detailed background and if the aperture is changing, it should be visible in the DOF change.
 
Marcel

Marcel

Happily retired
Staff member
Super Mod
I agree with Hark and think the aperture blades on that particular lens could be just a tiny bit sticky. Just enough so that in a fast continuous sequence, they don't have time to close to their supposed f stop after the second shot.
 
Top