D700 Underexposure Puzzle

Discussion in 'Nikon' started by rob f., Mar 13, 2012.

  1. I want to first show the scene that I used for this test. I think you'll agree it easily qualifies as the proverbial "average scene;" the kind that exposure meters are calibrated for. I imagine it probably reflects somewhere close to 18%. Blue northern sky; fully frontally lit red brick wall; white garage doors; medium gray asphalt; black car. Picture taken maybe 10:45AM on a sunny day. The fire station wall faces South.
    This picture was taken at f/2 with a 35mm MF lens with a CPU. I adjusted the exposure with the exposure compensation until the histogram was as close to the right as possible without crashing into the right edge. It took 1.3 stops of exposure comp to get this. OK that's just a little background. I haven't told you the problem yet.
    Standby while I try to get the image inserted.
    00a8am-450061584.jpg
     
  2. The next point is that I discovered this phenomenon by accident. I wanted to see how a couple of lenses performed at their wide apertures. I set the camera to "A" and the ISO to 400. Trusting the camera's auto-exposure, I shot the picture at f/2. f/2.8, etc. up through f/8 or f/11. When I downloaded the pictures, I saw that the ones at the widest apertures were badly underexposed. The wider the aperture, the worse the underexposure.
    So I went back the next day and re-shot the experiment, this time at ISO 200, and previewing and setting the exposure comp for a perfect histogram. Here are the settings needed:
    f/2: +1.3
    f/2.8: +1
    f/4: +.7
    f/5.6: +.3
    F/8: +0
    F/11: +0
    I then repeated the experiment with my 35mm f/2/8 PC-Nikkor and got the same result.
    Then I tried my 35/2 AF-D Nikkor and got the same result, except the AF-D required rather less exposure comp.
    Obviously the D-700 meter was overcorrecting for the additional light at wide apertures. But why?
    I tried with both matrix metering and center weighted. Not much difference there.
    Ever seen this one? Any ideas?
     
  3. Was shutter speed max/min out, say, via auto ISO or the like? At the scene, why not use manual exposure to figure out the problem right then?
    Just using sunny 16 (at EV16), the camera at f4 or more open, the SS would have been top over 1/8000...but then, it would have been over exposed, right?
     
  4. Some random thoughts:
    Half underexposed is not as bad as double overexposed. You could argue that filling the histogram only to the middle just loses one bit out of the 10 or 12 or 14 density/exposure/ADC bits you have. No big deal.
    (Some theoretician at U.of Chicago has a pretty good web page about exposure and noise and all that; not bad to read at all. Not sure it will answer your question, but it might make you feel better about the underexposure. On the other hand, that guy still hasn't found superstrings, so maybe his photography page isn't all that great either. I keed I keed....)
    Second random thought: besides the normal inherent imprecision in exposure measurement and the slight (or larger) fluctuations that come with cloud coverage etc., there are also errors having to do with what exactly makes it into the scene. Those errors might be harder to quantify. A slight change in direction might exclude a few very bright pixels that change the (matrix) metering situation significantly. This 'systematic' error could be minimized by doing a test with a tripod.
    And third it could also be that it's not the metering but the stop down that's not exact. I don't know how exact AIS and better lenses are, I know that non-linear stop down is supposed to be a problem with AI lenses.

    I know that the viewfinder screen in my D70 doesn't really get darker until I stop down to f/4 or so. But that's the opposite end of the mirror box from the meter, and (supposedly) the texture of the screen, or lack thereof, is a perfectly good explanation for this. I'm only mentioning it because I've been told there could be some feedback from the viewfinder to the metering. Any special screen in your camera?
    Well, let's hope that one of the actually knowledgeable people here answers soon.
     
  5. The shutter speed was probably maxed out at f/2. It was 1/8000. But that wouldn't explain the progressive incremental 1/3 exposure reduction from f/8 down to f/2. I don't guess auto-ISO was enabled. I set 200 manually. I could check the camera, though.
     
  6. Nope, auto-ISO is off. This camera must have 42,969 menu things to set up, though. I wonder what else could be set wrong?
     
  7. The shutter speed was probably maxed out at f/2. It was 1/8000. But that wouldn't explain the progressive incremental 1/3 exposure reduction from f/8 down to f/2.​
    Bracketing enabled?
     
  8. Bracketing. No, I don't think so. If it were, I guess I would be able to see all the bracketed shots in my download, no?
     
  9. If you're taking one shot at a time, this could happen; changing lenses doesn't reset the bracketing. By default, bracketing is tied to the FUNC button, which is the lower button on the front of the camera near the lens. Press that and see whether it says "0F" on the top display. If not, bracketing is enabled; rotate the main control dial (the one on the back) until it says 0F then test again.
     
  10. Maybe you didn't insert (or change) different (AIS) lens info i.e. largest aperture of each lenses into the d700?
     
  11. Bracketing makes sense...
     
  12. Leslie, that would only be an issue with the 35mm PC-Nikkor. The other two lenses are chipped. And I did have the correct non-CPU data entered and selected.
    Mark, my FUNC button is set up for non-CPU lens data. Could the culprit be lurking somehow in that choice?
     
  13. Leslie, that would only be an issue with the 35mm PC-Nikkor. The other two lenses are chipped. And I did have the correct non-CPU data entered and selected.
    Mark, my FUNC button is set up for non-CPU lens data. Could the culprit be lurking somehow in that choice?
     
  14. Leslie, that would only be an issue with the 35mm PC-Nikkor. The other two lenses are chipped. And I did have the correct non-CPU data entered and selected.
    Mark, my FUNC button is set up for non-CPU lens data. Could the culprit be lurking somehow in that choice?
     
  15. Leslie, that would only be an issue with the 35mm PC-Nikkor. The other two lenses are chipped. And I did have the correct non-CPU data entered and selected.
    Mark, my FUNC button is set up for non-CPU lens data. Could the culprit be lurking somehow in that choice?
     
  16. Leslie, that would only be an issue with the 35mm PC-Nikkor. The other two lenses are chipped. And I did have the correct non-CPU data entered and selected.
    Mark, my FUNC button is set up for non-CPU lens data. Could the culprit be lurking somehow in that choice?
     
  17. Leslie, that would only be an issue with the 35mm PC-Nikkor. The other two lenses are chipped. And I did have the correct non-CPU data entered and selected.
    Mark, my FUNC button is set up for non-CPU lens data. Could the culprit be lurking somehow in that choice?
     
  18. Mark: my FUNC button is set up for non-CPU lens data. I wonder if that is somehow related to the problem?
    Leslie: I did have the non-CPU data selected: 35mm, f/2.8 for the PC-Nikkor. The other lenses have CPUs.
     
  19. Rob, I'm not sure I should enter this debate about aperture inaccuracy, but you're not the first person to experience this phenomonen.
    Repeat the test in manual, ONE stop down on the Aperture, ONE stop up in shutter speed. they should be identical exposures.
    Can you borrow a modern G lens and run the same test. That would remove an chipped-or-not data errors and any aperture ring problems. If it still does it, it a metering issue. If it doesn't, it's a physical issue.
     
  20. I think every lens I have used in my life show different densities wide open; they all require compensation to get the same exposure as stopped down. Some just a bit, others a little more.
    I don`t know why. I use to think that the nominal aperture could not be the real one. Maybe lens testing must be done under laboratory conditions. Sincerely, I never care so much about it.
     
  21. Well, I guess that comparing the in camera jpg histograms at different apertures may not make sense. There are too many settings that can effect on the exposure.
    Method: use all manual settings: time aperture, ISO. Make comparisons then.
    Jose, Mike, I agree.
     
  22. OK. I will repeat the experiment on manual later this week.
     
  23. Rob, even if bracking isn't tied to your FUNC button, bracketing could still be your problem. I don't remember how to get there through the menus, but one way or another you should ensure that bracketing is disabled.
     
  24. Gup

    Gup Gup

    Rob, perhaps returning the body to 'default' settings and begining again with your own would eliminate some of the variables.
     
  25. I have a number of comments.

    First, I'm not sure that I understand the objective of your experiment. Could you please state that?

    Secondly, it's not a mid toned scene. White doors and clock, large shadow on dark pavement, reflective windows.

    Auto exposure in some cameras is linked to the location of the selected focus point. Did that change?

    1/8000 is at the limit of the shutter's capability. Systems don't perform optimally at their limits. It looks to be performing better closer to
    the middle of the range of shutter speed values.

    Auto exposure systems are somewhat random in my experience. Take ten shots in a row of the same scene in the same light, and there
    will be some variation.

    Did you set the aperture for your manual focus lenses in the menu system as recommended in the manual?

    Histograms depend on settings such as contrast and saturation.

    Auto exposure does not use the histogram, as the histogram is calculated after the shot, not before.

    There's no such thing as a perfect histogram.
     
  26. Dan, I'm curious to ..

    'Take ten shots in a row of the same scene in the same light, and there will be some variation.'

    Do you attribute this to shutter speed variation or inconsistant aperture closure?
    It's not meant to be a trick question, I happen to believe you!! I've just always wondered why??
     
  27. "... 35mm lens with a CPU.."​
    Did that lens start life as an Ai - not Ai-S - lens by any chance? If so that completely explains it, because Ai lenses don't have a linear aperture actuator and really aren't suitable for chipped use with camera controlled iris.
    Another thing to take account of is vignetting, which can easily be confused with under exposure and will change with every stop.
    BTW, what shutter speed was the camera giving for each aperture with no compensation applied? If the shutter speed was doubling or halving for each stop of aperture change, then the metering is OK and the camera's aperture coupling may be faulty.
     
  28. Rob, your D700 boost the ISO behind the scenes to compensate for the fact that the sensor's photosites are not receiving light that comes from an angle. This is only relevant at larger apertures.
    When the camera doesn't know exactly what lens you have, it can't do its behind-the-scene trickery. That results in underexposure that is more prominent as the aperture opens up. And that is what you are seeing.
    This can easily be tested. Take a f/1.4 G lens and shoot it wide open with manual settings. Put some tape over the contacts and shoot the same image again with the same exposure settings. The exposure should be the same but it's not - the second image will be underexposed. This applies to DX cameras as well but not to the same degree.
    PS. While apertures might be non-linear, incorrectly aligned and what not. But when you shoot wide open the aperture is not used.
    BTW, dxomark has an article about the behind-the-scenes iso boost to compensate for the sensor "problems" at wide apertures.
     
  29. Mike, I think the AE sometimes toggles between two borderline exposures. For instance, 1/250 and 1/320 at the same f-
    stop. Manual exposure prevents toggling by locking in one value or another. If you're shooting a small number of
    images, particularly raw files, it's not a big problem. This is one of the reasons why it's recommended to shoot a time
    series in manual mode.
     
  30. Wow, take a number, guys. It may take a few days to respond to all this. OK I'll start with:
    @Rodeo Joe: OK, it's a Zeiss ZF.2. AIs from the start, AFAIK.
    @Dan South: The objective is to find out why some shots I took at wide apertures were grossly underexposed. I shot with Aperture Priority, so you would think the camera would give the right exposure. OK no such thing as a perfect histogram (or parent); but (as with parents) there is such a thing as a good enough histogram. And I think mine were good enough. They were not crashing into the left wall or the right wall. And when I adjusted the exposure to achieve that, the resulting picture was fine. I think that my selection of subject qualifies as the so-called "average scene." An average scene can contain some white, some black, some shadows, some sky, etc. etc. I don't know if mine reflected 18%; but I'm sure it's not 5% or 65%. It's in the ballpark.
    As to those who question that the camera ran out of shutter speeds: well, I think t did, at 1/8000. BUT: (this is a big but) the inability to give a shorter exposure at wide apertures should cause increased exposure, not decreased.
    At this point I think Mark's suspicion that it could be bracketing should be followed up. And Gup Jeffries suggestion to return to default may be necessary. And trying manual exposure should be done.
    Thanks for all the interest! I'm getting a lot of help here. I would hate to be in the Galapagos or even Southern Missouri, and get home to find out my exposures were all blown. Score one for chimping and checking histograms, I guess. More later . . .
     
  31. About some of the variables: lens contrast depends on aperture. So at bigger apertures contrast is lower. This leads to different shape of the histogram. Also vignetting at bigger apertures has it's effect on the histogram and histogram shape. And active contrast adjustment and auto vignette control,,,,
    So comparing exposures of different apertures based on histograms are not generally comparable side by side.
    Just too many variables that may be interconnected.
     
  32. >> I shot with Aperture Priority, so you would think the camera would give the right exposure.

    I would never make such an assumption. I've seen auto exposure modes create exposures that are off by 2.5 stops on
    many occasions.

    >> I think my selection of the subject qualifies as a so-called "average scene".

    It doesn't matter whether the scene was 'average', because the camera is not taking an average of the scene. Rather it is giving different weights of importance to each part of the scene based upon its own internal algorithms. Maybe it's giving more weight to the white areas. If you rearranged the
    scene, you could probably find a layout that produced over exposure rather than under exposure. Auto exposure is highly
    unpredictable even in the most sophisticated cameras, because we don't know how it's making these evaluations.
     
  33. Rather it is giving different weights of importance to each part of the scene based upon its own internal algorithms. Maybe it's giving more weight to the white areas.​
    Dan: Yes! I think that may be exactly what has been happening. I took another set of shots, this time in my backyard. It contains a small white area in an otherwise fairly medium-refectance scene. After examining those shots, as well as the ones of the scene I posted, it looks like the camera wants to make sure that no highlights are blown, even if they only occupy a small, insignificant part of the picture. The camera is willing to sacrifice the darker areas if necessary to accomplish this. My shots of the fire station had a small white area too: the garage doors. Not that large, but enough to ruin the exposure.
    Switching from Auto to Manual didn't make a lot of difference, nor did switching between Multipattern and center-weighted. Spot is more useful, because I can pick the area I think is of a medium reflectance.
    There are times when it's hard to beat an incident light meter.
    I think I'll re-shoot my tests with these ideas in mind. I bet it will go better. It's good to try these things while walking the dog in the AM. It saves blowing shots, during a trip, that would be hard (expensive) to go back and redo. Know your camera.
     
  34. Rob,
    'Switching from Auto to Manual didn't make a lot of difference......'
    Not sure I get you there, Auto...camera sets stuff. Manual......You set stuff....!!! Do you get incorrect exposures if you do a manually set sequence of shots? The set should be one stop up on aperture AND a stop down on shutter, CLICK, next pair....... or visa versa depending on your preference, as long as it's up one and down the other... :)
    I wish my D700 had your metering that prevented highlights blowing out. Full 3D matrix and if I happen to have a mid-tone rider with a dark horse and a nice white saddle cloth, set against leafy trees, it's gone. >255...blinky blinky..... :-(
     
  35. Switching from Auto to Manual didn't make a lot of difference......'
    Not sure I get you there, Auto...camera sets stuff. Manual......You set stuff....!!!​
    That's right. Agreed. It makes a difference in the procedure, not in the results.
    Do you get incorrect exposures if you do a manually set sequence of shots?​
    Yes. Underexposed either way at the widest apertures.
    The set should be one stop up on aperture AND a stop down on shutter, CLICK, next pair.......​
    Right. That's what I did. e.g. 1/5000 at f/2, 1/2500 at f/2.8, etc.
    I wish my D700 had your metering that prevented highlights blowing out. Full 3D matrix and if I happen to have a mid-tone rider with a dark horse and a nice white saddle cloth, set against leafy trees, it's gone. >255...blinky blinky..... :-(​
    Not sure about this. After reading Dan's post, and reviewing these two sets of shots, it looked as if that might be happening. But then again, I have gotten blown highlights in other shots, with >255--> blinky blinky. And the protection-of-highlights theory would not explain why the camera underexposes at the widest apertures. By f/8 and f/11 it's OK. Starting from f/8, each wider aperture needs another 1/3 stop compensation to get the same exposure as the previous stop. It should be noted that I was using apertures not often used in bright daylight. That could be asking the camera to do something it wasn't designed for. (???) Maybe it got overwhelmed, and overreacted by raising the shutter speed too much.
     
  36. Rob, your camera doesn't care about "blown highlights", because it can't whether a pixel will reach its limit until AFTER it makes the exposure. The meter doesn't do a pixel by pixel check of light intensity. It meters big chunks of the scene. Only after the image has been captured and the camera can examine the image pixel by pixel does it notice pixels that exceed the range of the sensor.
    To complicate matters further, the blown highlights are blown in the JPEG preview file. They might also be blown in the RAW file, or they might not. Even your camera doesn't know at this point. Histograms are not based on RAW data; they're based on the JPEG preview.
    A blown highlight is not necessarily an error. Specular highlights such as the shiny spots on a clean automobile in direct sunlight will exceed the sensor's dynamic range. That's not an error. If you exposed to capture these meaningless highlights, the important parts of the frame would be grossly underexposed.
    Your experiment leads to one important conclusion: exposure metering is not foolproof. Even if you used a gray card, some shadows highlights might have been lost. Ditto if you had used an incident meter. In part this is because there's nor really any such thing as a correct exposure. You might want to expose for a person's face and not worry about blowing out a white sky behind them. Or you might want to keep the sky and boost the level of their face in post production. The camera doesn't know what you want to do.
    And what about that perfect histogram? Think about the gleaming car in the sunshine. What would the perfect histogram be for that shot? Expose to the right? Not unless you want your red car to look black.
    Unexplained over and under exposure? Welcome to the real world. :) That's why we bracket. I'm not being facetious. I'm being practical. Getting a good, workable exposure requires experience and a good knowledge of how the equipment works under various conditions. And it helps to have multiple samples (bracketed shots) from which to choose after the fact.
    Good luck, and keep experimenting!
     
  37. I have noticed consistent underexposure when using some manual lenses on my D700 for a couple of years at least - irrespective of metering mode but contingent, of course, on the subject metered. I notice this particularly on a 200 mm F4 converted to AI and a 20mm F3.5. My 16mm 2.8 FE doesn't seem to be as susceptible. I adopt similar compensation values to that suggested by the OP.
    I posted a question about this in a couple of places but no one came up with any suggestions that I hadn't already tried so I just live with it. For reference the camera gives consistent and expected exposures using my AF S and D lenses. It does however underexpose to the same degree when using my Sigma 150-500, which might be a clue, if I was inclined to investigate further.
    Shooting RAW with the D700 gives a surprising amount of latitude for overexposure (> 1 stop) so I tend to err on the + side anyway, unlike in the good old days of slide film.
    Roy
     
  38. I agree with absolutely everything that's been said. Still, there is something funny going on. Here's another example. Today it's the 24mm AF-D Nikkor at f/2.8. No exposure comp in this first picture.
    00a9bR-450997584.jpg
     
  39. Lovely exposure, eh? That's what I get for trusting the camera. Let's try again, with exposure compensation:
    00a9bU-450999584.jpg
     
  40. Those two were with spot metering. Here's one with matrix metering, which was a better reading for this scene. This one required 2/3 stop correction to get this exposure:
    00a9bd-451001584.jpg
     
  41. See my point? After we have said all there is to say about there being no correct exposure; no perfect histogram; pixels; metering modes, etc., the camera still wants to underexpose at the wide apertures. A shot taken at f/8 needs little or no exposure comp, at least for this scene and several others I have tried in the last few days. So I think there is something going on. I ought to post one at f/8 or f/11 so you can see.
     
  42. Here's one at f/11, no exposure comp. Matrix metering.
    00a9bs-451003584.jpg
     
  43. Spot looks about right for the 5mm or so central area........making that bit 'mid-tone' ... and ignore the rest!
    How about Centre-Weighted? I often find that a good balance between too-tiny centre bit and all!
     
  44. What exactly are you spot metering on in the under exposed image. Running diagonals across that image, it looks like dead center might be just below the two rows of dark windows in the left door. If that is the spot location, the exposure is not as far off as it seems, as that lower door is visually in the ballpark of an 18% grey.
     
  45. @ Mike: Yes I am beginning to feel that center-weighted is probably the best single metering method.
    @Bob: Agreed, I probably shouldn't have used that spot reading as the basis for illustrating the underexposure phenomenon. If there seems to be a reason, and I can find the patience, I may do another pair with a center-weighted exposure as the baseline. But I think I should stop pretty soon!
     
  46. there is something funny going on.​
    The underexposed shot a few posts back was completely predictable. You said that you used spot metering. The spot was pointing at the garage door which appears to be quite a bit more reflective than 18 percent gray. Spot meter something that's more reflective than 18 percent gray, and you'll get an underexposure every time. The spot meter ignored the rest of the building, the sky, and the roadway and focused on that one bright point.
    Center-weighted metering looked at more of the scene. It still gave the highest priority to the reflective garage doors, but because it considered the rest of the scene, it underexposed less than the spot meter. Again, an entirely predictable result. Nothing strange or funny about it.
    I'm not sure why you even bother to try to quality these scenes as "average" when you use a spot meter and point it at some highly reflective object within the frame. The camera is doing exactly what you're telling it to do.
    If you want to understand your cameras metering system better, find a much simpler scene, i.e. one that doesn't contain white and highly reflective objects, shadows, dark pavement, etc.. A single color and a single tone would be a great place to start. Meter a blue sky, a white wall, a brown cardboard box, a yellow T-shirt, or a field of green grass. Understand how your camera's metering modes work in isolation first, then work your way up to more complex scenes with mixed tones and levels of reflectance.
     
  47. Dan: the spot reading was taken off the brick wall and then I recomposed for the shot. Trust me, the camera underexposes more and more at wider and wider apertures. That's the point of all this. Each progressively wider aperture seems to need about 1/3 stop more compensation to get the same exposure as I get without exposure comp at f/8. I appreciate the difference metering off of various areas makes, but I know all that already.
     
  48. Hi Rob,
    Maybe these will be silly questions, but:
    Do you have D-Lighting ON or OFF?
    When you half-press the shutter-release button does it block the exposure reading when you recompose?
    Have you tried to start with mode P (the camera on a tripod) in a way the chosen aperture is f:8 (as you say it wouldn't need compensation) and then modifying the measures on the wheels: first cutting speeds and afterwards doing the same changing the apertures till you get to the wider one?
    I never experienced something like the experiences of yours.
    Regards
     
  49. OK, Have 'we' done the M set of pix yet? On the D700, in Manual with an AI lens you can turn the aperture ring to set the f stop and take a pic.
    Find a nice stable, but not too bright seen. Pop on tripod. Lock ISO at 200. Turn off Active D-Lighting. Assuming something like 125/f8 being optimum... Set it to 1/4000@1.4 click > 1/2000@2 click> 1/1000@2.8 click> 1/500@4 click> 1/250@5.6 click > 1/125@8 etc etc... they should all be a very, very similar exposure. Histograms will vary in width depending on vignetting and contrast will increase as the aperture closes and decreases as diffraction starts at f11 or so. I would expect the peak to be in the same place histogram wise.
    If the peak moves across the histogram field then the 'stops' are inaccurately 'notched' on the lens or the shutter is off. I suspect the former.
    If it remains stationary, then the variation you're seeing in 'normal' use is the camera's doing and that's a metering variable.
    Without discounting the physical aperture sorting out the metering is impossible.
     
  50. OK. Starting from scratch. No sun today, and a new subject: my backyard/rose garden. By request: ISO at 200; auto-ISO off; Active D-lighting off; bracketing off. Also: no exposure compensation. We are just going to look to see what the camera does, not compensate for it. And therefore, no histogram peeking. Histogram, shmistogram; we will just look at the pictures. No more spot metering. The first shot is with center-weighted metering.
    I deliberately chose my worst vignetting (and most expensive) lens, the Zeiss 35mm ZF.2.
    00a9n4-451157584.jpg
     
  51. Now we will look at f/4. Still with Center-weighted. Note how the vignetting clears up, doing away with much of the underexposure.
    00a9nC-451161584.jpg
     
  52. Next is at f/8. Everything else is still the same. Camera on tripod for all shots, as requested.
    00a9nH-451163684.jpg
     
  53. You would think the center weighted metering would notice the dark, vignetted corners and respond by lightening the picture a tad. Well, maybe it did. But wait: maybe matrix metering is the answer. Let's see if it will notice those dark corners and do something about it. Back to f/2, with matrix metering:
    00a9nT-451167684.jpg
     
  54. Gee, not so good, eh? Center-weighted was better. Let's save Photo.net a little disk space and just skip to f/8:
    00a9nd-451173584.jpg
     
  55. Oh, I almost forgot--just one more thing (Lt. Columbo). Let's go back to f/2, and center-weighted metering, and this time bump the vignetting control up to maximum:
    00a9nj-451175684.jpg
     
  56. Does it look any better than on normal vignetting control? Maybe a little.
    In this series vignetting seems to explain a lot of the underexposure. That's got to be part of what is going on. But in the previous series of the fire station, when I used the 35/2 AF, which does not vignette so much, I still had this effect of underexposure at wider apertures. It just wasn't as pronounced. I wonder if I should try with a different lens now? Maybe the 35/2 AF, or the normal 50/1.4, or the 50/1.8? But first, does anyone want to see that?
     
  57. Rob, Manual means No Metering, inc No Vignette Control! Still don't know if it's a lens or metering issue.....?
     
  58. Rob, Manual means No Metering, inc No Vignette Control!
    I don't agree. I see no reason why you can't use the meter during manual exposure. I've been doing it for years--since I got my FE2. Before that, I did it with a Weston Master IV, with Leica M2 and Nikon F. You need a meter reading, for goodness sake.
     
  59. Rob, you completely miss the point of my post. Use the meter ONCE to get a generic exposure and then take a separate picture with each of the different aperture/shutter speed combos of the same exposure to see if the variation is the lens' aperture. They should all be of identical brightness with a very similar histogram position.
    You do get that 125th@f8 should give exactly the same exposure 'brightness' as 250th@f5.6 right ?
    If this makes no sense.... :-(
     
  60. No sun today​
    Unfortunately, overcast conditions are a poor choice to do controlled tests. Of course there there is sun today. It's above the overcast. If there are scattered clouds between the overcast and the sun, you will have variable lighting conditions.
    When using aperture priority mode, you should still pay attention to the shutter speed selected by the camera. Close down the lens a stop and you should see the shutter speed slow down a stop.
    You don't even need to take a picture to see if the metering system is working sanely. Put the camera on a tripod and watch the shutter speed change as you change the aperture.
    In the three photos above with center weighted metering at f/2, f/4, and f/8, the camera chose the following shutter speeds
    • f/2 - 1/2500
    • f/4 - 1/640
    • f/8 - 1/125
    In aperture priority mode, the displayed shutter speeds might be rounded from the actual shutter speed selected by the meter. f/8 @ 1/125 will collect about 1/3 stop more light than f/2 @ 1/2500.

    But the two samples show way more variation than 1/3 stop. The most likely explanation for this is that the ambient lighting changed.

    If the ambient lighting were consistent, then one explanation for the brighter image at the small apertures is that the aperture blades are sticky and aren't stopping down properly. A problem with the stop down lever could be another explanation.
    What you should do first is to confirm that the camera is working correctly when using manual exposures. Pick an indoor scene with constant lighting. Manually meter to set correct exposure then take a series stopping down the lens one stop each time the shutter speed is reduced one stop. All the photos should be consistent and the RGB histograms should be nearly identical.
    After doing the manual exposure test, you can proceed to aperture priority with matrix metering. Changing the aperture one stop should result in a shutter speed change of one stop. All the images should be consistent and have nearly identical histograms.
    After passing each of the above tests independent, you can begin to wonder why matrix metering isn't giving the same image as as with center weighted metering. For an average indoor scene, I think matrix metering will give similar results.
     
  61. Any chance the settings on this menu item were changed:
    b6 Fine tune optimal exposure
    I mention this only because A) this will not be recorded in the exif as exposure compensation B) spot, center-weighted, and matrix can be changed independently (causing greater that expected variations between the 3 types), and C) neither the 2-finger reset nor the other menu reset option will restore these values if they were changed.
    Just a WAG. Be a shame to do all this testing if it was something you or someone else set and forgot about.
     
  62. Sounds as good as anything else, Steve. It might explain why the multi-pattern metering didn't do what I thought it should. I'll check it out.
     
  63. Rob, you completely miss the point of my post. Use the meter ONCE to get a generic exposure and then take a separate picture with each of the different aperture/shutter speed combos of the same exposure to see if the variation is the lens' aperture. They should all be of identical brightness with a very similar histogram position.
    You do get that 125th@f8 should give exactly the same exposure 'brightness' as 250th@f5.6 right ?
    If this makes no sense.... :-(​
    Umm . . . I understood the reciprocity of shutter speed and aperture by the time I was 12. My dad explained it and I grasped it instantly. That's when I started shooting with his Leica II and Weston Meter. I'm 71 now and still have the M2 that I bought at 19 with my savings out of my USAF paycheck(s).
    Right, I didn't do it your way. Not yet. Doesn't mean I won't. But I wanted to demonstrate what the camera is doing under actual shooting conditions. One does not expect, in the normal course of things, to set shutter speeds by mental calculation when using such a sophisticated camera. With a manual Leica, I do it all the time. So far I am just trying to demonstrate that the D700--my D700--makes the picture darker when you would think that, if anything, at a wider aperture, it would make it lighter. Or hopefully give the same exposure.
    At this particular point, I looking at vignetting as a contributing factor. we'll see.
     
  64. If the ambient lighting were consistent, then one explanation for the brighter image at the small apertures is that the aperture blades are sticky and aren't stopping down properly. A problem with the stop down lever could be another explanation.​
    The outdoor light was pretty consistent, especially on the cloudless days of the first series. We can rule that out. As to the possibility of mechanical stickiness, I did get the same effect with three lenses, which would seem to rule out the lenses. As to the stop-down lever, well I got pretty normal exposures--I think--with those three lenses plus a fourth, at f/8 and f/11 yesterday. That seems to rule out the stop-down lever (seems that way to me at the moment). However I am going to go ahead now with the fully manual experiment, and we will see what happens.
    Good night, everyone. Thanks for the pointers.
     
  65. When you do the "full manual" experiment, set the aperture with the aperture ring, instead of electronically. That should
    help you isolate whether aperture control issues are part of your problem.

    Also, you mentioned that the camera might meter differently near wide open because it sees vignetting. Since the camera
    always meters with the lens wide open, this should make no difference.
     
  66. OK. Here we go. Camera on tripod, set to manual. Sunny day, cloudless sky. By Sunny 16, exposure should be 1/200 at 16. Spectra Professional meter confirms: 1/200 at 16. D700 set to 200, f/16; exposure indicator zeroes out at 1/160. So be it, we will use 1/1/60. here's what I got at f/2:
     
  67. Well, the picture didn't upload. What I got was severely underexposed at f/2. Stopdown mechanisms shouldn't matter, since the diaphragm is wide open for both metering and exposure. I'll try again:
    00aAL8-451613584.jpg
     
  68. Vignette control was still set to "high" since I forgot to turn it off. Here's what I got at F/16. Looks good to me. I can skip the in-between pictures. The exposure increased with each successive stop, until f/11, which is nearly the same as f/16.
     
  69. Vignette control was still set to "high" since I forgot to turn it off. Here's what I got at F/16. Looks good to me. I can skip the in-between pictures. The exposure increased with each successive stop, until f/11, which is nearly the same as f/16.
    00aALB-451615584.jpg
     
  70. The problem is the shutter is not accurate at higher shutter speeds. The camera needs to be serviced.
     
  71. You began the series above with a manual camera setting of 1/8000 @ f/2, so the series of photos you should have taken was:
    1/8000 @ f/2
    1/4000 @ f/2.8
    1/2000 @ f/4
    1/1000 @ f/5.6
    1/500 @ f/8
    1/250 @ f/11
    1/125 @ f/16
    I'm not sure whether you understand that the metering mode is irrelevant since you pointed out you were using center-weighted metering.
    Whether or not the exposure is even accurate doesn't really matter for the above series.
    What matters is that the above series should give images with approximately the same density (film) or histogram (digital). They could all be underexposed (by the same amount) or overexposed (by the same amount) or even perfectly exposed. What matters is that each of them should be pretty much the same.
    All of the other camera settings are taken out of the equation. You don't have to puzzle about how matrix metering magically decides what exposure to use. You don't have to wonder whether you spot metered off a good spot. It won't matter whether vignette control is on or off as long as it is at the same setting for all shots.
    You did make a minor mistake on your f/16 image as it was shot at 1/160" instead of 1/125" which is what was needed to get the same exposure as at f/2. But the f/16 image would have been even brighter which would still indicate the problem at high shutter speeds.
    I've heard that replacing a shutter isn't terribly expensive but if you're lucky, maybe there's a calibration procedure that can be done.
    Good luck.
     
  72. 'So far I am just trying to demonstrate that the D700--my D700--makes the picture darker when you would think that, if anything, at a wider aperture, it would make it lighter'
    I suspect the entity, known as your D700, is NOT doing anything of the sort. It's telling the shutter to stay open for X, and the aperture to shut down to Y. Full stop. However, one or the other, or possibly both, are NOT doing as they are told. The D700 won't know a thing about it.
    Metering is done wide open, so any failure to close quickly enough is an aperture error and usually results in overly bright or erratic exposures.
    Shutters are usually very reliable, but when beginning to fail, the very top end and the very bottom tend to suffer first.
    Usually the way of differentiating is to see if it effects all lenses the same = shutter
    or one lens, maybe 2 if old lenses, = sluggish apertures.
     
  73. Tom: Although the series was presented with f/2 first and f/16 last, I actually shot the other way, beginning at f/16 and 1/160. Then it went: f/11, 1/320; f/8, 1/640; f/5.6, 1/1250; f/4, 1/2500; f/2.8, 1/5000; and f/2, 1/8000. Note that the final speed should be 1/10,000, but the camera only goes to 1/8000. That error, however, ought to be on the plus side. It should have made the picture 1/3 stop more exposed.
    The points about the metering mode, vignette control, etc are well taken--all we really care about is that the exposures are either consistent, or else they are not. And they are not.
    Mike, I see you have reached the same conclusion, concerning the shutter. One thing bothers me, though: Let's note that the correct shutter speed would have been 1/10,000 at f/2. And the picture at f/2 was, just offhand, maybe 3 to 4 stops underexposed. To be even 3 stops under, the shutter would have to be operating at 1/80,000. For 4 stops, it would take 1/160,000. That's a pretty fast shutter, guys! The D700 only goes to 1/8000. Problem there: how can it be the shutter?
    This makes me think more along the lines of some sort of reciprocity failure of the sensor. It's not putting out the full quota of electrons when the exposure is that brief. Does that make sense?
    00aAPO-451669584.jpg
     
  74. I guess those are indicated speeds. If the first blind is 'late' or the second 'early', it would act like a faster shutter.
    However, I still favour the lazy apertures.
    Borrow a new(ish) G lens and do a Manual Series (once only metering to set a basis) series of shots. If they are not near identical, I'd think Shutter. However, if they are very close, it's sticky old Apertures in the other lenses.
     
  75. It's not a question of lazy apertures. The meter reading on the scene was 1/160 at f/16 and that image was exposed properly. A lazy aperture would have caused a shot at f/16 to be overexposed while the shot at f/2 would have been correctly exposed. Instead the opposite happened.
    The posted example labeled f/4 at 1/2500 is actually the same file as f/16 at 1/160. Rob - you said "I can skip the in-between pictures." Did you even bother to look at them? You should have noticed the progressive underexposure as you moved form 1/160 to 1/8000.


    And your comments about shutter speeds of 1/80,000" and 1/160,000" suggest you don't understand how a focal plane shutter works.


    The curtains of a focal plane shutter move at the same speed irregardless of shutter speed. Exposure is controlled by varying the delay between the first curtain opening and the second curtain closing. At speeds above 1/250" (flash sync speed), the second curtain begins closing before the first curtain is fully open. As the exposure time decreases ("faster" shutter speed is really a misnomer), a progressively smaller slit moves across the frame. For more detail, search on something like "How does a focal plane shutter work".
    Your shutter isn't working right. Send it in for a repair.
     
  76. Hi Rob,
    Never eared of something as "reciprocity failure of the sensor" and the concept applied to film was at the opposite side - long exposures.
    As you started from F:16 to maximum aperture and you did use lenses with automatic diaphragm there was no reason to have maximum effect with the lens wide open - as at this set it would need to move the blades.
    It doesn't seem to be a metering problem as you measure a correct exposure at F:16 and adjust the other ones manually.
    On the speed side it would only make sense if the camera was fastening the speeds too much and when you cut from 1/10000 to 1/8000 you still get the effect - and 1/8000 is the maximum speed your camera can get.
    Considering all this, the situation is rather strange and we can make only guesses but I thing only service can give you an answer. Maybe a processing or circuits problem, related to a component or charges control, but who knows?
    Regards,
     
  77. Well, thanks for all the time and trouble you all have put into your responses. The vast majority were not only helpful, but polite and respectful. The D700 is now on its way to Nikon, and we will see what they come up with.
    The repair will take 6 to 8 weeks! At least I have my D300 in the meantime.
    Thanks again.
     
  78. I have that problem with a lot of my cameras (film SLR and DSLR), it's shutter speed inacuracy. The fastest speed (1/8000 in your case) is mechanical speed without any delay. This speed is supposed to be calibrated well in the factory but it is not accurate now. You got underexposed because this highest speed is too fast, almost like the shutter opens and closes right away. All the other speeds is calculated after that by adding a delay. This delay time is usually accurate but because of the involvement of the wrong 1/8000, the slower speeds are also wrong. However, the slower the shutter speed, the less effect of the mechanical speed and the result is better
    It's like you lost 1/10000 sec of exposure time for every exposure and that is nothing when ss is 1/250.It's more important when ss is 1/3200
     
  79. In which case put a 4 stop ND filter on and redo the test. That will take all of the high end shutter speeds out of the equation.
    If the 'problem' still exists, it's not the high shutter speeds.
     
  80. John, that sounds like the clearest explanation yet. Not a very happy one, but clear. If 1/8000 is only a mechanical speed, not electronically timed, and all the electronically timed speeds are dependent on 1/8000, it sounds like ongoing trouble, since mechanical speeds--especially fast ones--get out of whack at the drop of a hat.
    Mike, my D700 is now in the hands of the Nikon techs. The ball's in their court now. Not a bad idea, though.
     
  81. Rob, thank God for your post. I have a d700 with the exact same darn problem. I have posted on a couple of sites and spoken to Nikon, but no one can figure it out.
    For me, anything below 50mm e.g. 35mm requires exposure comp. The 35mm F2 requires between .3 and 1 depending on setting, while my Sigma 12-24mm at 12mm can require +3! I also used 28mm AIS and 20mm both which had the same results.
    Please contact me through website to see what Nikon finds!
    I love my D700 but have always been frusted with this issue.
     
  82. So . . . the plot thickens. Could be it's a design glitch, in which case I sent my camera in for nothing. Funny I was never aware of it until recently. I shoot wide-angle all the time, every focal length from 15mm on up.
    We will see what they say. I'll let you know. Thanks for that info!
     
  83. Nikon returned the camera with a note saying there were no exposure issues. Then the camera store staff figured out that the problem was being caused by my having aperture control set to the aperture ring. While it was at Nikon, they changed it to the command sub-dial. The camera now exposes normally at wide apertures/high shutter speeds. The mystery is apparently solved.
     
  84. Rob, thanks for the update. It's great that you can get correction exposures across all f-stop and shutter speed settings now.
    But wouldn't this mean that there is some sort of problem related to the AI meter coupling ring. My D700 is out on loan. But after I get it back I'll put some non-G lenses on it and see what results I get using the aperture ring to set f-stop.
    Good catch by the store staff by the way.
     
  85. So, nothing wrong what-so-ever with the metering, just mechanics. Excellent.
    If setting the aperture via command sub-dial has 'corrected' all errors, as Tom says, there's still something very hinky with the mechanism. Using the A-ring to set the aperture doesn't actually move anything other than the 'position-to-stop' for the blades on stop-down when the shutter is tripped.
    Testing! If you do a simple manually set test 1) by using the A-ring, and the same set with 2) the Command dial and 'pair' them, you might see where it's going funny.
     
  86. Not a bad idea. I might just try that!
     
  87. Could this be what the new firmware update is for? Or is it just that without the D lens info enabling 3d matrix metering
    it's not as good at compensating for highlights and back lights?
     
  88. Looks like more testing for Rob! :)
    My D700 has -0.3>-0.7EV set and locked for nearly all outdoor, ie not studio, use. I'll see how it is this weekend then I'm out again.
     
  89. EricM

    EricM Planet Eric

    I'm going through the same problem with a new used D700. 24mm af-d. It doesn't matter what meter modes, but I have unpredictable under exposures when wide open. Is there a solution to this, Rob?
     
  90. Eric, does this result in shutter speeds over 1000th sec? I wonder if the correspondingly fast shutter speed is proving problematic to achieve? I'd expect this to over-expose, ie can't go fast enough, but the leading and following shutter blades work kinda differently.
    How many actuations does your 'new' used D700 have?
     
  91. EricM

    EricM Planet Eric

    Mike, it has 31,000 actuations. I just put on my my 24 prime and started shooting a friends kitten and dog under even lighting condition, in a living room, on a couch, all at f2.8, A priority, and most frames are releasing around 1/60th then sudden random under exposure at 1/250 and then the next at 1/640, then the next back to normal around 1/60. Then I moved on to anther part of the living room and it starts at 1/800, 1/1000, 1/800, then back to normal at 1/250. I can't replicate the problem on either matrix or CW metering. It just does it on either or metering option.
     

Share This Page