Permanent slight overexposure with D700

Discussion in 'Nikon' started by monika_epsefass, Jun 11, 2009.

  1. Hello,
    I've noticed a strange phenomenon with combinations of D700 and some lenses, most notably the 24-85/2.8-4 and the 50/1.4: most of my photos will always be slightly overexposed. Adjusted in postprocessing to -0.4 or -0.5 they will be fine. Is it a lens issue?
    My most current way of working is in A mode (selecting f-stops manually). I have no indication while shooting that my pictures may be too bright.
    Any advice? Thanks.
  2. Monika,
    I cannot explain this, however, it may give you a little reassurance if others report similar results. I have a Sigma AF 14mm f/3.5 lens which consistantly over exposes by a full stop. When using this on my D700 I simply use M mode and adjust the meter to underexpose by exactly one stop - this gives me the exposure I expect from all my other lenses. I don't know why this is so as the lens appears to have healthy aperture blades and the aperture arm adjuster operates freely and cleanly. i just make the compensation and live with the lens. I'm interested to hear of others experiences with this....
  3. The standard questions are:
    1) How do you judge overexposure?
    2) Are you shooting RAW and if so how do you convert raw files and post process?
    Do you evaluate exposure by the histogram in post processing? Is it possible to recover over exposed areas?
    The problem is that not only the camera and its settings (including the interaction of the metering with the lens) are involved but also the raw conversion and the post processing.
    Be aware that matrix metering is not "just light metering" but it includes a decision by the camera to give the "best" exposure to the scene. This will be different for different angles of view even in the same scene and light conditions. "Just" metering can be done by selecting spot metering or to meter by a hand-held light meter. Center weighed metering already averages usually a number of objects in the frame and can result in variations of exposure with different angles of view. But it eliminates the "intelligent guess" by the camera and therefore is a mode of choice for many photographers who prefer to makes their own intelligent guesses about "best" exposure.
    To test if it is a simple problem of the two lenses you could shoot a uniform object covering the entire frame in manual setting with different lenses across the whole range of f-stops. The fact that it applies to two lenses not just one is an indication that the problem may be not so simple and involve many of the mentioned factors.
    Last but not least examples posted are most helpfull to give a helpful suggestion.
  4. Thanks for the insight Walter. I actually did try a side by side test with the Sigma 14mm against the 17-35mm Nikkor @17mm whilst I was on holidays a few months ago. Shooting the same scene, tripod mounted in steady ambient lighting using centre weighted metering and same camera settings in M mode with the centre focus point on the D700 the Sigma was a clear 1.0 stop over the 17-35mm. I tried different aperture settings on both lenses from f/3.5 to f/11 in 4 steps, the results were uniform over exposure for the Sigma. This was when I first had opportunity to shoot with the Sigma 14mm, when I returned home and opened the initial over exposed files from the Sigma which I had not deleted, the exposure slider in PS CS Raw converter brought back the exposure to a point where it was comparable to the same shots taken with the Nikkor at 17mm. I did not look at the in camera histograms, I use the PS RAW converter histogram on occasion to gauge my exposure, the Sigma exposures if not dealt with in camera are always needing shifting to the left.
    I know it's only a rudimentary testing, however, on the subsequent occasions I've shot with the Sigma in all light and scene types the same exposure anomaly has occured and to the same extent without fail so I live with it and continue to make the necessary adjustments..... If I had kept an overexposed file I would have posted it here - maybe I'll re-visit this on the weekend.
  5. I have a couple of lenses that expose quite differently on the same camera - one tends to underexpose a bit while the other tends to overexpose. I simply adjust my exposure compensation on my camera before I start shooting as needed for the lens I am using.
    You may want to consider setting your expose compensation to -.3 or -.7 when using that lens.
  6. @ Walter: I'm not too good with histograms, albeit knowing the basics. Mostly, I judge by overall impression, which I consider to appear much brighter than I had in mind.
    I shoot raw and postprocess in Bridge. Usually, lights can be recovered, which often is not even necessary after having readjusted the exposure values.
    I haven't said a word about matrix measuring. But, given that you mention it, I mostly work with single field measuring - might it stem from that? (Don't like matrix too much, as it does as it pleases) Just measuring a spot which does not express 18% gray? Gets me thinking here... Nevertheless, I've experienced the same phenomenon with one sole lens on my Canon before, and not with all the others I had, so as others reassure me, it may as well be the lens, and not necessarily me. I do not think that I measure so differently when using different lenses... :cool:
  7. Monika it may still be worth a short test shooting in manual shooting the lenses with same aperture and exposure time.
    On the other hand I once tested a sigma 12-24mm lens and a Nikon 12-24DX lens.
    I could see a clear 1/2 stop differene between the lenses.
    My AFD 50mm f1.4 did not give me the impression of a general differece in exposure to my other Nikon lenses. But this was on a D200 and a D3.
  8. Will try. Should suffice to shoot a white wall with different lenses to achieve an identical look, shouldn't it?
  9. I notice that I usually need to dial in -1.3 EV of compensation on my D40 when using my 35mm f/1.8 DX as opposed to the -0.7 EV I usually have with my kit lens and 55-200 VR. Not sure why.
  10. I don't really understand why, but I have the same problem with one of my fairly new Nikon lenses that I bought for my D300. Most of my other Nikon lenses and those of some of the less pricey knock offs work with the camera to help perform automated exposure just fine.
    It just doesn't make sense that one should have to shoot in manual mode when you have spent this much money on cameras and lenses that are intended to do almost magical high tech TTL metering.
  11. I find my D700 overexposes compared to my old D200, by around half a stop. I've dialed in automatic compensation for that, and I've been meaning to test more fully.
  12. The odd part that I don't understand, considering modern TTL metering, is that this phenomenon is very lens specific. Most Nikon lenses work just fine, but one in paticular almost always overexposes.
    And the word "almost" is the big problem. If it were consistent you could just get used to always dialing in a little correction.
  13. You're lucky it's only slight overexposure. I got so fed up with my D700's matrix metering constantly blowing highlights that I returned it (twice) under warranty. Nikon kept my camera for a total of 7 weeks before claiming that it had been thoroughly tested and met Nikon's specifications for exposure metering.
    I now have minus 2/3rds of a stop permanently dialled into the matrix metering "fine tuning" menu, and still have to wind the exposure down even more on ocassions.
    Nikon, I'm telling you publicly, both your warranty service and your 3D colour matrix metering stinks, and both need a severe rethink. Try consulting a few PHOTOGRAPHERS about your metering algorithms, instead of putting a weekend snapshotter in charge of developing your exposure systems!
  14. One thing I have noticed in mine is how often it has proper exposure. My D200 often underexposed, so it almost seemed at first that my pictures were overexposed until I downloaded them for the first time.
  15. Read this:
    Note Takeuchi San's profile.
    Then read how much apparent effort has gone into developing a metering system that actually performs worse than the one fitted to my ancient F801s. It's a pity that complexity doesn't necessarily result in improvement, which is an insight that many companies seem to lack these days.
  16. My D700's matrix metering works just fine, for me. Now, flash exposure is a different, but contollable story.
  17. I've just recently bought a D700 and had the exact problems you described.
    After a bit of investigation I found that when in Aperture Priority Mode the sub-command dial (thumb wheel which usually does shutter speed) becomes an Easy Exposure Compensation Dial. It allows you to dial in compensation without holding down the +/- button.
    You don't get any obvious warning that this is on apart from the meter indicator bar is off to one side (+ or -). Not even the exposure compensation value on the LCD changes, that only changes if you dial in the compensation via the 2 button method (the old way).
    Turns out I had in-advertantly made an exp comp adjustment without realising and shot away happily..... well until I noticed there was something wrong.
    You can turn this feature off I think or set it to RESET which resets the compensation to '0' when the meter turns off.
    Good luck.
  18. Hi Monika,
    Your story sounds familiar. When I first started shooting with my D700, I was convinced that it was overexposing all the time. But it really wasn't.
    The D700's sensor can capture a lot more detail in highlights and shadows than any other camera that I've used, especially since I'm used to shooting slide film. That means that if you shoot in RAW mode, in post-processing you can lighten up shadowy areas and recover lots of detail. You can recover detail from those apparently overexposed highlights, too. So, while the images might look overexposed, they're not really overexposed. You can adjust the relative exposure level on your computer later if you want the images to look darker and more saturated. But if you employ exposure compensation on the camera itself, you're throwing away detail that the D700's sensor can capture, detail that you might wish you had later.

    Do you have Lightroom 2 or Capture NX2? Shoot some RAW (NEF) format images and then experiment with shadow and highlight recovery in these programs. It's amazing what the D700 can let you capture in a single exposure.
  19. Another factor to consider when comparing exposure on different lenses is falloff; are you comparing the center of the frame or the overall appearance? Metering is done at full aperture, where falloff is usually strongest, but the falloff often diminishes as the lens is stopped down and for zooms may also change at different focal lengths. If you're using the spotmeter, it is likely metering where there is minimal falloff but center-weighted or matrix metering will be experiencing more of that falloff which may affect exposure to some extent. If you consider all the variables with interchangeable lenses, it's not surprising that exposure can vary lens-to-lens, TTL metering or not.
  20. you know you can calibrate those meters.
  21. What Dan says about the dynamic range of the D700 is true, and it's one of the reasons that I still use a D700 despite the awful metering. However, it doesn't alter the fact that blown and compressed highlights are a common occurrence and widely reported with this camera. Recovery in RAW gets you about another stop of headroom, but sometimes the highlights simply become too badly posterised. I wouldn't mind if it was the odd little specular highlight that got blown, but often it's an entire cloudscape, pale blue sky or side of a building that goes "on the blink" and disappears off the righthand edge of the histogram. The variation in exposure from lens to lens is annoying too. I never got any such exposure problems with an Eos 5D.
    Another thing to ponder. If Nikon have such faith in the D700's matrix metering, how come all the sample pictures on their website are taken in manual mode? You'd think they'd want to show just how good their overhyped metering system was, wouldn't you?
    Anyway, here's an illustration of the dynamic range you can get from a RAW file. The top picture is the standard JPEG from the camera and the bottom is manipulated in ACR from the raw file. NOTE: This is not an HDR combination, just an alteration of the tone curve. I estimate that there's about 11 stops of subject brightness contained in this image, however the exposure was wound down by about 4 stops from what the matrix guesstimeter suggested.
  22. Joseph Braun, are you saying you can calibrate the exposure sensors without sending the camera to Nikon? If so, how?
  23. A couple of things spring into my mind. When considering constant overexposure, and with an older lens, never exclude the possibility of sticking apeture blades. With the high working ISO's of the D700 many are working at small apetures of f11 and more, and all it takes is slightly sticking/delayed apeture blades to give you overexposure at those apetures. Check the 'snappyness' of the blades very carefully.
    Another thing I can't help wondering is, how is everyone judging the exposure ? By the image in the pre-view screen, by prints, or by the cheap over bright LCD monitors that most people seem to have nowadays that i would not trust to judge image brightness on. I only trust the exposure viewed on my old Trinitron crt. On the cheapos everything looks overexposed. Consider these things before jumping to conclussions about the D700's metering system.
  24. Three additional comments about metering and the D700.
    (1) Some situations just have too much difference between light and dark for the camera to capture. In those cases, it needs to over-expose and/or under-expose some portion of the image unless you use filters to tone down the highlights. In these situations, try spot metering on the most important part of the scene and let the rest fall where it may. Or if the subject isn't moving, bracket and use HDR to make a composite image. Or use a graduated ND filter or (in some cases) a polarizer.
    (2) Nikon's Matrix metering pattern takes the level of brightness at the active autofocus point into consideration. It assumes that the active AF point is on something important and tries to place that point toward the middle of the range. If you focus on something in the shadows, you may over-expose the highlights and vice versa.
    (3) NEF (RAW) format has more latitude than JPEG files. When you convert a high contrast image to JPEG format you'll lose some highlights and/or shadows. You might need to do some highlight and/or shadow recovery before converting the image into a JPEG file. For instance, some files that display fine as NEFs on my computer look totally washed out if I convert them to JPEG format and post them on the web. Compress before you convert.
  25. "Should suffice to shoot a white wall with different lenses to achieve an identical look, shouldn't it?"
    It should, but in my experience, don`t. In many lenses, like the 50AFS (many other lenses too, I remember this issue at least on faster primes) there is a noticeable diference from wide open to say, two stops closer. Looks like at wider apertures for whatever the reason the resulting exposure is different than from f2.8 on.
    Also, there are noticeable color differences between lenses. This could help to have different perceptions of the same scene, I think.
    Guillaume advice is what works to me. I like to shoot on AE mode or manual... that "easy exposure compensation" feature" is one of the wonders of this cameras. I like the check the histogram to see how highlight are rendered (the "highlight" screen is also great but I need the histogram to know how much I could be off) and then I compensate via command dial, looking the stops inside the viewfinder.

    Although I need to compensate almost every shot I do, I`d not say the metering system is that bad... for most casual shooting situations, works for me. Anyway, shooting RAW is currently a must to me.
  26. Thank you all.
    Guillaume, your hint is of real value to me. So far, I've customized this camera to my taste, but after only a short time, some things are still not something I know "while sleeping", as we Germans say. I'll check this immediately - I had something similar happening when I had just bought the camera: being used to Canon, I apparently had fiddled with one of the dials inadvertently, and on some pics, you can see the shutter moving through the scene. Yuck!
    Dan, your point 2 is something that I must consider, too. For the workflow, I always, *always* shoot in raw, and then postprocess to my taste (for such reasons as recovering seemingly blown out lights) and I've often noticed that, while something might look bland in ACR, after adapting curves and such in PS, it's a miracle.
    José, the difference in lens brightness and colour is something that stuns me, and that I've often noticed, but not only with Canon.
    I know that it's not a real big issue problem, and I can certainly live with it. It may also be the totality of factors: moving to a different camera system, moving to full frame, AND moving to a different measuring system.... but I'm certain that me and her (yes, it's a male, this camera...) may be damn good friends in the end... :)))
  27. I notice that nearly all of the workarounds suggested above require time and test exposures. If you've all the time in the world to take the shot, then using spot metering, test exposures, a handheld meter or fiddling with exposure compensation is absolutely fine and dandy. But surely the main (or only) reason for wanting to use matrix metering is that it should get it right first time, most of the time, with subjects that don't hang around to be carefully metered. If it doesn't do that then it's a pretty useless feature.
    I've been a Nikon user for over 30 years, but brand loyalty can only stretch so far. It seems to me like a lot of fellow D700 users are simply making excuses and workarounds on behalf of Nikon. Tell you what, if I'm ever shooting sport or action it'll be the reliable old 5D that's hanging round my neck and not the flakey D700.
    PS. FYI all my lenses have a snappy aperture action. Do you think I didn't try that? And I get overexposure with a brand new 14-24 AFS ZoomNikkor as well, and in fairly low contrast lighting. I'm also not in the habit of accidentally poking the buttons on the front of the camera while twiddling a thumbwheel, or setting menu options that I promptly forget about! So have you Nikon apologists got any further suggestions as to why it's not the camera's fault that it can't be relied on as a "point'n'shoot"?
  28. I empathize with your pain, Joe. Can you post an example of one of your "overexposed" photos? Is the camera overexposing near the point of focus, or is overexposure happening on the edges of the image?
    If you're shooting fast-moving objects, can you preset the exposure by metering the scene before the object enters it? I don't have to meter a football player if I can meter the field before he runs for a touchdown (provided the sun doesn't go behind a cloud in the meantime.)
    I have yet to see a "perfect" matrix metering system, but I'm amazed at how well they can meter a scene in about a hundredth of a second. In tricky situations, we still may have to give the meter a helping hand.
  29. I also bought D700 and spend some time with it. I have noticed that the exposure is always 2/3 stop more than D200 with the same lense and using spot metering. I tried both camera using 85mm F2.8 wide open. So there wouldn't be any problem with sticky blades. A bit disappointing because shots seems to carry a bit washed out shadow. with -0.7 stop compensation dialed in, the pictures seems to be 'normal' or what I am used to seeing without washed out faces and shadows. Will need to ask around a bit more before I take the camera for any adjustment or repair.
    - Young -
  30. I fully agree with Joe
    I think nikon meters should always match the classical 18% grey, no matters how good the D700 sensor is at retrieving info in highlights.
    18% grey perfect matching of a spot meter is something you really need when you have high contrast and you still need to expose for a critical subject.
    Moreover, how do you explain this overexposure to students and people who want to avoid postprocessing ???
    Consistent overexposure with a body is illegitime in my opinion.
    Just like most of you, I tested my D700, and meter clearly overexposes by 0,7 EV, what I compensates with the personalized setting (hopefully this setting exists!!!)
    Kind regards,
  31. Interesting reading all the threads. I've had my D700 for around a year and using Nikor AF lenses. Some images are fully overexposed and others are okay. Also, I can shoot the same image twice, immediately after each other, with centre weighting, camera on a tripod and get one image overexposed and the other correctly exposed? I shoot in NEF. It does not always correct on post production. The lenses are working fine as use them on a film based Nikon F80.
    I am very disappointed with the overexposure and the lack of sharpness that I obtain with the D700.

Share This Page