Wednesday, October 17, 2012

Nikon D600 Dynamic Range in the Real World

You saw from our Nikon D600 dynamic range test that the Nikon D600 showed exceptional shadow recovery capabilities.  In this post, we'll show how that capability can be used in the real world, using some of the samples from the Nikon D600 and the Nikkor 24-70 2.8G.




Here are a couple of those shots again:



Here are how those shots looked like straight-out-of-the-camera:


If you're wondering why the SOOC shots looked underexposed, it's because I wanted to protect the highlights.  With digital sensors (or slides), when highlights are blown, they are lost forever.  To avoid clipping the highlights, the shot has to be underexposed to the point where the highlights are not clipped.  See this post discussing this.  As you can see in the final shots, the highlight details were preserved.  The downside of this approach is that the image will have more noise on those areas where the shadows were lifted.  In the case of the D600, however, the shadow recovery is so incredible that there is not a lot of noise even for extreme adjustments.


Here are a couple of other, more extreme examples, comparing the SOOC version and the final result.  The noise in the final versions looks negligible to me.  I did not apply any noise reduction.




You might wonder how the exposure is set to avoid clipping highlights.  Setting the Active D-Lighting to 'Auto' or 'Extra High' is sometimes sufficient.  Check out this example:


Here is the SOOC version.  The camera chose an exposure that avoided clipping the highlights on the glove.

Here is the glove (exposure decreased to show that no detail was lost).

Other times, especially for very contrasty light, the D600 tries to make a guess: it either thinks you're doing a silhouette in which case all highlights will be preserved, or it thinks you're doing a more typical image, in which case it will try to get a correct exposure for the subject (based on where the focal point is) and let the highlights clip.  If the camera does the latter, I just have to rein in the exposure by dialing it down:



(If you'd like to know a more reliable way to confirm that highlights were not clipped, check out this tip.)

The second part of this solution is how to lift the shadows to where they should be.  I'll discuss a few alternatives in the next post.

12 comments:

  1. These photos of your beautiful children in difficult light do a fine job showing what the D600 is capable of with fairly simple post processing. As someone who shot Kodachrome 64 for 20 years, the ability to show so much detail in deep shadows continues to amaze me. For all but extreme backlighting, fill flash is more an option than a necessity. I was going to get an SB-910, but decided to hold off since I can take so many appealing photos with the D600 in low or strong light without flash. This is my first DSLR. I am very impressed with Nikon and Sony's accomplishment.

    ReplyDelete
    Replies
    1. Hello again friend! Thank you very much. You're absolutely right. In fact, there was a time when I wanted my [digital] shots to look like Kodachrome and part of my recipe was to deepen the shadows to the point where there was little detail :)

      These new DSLRs are really amazing - with their extreme total (shadow + highlight) dynamic range, you can keep delicate highlight detail like print film while also simultaneously getting noise-free shadow detail.

      As for fill flash, I completely agree. That's one way the D600 has had a significant impact on my shooting style. I used to need flash for fill light but with the D600's insane shadow recovery, I have not really needed it. In fact when I think about it, I don't recall having used flash as fill yet for the D600 (!). :O

      Best regards,
      Mic

      Delete
  2. Yeah, that's one thing I am quite afraid, I'lose a bit my mark, I am so much used to expose to the right with the S5, blowing out everything in order to get more details in the shadow, with the D600, I'll need to do the complete opposite if I understand correctly.
    Thanks

    ReplyDelete
    Replies
    1. Yes it's pretty much the opposite of the S5, but the good thing is that the end result is the same as the S5. On top of that it doesn't have any of the S5's weaknesses such as speed, handling and high ISO noise.

      Best regards,
      Mic

      Delete
  3. I don't know about other cameras, but that's what I used to get from my 5D2 & 5D3, even the 60D could get close to this, as long as you're using low ISOs, you could open up the shadows this much without introducing noise.

    I pretty much do this all the time when shooting in harsh sun.

    ReplyDelete
    Replies
    1. Hmm. I did not know that even the 60D could do that. My older cameras (D70, D80, S5) aren't as good with recovering shadows without noise.

      Good to know. Thanks!

      Best regards,
      Mic

      Delete
    2. Well, I went to lightroom and picked up a couple of pictures that express what I want to show, both pictures were taken with the 60D (same story with 550D, 600D & 650D), first one taken at ISO 400, second one taken at ISO 200. All pictures are re-sized to 2500px.

      Sunset (SOOC)
      Sunset (Processed)

      Father & Son (SOOC)
      Father & Son (Processed)

      Delete
    3. Nice examples, Mohammad. Very clean. Thanks!

      Best regards,
      Mic

      Delete
  4. Thank you for this post, I found it helpful. I recently moved from a D90 to the D600. I've only had the D600 for about a week, and while I am impressed with it in most every way over the D90, I became a bit confused and concerned over the weekend when I shot some comparison photos between the two cameras.

    My concern was - using matrix metering, when I set what I thought was the correct exposure for a shot in Manual mode, the same D600 shots seemed to be consistently underexposed when compared to the D90. (for the most part, the shadows seemed darker). Then just for comparison I shot the same image in flash off/auto mode on both cameras to see how the cameras would handle it automatically, and again the D600 looked underexposed compared to the D90 images. For instance, I shot my camaro in my garage under typical incandescent lighting. The shadow on the cement floor under the wheelbase of the car looked about natural on the D90, and I could make out the oil drip spots on the floor, etc. Didn't seem like any post work would be necessary on the SOOC shot. But on the D600, the wheelbase shadow was noticeably darker and the spots on the floor seemed "buried" in the shadows. I did not have time to try on those photos but I am confident I could correct the D600 photos in post, like you did here.

    My question is, am I going to have to do post-work on every SOOC shot from the D600, even when I have what I believe is the correct exposure reading according to the meter?
    Thanks!

    ReplyDelete
    Replies
    1. Hi Brian. There could be many possible explanations for what you are observing. The two main ones are metering and processing.

      Metering: the metering of the D90 and D600 could be different. I haven't tried comparing the D90 and D600's metering side by side to test whether they choose the same exposure. However, I would not be very surprised to find if they chose different exposures. All the DSLRs I've owned have idiosyncrasies with respect to metering. For example, the S5 is ultra conservative and will avoid blowing highlights at all costs. In in my opinion, the D600 appears to be on the aggressive side (choosing a well exposed subject even if the highlights in the background are blown). I haven't given much thought to the D90's exposure but I'm curious to see, so I'll check it out. To eliminate this as a factor, shoot the same exact scene with the exact same exposure (I don't mean 0 the meter - I mean exactly the same ISO, shutter and aperture). Also, if you are finding that the D600 is consistently metering below what you want, you could double-check if the Active D-Lighting settings are the same, and if it still consistently meters below, you could you can fine-tune the metering using option b5.

      2. Processing: the other possible reason is processing. To eliminate this as a factor, I am assuming you are comparing raw versions of the shots on your computer? If you are looking only at the LCD, fyi that's just a jpeg preview (even if you shoot raw). It is possible that the JPEG setting of the D600 is different from that of the D90. If the processing is the issue, and you find that the D600 is darker than how you like it, then you can use Lightroom and apply a preset upon import to automatically apply adjustments that would bring the exposure to how you like it.

      It could be a number of other things. It's really hard to tell without seeing the images... Pls feel free to email me at info AT betterfamilyphotos.com

      Best regards,
      Mic

      Delete
    2. Thank you for the explanation - when I get enough time, I'll do some 1:1 comparisons of D90 vs D600 shots and post them. I'm hoping it's just me and not the camera!

      Delete
    3. Thanks Brian. You can also email them to me instead to info AT betterfamilyphotos.com (raw format, if you like).

      Best regards,
      Mic

      Delete

Thanks for your comment. It will be published as soon as we get a chance to review it, sorry for that, but we get lots of spam with malicious links.