Thursday, 17 January 2013


Understanding the Nikon D600


I promised to fill in a little on what software works and what does not. The differences are much greater than I would have thought.


An example – another jpeg picture that came out rather horrible, just a casual street shot, a little light from a restaurant, looked interesting and I took a picture as we walked by. This is the kind of picture that I would normally just delete but the D600 was new and I was curious…

Nikon D60, 35mm f/2,0 Ai-S shot @ 2,8, A mode, -1,0 Exposure Comp., camera chose 1/30s and ISO 320

This is the same picture pulling the shadows slider by a lot, and tweaking the brightness and contrast only slightly:

The first software I tried was Nikon ViewNX 2, mostly because I use that (together with Transfer) for downloading, viewing and deleting pictures. Anyhow, I wanted to try the different types of software so here is another attempt, this time in DxO Optics Pro 6:

The DxO version may seem ok when seen this small but the details don’t hold up. Zoomed in to 100%, same DxO adjusted picture, it looks rather bad with, not banding but ‘spotting’ of levels – looks almost like a skin disease or a pointillist painting by Paul Signac or Vincent van Gogh. 

View NX 2 handles this much better, especially considering that I was able to lift shadows much more. Levels look almost ok, noise is present but not too bad:

I was baffled how great were the differences – one picture turns out usable, the other not. This had me wondering how other kinds of software would work. So, another picture, same evening, difficult street photo – again, a crappy picture but might be interesting to see what you can do with it.


Same settings as before. This time, the camera chose ISO 160 
Exposure Comp at -1,0 here as well (I usually want very dark night scenes to come out very dark – like night scenes!) but this really came out too dark – Matrix metering was of course fooled a little by the light sources. 

First, jpeg straight out of the camera:

Second, DxO Optics Pro 6. The overall impression may, again, at first glance, seem ok.

Third, Photoshop CS4:

Fourth, Nikon ViewNX 2:


Looking down the street at details, at 100% magnification the very different behaviour of the different softwares becomes apparent.

Original SOOC zoomed in down the almost black street. It was even difficult to crop this picture because I couldn't see the details:

DxO Optics Pro 6 – still, looks ok with lots of more detail visible:

Photoshop CS4 – this legendary software turned out rather disappointing. Even despite limiting how wide the shadows slider tool set in (by pulling down the tonal width and radius controls) it lifted the brightly lit rooms so much that, in order not to ruin the picture, shadows had to be lifted with extreme caution:

Nikon ViewNX 2 – maybe a little less impressive than DxO, perhaps due to a ‘flatter’, less contrasty appearance. Contrast was lifted slightly but not too much in order not to lose detail elsewhere in the image:


This is what happens in the same pictures as above with the finer tones of bigger, dimly lit areas of the wall, and the brighter rooms inside

Original SOOC zoomed in:

DxO Optics Pro 6 – at pixel level the image now falls apart. The fa├žade of the house is again shows artistic ‘pointillist’ levels artefacts:

Photoshop CS4 – this shows how the brightly lit rooms become brighter than desired, despite a very cautious handling of the shadows slider (and its settings):

Nikon ViewNX 2 – impressive performance; shadows much brighter without almost no apparent noise and no disturbing levels artefacts, despite more shadow lifting than any other example – and the room is left more or less intact:


Finally, a highlight area. I noticed that the blue light at the top of the building looked funny from DxO, with very disturbing banding artifacts:

This is the same lamp in the ViewNX 2 picture - look also at the faintest upper parts of the building that are visible here. This is not pretty at 100% but may be passable at smaller print sizes:

So, what software to use?

I cannot answer yet but this is what I see, and I will give you a few more examples. But, most of all, it would be of great interest to check the software I haven’t been able to test. 


The Shadows / Highlights sliders in CS4 are advanced tools which allow you to adjust tonal width and radius – this ought to lead to great control over the result.

Surprisingly, Photoshop has not been impressive and, in particular, despite very careful tweaking of tonal width and radius, it generally tends to produce some very flat parts of the images. 

Furthermore, it seems almost impossible to make the shadows slider work on shadows only; and the highlights slider to work on highlights only.

I will show highlights examples later on.


I didn’t have time before we left for Cuba to take delivery of LR4 but I would have loved to. Apparently, it is a very user-friendly tool, and I have no doubt that it is one of the best for tweaking a raw image.
Still, I have not heard about anybody using Lightroom for jpegs, and cannot even guess how capable it is for this particular use. Since it is produced by Adobe, just like Photoshop, it might perhaps be built around the same engine, or algorithms. Bad news, in such case.
It would be very interesting if anyone would care to try lifting shadows in a D600 or D800 jpeg, using LR4.


The good: DxO software (I use version 6) has been amazing for improving jpegs when I was using my D90, and still is for my Panasonic Lumix LX5. With cameras such as these, it automatically manages to bring out a level of detail and clarity that I believe you cannot really get in other ways. With camera / lens combinations not supported, you can have many of the improvements by manual fiddling (and with DxO, unfortunately, fiddling is the word).

As a raw converter, I believe it is fine but I haven’t compared it to enough others to pass a comment and, besides, raw converters is not really the topic of this blog.

In addition, DxO has a geometry correction module which is absolutely fantastic. For architecture and generally distorted pictures I use this all the time. It is worth the cost of a license alone.

The bad: If I want to use DxO for my D600 I would have to upgrade to the Elite version, which is a bit costly. This makes me hesitate, 1) because the D600 does not really need the kind of jpeg image enhancement DxO usually offers, and 2) because although the camera+lens modules in DxO work extremely well, there aren’t any for the manual focus lenses I prefer.

When trying to use DxO to lift shadows in jpeg there is one very obvious limitation: It does not have a separate shadows slider so you will have to tweak the entire picture, bright and dark, using a tool which seems to be related to the ADL of the jpeg images. Compared to separate shadows and highlights sliders, this is very primitive.

But the biggest flaw with DxO, for this particular use, is how it cannot handle the finer tone levels, creating artifacts, as seen above.


I am very keen to know how this works, considering that it probably has most of the qualities of the free-with-your-camera ViewNX 2, and greater room for control and customizing. This could well prove to be the high-end solution for bringing out the best of your jpegs.


Amazing software, comes with your Nikon camera, version 2 is stable and works fine. I use it for reviewing my pictures, but have come to use it all the time as my preferred software for tweaking my D600 jpegs.

Some of the controls are excellent. Pushing shadows brings a slightly flat image, so it usually needs to be paired with a trifle (+2-3) contrast enhancement and a touch (-2-3) lowered brightness in order to preserve highlights as well as possible.

Other sliders are a little primitive. Pulling down brightness a lot, for example, pulls up the contrast and colour saturation to unusable levels so a shot that comes out too bright cannot be salvaged by this software.

A couple of  more examples


Please excuse me; in order to provide examples for this article, I am using more or less failed photos.

This is another shot, taken in the Camera Obscura in Habana Vieja. In this configuration, a camera obscura is basically a lens projecting the outside scenery via a periscope onto a dished white screen. You see the outside world in real time, it is all optical and mechanical, there are no electronics, and it is very spectacular – there are a handful of camera obscuras in the world.

In order for it to work the room has to be absolutely dark. I took this random shot and, of course, the persons standing on the opposite side were hardly visible in the almost pitch black darkness.

Nikon D600, 35 mm f/1,4 Ai-S @ 2,0, camera chose 1/60 s and ISO 2800. 

First, the straight-out-of-the-camera jpeg:

Second, using Nikon ViewNX 2, pushing shadows while maintaining the dish more or less unchanged. Obviously, there is some noise but, still, with a grain much like an old Ektachrome 200:

This is the same picture using DxO. As you can see, the dish has started to look bleaker. In order not to wash out the dish too much, and in order not to wake too much of the artefacts to life, I could not go further than this:

Which of these three versions (if any!) is better is a matter of taste - I am just showing the possibilities with different software, and the power of Nikon ViewNX 2 in particular.


So how far can you in fact go?

Let’s focus on the software that does work well, the ViewNX 2, and see where its limits are.

In this example, I reacted spontaneously to the autumn afternoon sun shining at the bottles in the background and instinctively tried a shot. Matrix metering was of course fooled by the background light; I should have predicted that but didn't. So I took another shot using raw and fill flash. 

This picture was one I normally would have deleted right away - but again I was curious.

Nikon D600 + Nikkor 28 mm f/2,8 Ai-S shot @ f/2,8, A mode, exposure correction 0 EV, camera chose 1/160 s and ISO 100,  jpeg straight out of camera:

And this is an attempt to salvage the picture – without success! The artefacts that seem to turn up all the time in DxO, with ‘pointillism’ (is there a proper term for it?) and banding, are both there. Obviously, the dynamic range in this motif was beyond the camera's capabilities, especially in jpeg. Black was indeed very black, and I was overdoing the correction.

What’s next?

I am thinking about writing another few posts, depending on reactions.

So far I don’t think I’ve been able to stir up much interest at all. The blog cannot be found via Google even if you enter for example ‘D600 dynamic range jpeg’, which no one would enter anyway…

Therefore, I posted on two forums I myself enjoy reading from time to time – on Fred Miranda, and on DPReview. These posts can in fact be found on Google, now 2 days later. (On FM my account was blocked, without explanation – I must have offended somebody?) 

If you know of  any other forums, please feel free to post something and link to this blog. The blog is not commercial & I am writing just because I happened to stumble on something, and I want to share it with you all.

The roadmap for this blog, depending on the interest, looks like this:

PART III – pulling highlights
PART IV – jpeg vs raw
PART V – other Nikon & Canon cameras
PART VI – in-camera HDR & very high ISO
PART VII – HDR colour; vignetting
PART VIII – retouching in camera

Please let me know your thoughts and reactions – directly on this blog, or on DPReview (


  1. Gabriel, I enjoyed your entire blog and I'm sure many others will too. My only question was in respect of the DX vs FX portrait of the man with glasses where I thought the DX image was far better than the FX - but as with all things artistic it is the author's opinion that matters :)

    Keep it up - I'm sure the yachts weren't successful overnight.


    1. Thank you Peter for your kind words. I tend to agree with you, but the portraits were actually the first exposure from my D600 and one of the last from my D7000, without any attempts other than to stay at more or less the same distance!

      Thanks for your kind words,

  2. very well written! Helps me to make decision on the D600

    1. Thank you,

      I have been amazed by the capacity of the D600 (you probably already figured that out...) and if I can help you decide on making the move, I am just happy.

      Now, at 6540 exposures, I have not encountered anything not to love about it and haven't had any problems whatsoever.

      Hope to find time to continue the blog at some stage but it will not be during summer, I'm afraid.