Jump to content
Sign in to follow this  

Hands On with the Nikon D4S in Norway

Andrea B.

How cool it is to walk into Nikon Nordic in Oslo with Bjørn and have handed over a D4S as an NPS loaner to test and play with for two weeks! But I immediately wondered, as a current avid, dedicated D3S shooter, whether I would find the D4S compelling enough to invest $6500 US dollars in the upgrade. That's about 39000 Norwegian kroner - before VAT. Whew!! Pretty steep in any currency.

I'll tell you right from the start: After a 2 week trial, I am impressed. The D4S rocks -- it is versatile, solid, reliable, customizable and has gorgeous image quality. The improvements over the D3S are useful. I want the camera. Whether I will actually buy it will take some further consideration. A D3S to D4 upgrade? No. A D4 to D4S upgrade? No. But a D3S to D4S upgrade is looking worthwhile.

All arguments about size and weight have already been made a gazillion times over. So we aren't going to get into that. I'm personally very comfortable with the big D cams and had no serious ergonomic problems with the D4S other than learning the new button placements. After a couple of days some new kinesthetic neuronal tracks were laid down in the old brainpan and I was happily using all the buttons and dials.

Only one button gave me some grief with a particular setting, the Sub-selector, formerly known as the AE-L button. The Sub-selector sticks out more than the old AE-L button did. That's good for distinguishing it from the AF-ON button, so I approve. But when I programmed the Sub-selector to use for selecting an AF point (via side-to-side or up-down pushing), I then kept bumping it and knocking my AF point awry. So I went back to the familiar Multi-selector method of moving an AF point.

I didn't get to fully test the various settings for capturing motion in such a way as to present examples. (So little time with the D4S - so much of beautiful Norway to capture!!!) A few of us had great fun performing rat-a-tat 11-frames-per-second burst shooting while Bjørn rolled his eyes. laugh.png But make no mistake -- these 11 fps are not for "spray & pray". If you are seriously shooting motion, you want this for use with the dynamic tracking capabilities of the D4S: Group AF, tracking with 9/11/21/51 point clusters and 3D tracking.

Applied to a moving object, the 11fps seemed to work as advertised with auto-focus and auto-exposure keeping up. I shot a thread spool rolling across the floor several times using the new Group AF (a cluster of 5 AF points). When *I* was quick enough to lock the focus on such a small object, the camera certainly performed its role well. Yes, there were some in each sequence which were OOF, but not many. And it is hard to say whether it was me or the camera who did not keep up given that it is not so easy to keep the Group AF points on a small, fast-rolling thread spool. It would be much better to test the D4S on moving cars or skate boarders or a bouncing pupdog. If I ever get my hands on a D4S again, I will thoroughly test shooting motion sequences (I love shooting motion!!) and come back and post the test results here.

Side Note: 3D tracking is useful for focus and reframe of a still subject. The focus point remains on the subject after reframing.


D4S: Compare RAW-S vs RAW-L Files

The D4S offers both a large 3280x4928px, 16Mp/14-bit, RAW-L file and a small 1640x2464px, 4Mp/12-bit, RAW-S file.

Seems like Nikon hid a D2HS inside the D4S. biggrin.png

Rawdigger has written an in-depth analysis of a RAW-S file here: http://www.rawdigger...l-raw-internals

This is well worth reading to learn how to turn 16 Mp into 4 Mp.

The practicalities of why one would choose to shoot a small NEF instead of a large NEF escape me just now. Seems to me like I would always choose to shoot the RAW-L size and downsize if and when needed? Wouldn't you? But there must be someone out there who needs RAW-S files. Oh well.

So, anyway, my goal in this test was simply to compare an enlarged RAW-S file to a corresponding RAW-L file and to note any differences. As Bjørn put it, how much would you lose in image quality should you happen to "accidently" shoot the smaller RAW-S file for an intended large print which would normally be made from the full sized RAW-L file? Answer: Surprisingly little.

For a proper comparison, I first PS enlarged the below RAW-S photo 100% (that's double the width/length) to match the RAW-L size. You will see that fine detail is slightly softened in the RAW-S enlargement crop but that no detail is completely lost.

Test Subject: Dragon Gong

I wanted to shoot something real and not newsprint taped to the wall.


RAW-S File: 400x400px 100% Crop from 1640x2464px RAW-S Dragon Gong Photo

f/16 for 1/1.3" @ ISO200

No sharpening.


RAW-L File: 800x800px 100% Crop from 3280x4928px RAW-L Dragon Gong Photo

f/16 for 1/1.3" @ ISO200

No sharpening.


RAW-S File Enlarged: 800x800px 100% Crop from Enlarged RAW-S Dragon Gong Photo

No sharpening in this crop.

But I did play with some low-radius USM and also with some .5 - 1.5 High Pass Overlay. Either method gave a small bit of contrast enhancement to the enlarged photo that was pleasing when viewed an arm's length from the monitor. Of course, choice of post-enlargement sharpening or local contrast enhancements can generate passionate discussion for many pages. So let's not get into that here.


The D4S as a Landscape Cam

Yes, having 16 megapixels is indeed better than having 12, in my opinion. Doesn't seem like much of a leap, but I like the extra detail it provides. YMMV, of course.

The D4S colours are superb - they seem nicely nuanced, but not much different from the D3S colours - that's my subjective opinion again. One might argue with the methodology of DxOMark's measurements of colour, but the results do serve for making relative comparisons - the D3S and D4S are close in colour capability.

I do definitely note that the D4S increased dynamic range better handles highlights over the D3S. I'd already seen this with the Df and D600, so I knew to expect some of that with the D4S also. It may only be a 1 - 1.5 stop improvement, but you will see it.

For the following Harbor photo I felt I needed higher ISO because I was balancing the cam on a bridge railing and not on tripod. Of course, everything held up very well as ISO 800 scarcely stresses the image quality of a D4S file. I admit to bluing up the sky just a tad in NX2. But it's my memory photo, so I'm OK with doing that this time. Other than that, I applied a bit of gentle contrast brightening here and there with U-points. (I did not correct for the 24 wide.)

After looking at the 100% crop, maybe you can see why I'm eager to move to 16Mp over the D3S's 12Mp.

Harbor at Justøy

Justøy, Norway

D4S + 24-70/2.8G @ 24mm

f/11 for 1/80" @ ISO 800

Standard [4,0,0,1,0]

Converted/edited in Capture NX2.


100% crop from preceding photo

229 x 352 pixels



The D4S as a Sepia Landscape Cam

Fellow Fotozones member Sten Rasmussen guided Bjørn and me to the Utstein Monastery, a still-standing mediaeval monastery. It was in active use while we were there for some confirmation services. We had a good time talking and shooting on the Monastery grounds. And we got to check out Sten's nifty Fuji X, too.

For this study of gate, stone fence and trees using sepia tones seemed the way to go because the spring lawn was an intense lime green which - although really lovely - did detract from the idea I wanted to capture. I "cloned out" a couple of fence poles beyond the stone fence. The Photo Ninja detail slider seemed to provide just the right enhancement here.

Gate in Stone Wall

Utstein Monastery, Møsteroy, Norway

D4S + 70-300/4.5-5.6G @ 200mm

f/11 for 1/50" @ ISO 800

Monochrome[4,1,0] Sepia[3]. But I altered that slightly when editing.

Converted in Capture NX2. Edited in both Photo Ninja and NX2.



The D4S as an Artcam?

Well, why not. The D4S is nothing if not versatile. This photo in its final form is intended to have some of that 'graphics' look. With 16 well-designed megapixels and Expeed 4 processing, you can later push on a photo in your chosen editor for (*koff-koff*) "artistic" effect and still maintain some clarity and detail to keep the resultant art(?) "photo-like".

Mirrored Landscape #1

Mirror Art Installation, Holandsdal, Norway

D4S + 24-70/2.8G @ 70mm

f/16 for 1/200" @ ISO 100



D4S Multiple Exposure

At the Altmark Incident Memorial both British and Norwegian naval pennants were flying. Typically a pennent is flown on commissioned naval warships. Sometimes homeowners fly a pennant to indicate that they are present. Flags and pennants waving in the wind are a natural subject for multiple exposures.

Norwegian Pennant #2

Altmark Incident Memorial, Jøssingfjord, Sokndal, Norway

D4S + 70-300/4.5-5.6G @ 195mm

9-frame Multiple Exposure

f/13 for 1/1000" @ ISO 400



D4S Trap Focus

It works.

  • Like 1
Sign in to follow this  


Recommended Comments

D4S and Df:  Comparison at high ISOs

Df Facts:  The calibrated ISO range for the Df is 100 - 12800.
Extended Df ISO ranges are:  Lo 1/.7/.3 and Hi .3/.7/1/2/3/4
where Lo 1 = 50, Hi 1 = 25600 and  Hi 4 = 204800.

D4S Facts:  The calibrated ISO range for the D4S is 100 - 25600.
Extended Df ISO ranges are:  Lo 1/.7/.3 and Hi .3/.7/1/2/3/4.
where Lo 1 = 50 and Hi 4 = 409600.

We set up a CC Passport for me to photograph in low ambient light using the 24-70 f/2.8G. The cameras were locked down on one of Bjørn's Sachtler tripods. Mirror-up and remote shutter firing were used.
Kindly note that no noise reduction was used either in-camera or in-editor for the photos and 100% crops shown below.

Under these low-light conditions, the Df produced a fairly large leap in noise between ISO 25600 and 51200, but I didn't find any banding until 102400 where it was very faint. At 204800 the Df banding was obvious.
The D4S produced very faint banding at ISO 51200. Noticeable D4S banding appeared a stop later at ISO 102400 and worsened at 204800 and 409600.

Of course, with judicious use of a good noise reduction app, you can probably make reasonably good use of low-light photos from both cameras when shooting up to ISO 102400. The D4S low-light photo at the highest 409600 is simply unusable, in my opinion, except for possible "spy photo" purposes.


[Note: I will post the Df series in this post with the corresponding D4S series in the next post.]

Here is a Df scene overview. You can see that the CC Passport was in the darker central area of the scene. From this scene a center 100% crop was made which includes both the Passport and some of the shadowed area behind it.

Df High ISO Series:  Full Scene resized to fit 1000px width
ISO 6400



Df High ISO Series: 100% Center Crops

ISO 6400


ISO 12800


H.3 = ISO 16000


H.7 = ISO 20480


H1 = ISO 25600


H2 = ISO 51200


H3 = ISO 102400


H4 = ISO 204800


Df High ISO Series:  More Full Scenes resized to fit 1000px width
ISO 25600. Things still look pretty good after resizing. Remember no noise reduction was used.


ISO 51200. The first big jump in noise is seen, although less apparent in the resized version.


ISO 102400. Banding is apparent now in the full-sized NEF, but you might be able to live with a resized version.


ISO 204800. Perhaps not so useable at any size?



Share this comment

Link to comment
Share on other sites

As before the first D4S overview shows that the central area is one of the darker areas of the scene. Again, a series of centered 100% crops were made.

D4S High ISO Series:  Full Scene resized to fit 1000px width
ISO 6400



D4S High ISO Series: 100% Center Crops
ISO 6400


ISO 12800


ISO 25600


H.3 = ISO 32000


H.7 = ISO 40960



H1 = ISO 51200


H2 = ISO 102400


H3 = ISO 204800


H4 = ISO 409600


D4S High ISO Series:  More Full Scenes resized to fit 1000px width
ISO 51200. This looks useable.


ISO 102400. Now the first big jump in noise is seen - 1 stop later than happens in the Df.

There is faint banding.


ISO 204800. More banding.



ISO 409600. Yikes!! I don't think so.







  • Like 1

Share this comment

Link to comment
Share on other sites
Guest nfoto


Good compilation, Andrea :D

Share this comment

Link to comment
Share on other sites
Guest PedroS


Thanks for sharing.

Good to have one already !!

Share this comment

Link to comment
Share on other sites
Guest Colin-M


Thanks for taking the time to do this Andrea.


As someone who's only ever used the smaller models, I had a chance to play with a D4 a while back and loved the way it handled.

Share this comment

Link to comment
Share on other sites

The additional weight does not faze me in the slightest – .

He he Ann, if you were a hunter you would be shooting a 500 Nitro Express :-)

Share this comment

Link to comment
Share on other sites

Andrea: Thanks for sharing your impressions. The D4s seems to be as expected. It would be nice to own one, but I hope to enjoy the patience to wait for the D5.

Share this comment

Link to comment
Share on other sites

I'm happy this small review was useful. It was such a treat to use this camera for 2 weeks!! Thanks again, Bjørn R.


The D4S's 16 megapixels clearly give more detail in landscape and close-up photos - something which I think is quite nice to have. (And one reason why I'm also loving the look of my Df photos.) Whether this better detail combined with better tracking is enough to pry the old D3S cams out of action photographers' hands remains to be seen.


I would like to rework the above high ISO series to illustrate how good (or not, as the case may be) the photos might look after de-noising. Hope I can find the time soon.

Share this comment

Link to comment
Share on other sites

I think the big D line has advanced enough that you will begin to be drawn to an upgrade

if you try one out and work with it for at least a week or two to overcome any ergonomic changes (easily done)

and to see the image quality it offers.


The D4S image quality is worth a look. I don't think we have good vocabulary (or perhaps training?) for describing IQ, but I find the D4S has a compelling IQ - more "depth", more 3D-ness? I suppose that is primarily due to its having a few more megapixels. IMHO, having 16MP is kind of a sweet spot between landscape (24-36MP) and full-tilt action (12MP). You can go in either direction with 16MP and get good results.


As I said somewhere above, the half-step D3S-to-D4 is a 'no', the half-step D4-to-D4S is a 'no'.

But the whole step D3S-to-D4S begins to make me think about changing.


Certainly if someone gave me this $6500 camera, I would accept it in a heartbeat. :D

I still haven't decided if I can fork over that much of my own money for a D4S

given the fact that I make almost nothing from photography.


I'm thinking about renting a D4S for another week of shooting so that I can test out the tracking.

I enjoy that kind of shooting quite a lot.


EDIT:  Yes, the card slot thing is kinda stupid. I'm hoping someone will make an adapter that permits using an SD card in the 2nd slot.

Share this comment

Link to comment
Share on other sites

the quality control on this model was much better than the D4.. mine was sharp out of the box... selling my D4 for 3400 USD with 100k actuations, dropped four times, main pcb replaced, f-mount flange replaced, af-tuned.. etc.. not too bad for a refurb (w/200 actuations) costing me 5000 USD.. so i can't complain.. like renting it for 50USD/month

the xqd card was designed for video... but i use the "S" version at 168MB/s often for 11fps shooting... it downloads faster to thecomputer and the pins (or lack of) don't bend..

SD cards are too slow..

Share this comment

Link to comment
Share on other sites

by the way, i have gotten better ISO performance than these examples

Share this comment

Link to comment
Share on other sites
Guest nfoto


by the way, i have gotten better ISO performance than these examples


Scene contrast influences noise and hence "ISO" performance. It's not hard to get much worse noise as it is to get lower.

Share this comment

Link to comment
Share on other sites

Scene contrast influences noise and hence "ISO" performance. It's not hard to get much worse noise as it is to get lower.

I should have said in the "right" light, I have achieved better.

Share this comment

Link to comment
Share on other sites

Bob, thanks for the input. It is always good to hear about field results from D4S users. I think that Bob's excellent bird shots made with both D4 and D4S tell the true tale about the value of moving to 16MP for the D4/D4S cams.


We set up a tough test for the D4S. And remember that I turned ALL noise removal OFF both in-camera and in-editor. It is certainly true that a scene's contents, contrast and ambient lighting can affect noise levels either for the better or for the worse.


It is also cool to hear about the benefits of the XQD card. I'm actually kind of surprised that this type card has not caught on more given the increased dslr/mirrorless video use these days. My quibble on the D4/D4S is with the difference between the two provided slots. Why not make them both the same - XQD, I suppose - and provide adapters which permit use of SD cards for those who want that? Oh well. No big deal.

Share this comment

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Join Our Small Community

    Like what you see on Fotozones? Join up here and make friends with like-minded photography enthusiasts from all across the planet without having to sell your soul to the Facebook monster. We are limiting our membership to no more than 2000 individuals, so if you are seeing this message there is still space available for you to join. We'd love to have you along. :)  



  • Create New...

Important Information

By visiting this website you are agreeing to our Terms of Use, Privacy Policy & Guidelines.