This post is about a luxury problem. Being an owner of a Nikon D70, a D700, and a D7000 (logarithmic ownership), together with the golden lens trio (14-24, 24-70 and 70-200) and some beautiful primes (50, 60, 85), there is still a wish for more… Smaller and lighter. Especially on family trips, on holidays, and photographic tours where ultimate image quality is not of ultimate importance.
I tried to work this out in with a P&S, first a Fuji Finepix F40d, too noisy), and then a Canon S90 (quite OK). In bright sunlight both do a good job, but they become noisy in common environment, like in-house. So these cameras are perfectly capable for bringing home some memories, but it is hard for me to use them for creative photography (which probably tells more about myself). Framing a picture by means of an LCD doesn’t seem to work for me. In most cases overall disappointment is the result when I come back shooting with these P&S cameras.
So, with the recent announcement of the Nikon V1, I thought it was worthwhile to have a look on the web and a travel along the local photography shop to look at various system cameras, and to get an idea whether the V1 is good value for money.
The 3 cameras that appealed most to me were:
Nikon V1 – It looks cute, and has a marvelous viewfinder! The grip is a bit slippery though, it has not build-in flash, and it is not easy to switch modes without going into the menu. From usage point of view (small + viewfinder) this has my preference.
Olympus PEN-EP-3 – Good grip, good photo quality. No viewfinder, and the user interface is reported to be cumbersome. It can be extended with a viewfinder, so if its photographic qualities are better, it may make more sense to go for an Olympus than a V1.
Panasonic GF3 – Same situation as the Oly3, however a cheaper feel (also price!). The GX1 is probably a good alternative, but hardly available. The G3, which has a viewfinder, is too big in my opinion.
There is of course other interesting stuff on the market. Take the Fuji X100, great retro-look, good quality APS-C sensor, and even a view finder, but it is big, pretty heavy, and limited to 24mm only. The Fuji X10 seems to address some of this, but it has a very small sensor, so it comes close to the Canon S90 in terms of quality, and its view finder is reported to be average. The Sony NEX3/5 will offer the best quality sensor at a competitive price, but the lenses are too big for my purpose, and they have no view finder. The Sony NEX7 has a view finder, but is rather expensive, and again lenses are big. In that case, the Nikon D7000 is a better alternative with the body being a better counter weight to the lenses. The same holds for the Samsung NX200, the APS-C sensor requires relative big lenses.
As you can see, the V1 is the highest, the EP3 the widest, the GF3 the smallest (see camera size). The Nikon J1 (which is a V1 without view finder) would be a more fair comparison, and would be the smallest of the three.
With the new 14-42 standard zoom the Panasonic shows a size advantage. Hence, the small CX sensor of Nikon hasn’t brought a significant advantage in lens size yet.
A comparison between the J1 and a smaller Olympus EP-3L shows that the Olympus 14-42 standard zoom is a bit bigger:
Overall I would conclude that the size differences are there, but not significant. In any case, the above cameras typically don’t fit your pocket, and need a small bag. For cameras that can be carried in your pocket, you’ll need a P&S. For instance, the Canon S90 including lens equals the size of the V1 body only.
There is definitely a size advantage compared to an SLR, see for instance the comparison to a Nikon D7000:
The Olympus EP3 has a good firm grip. The Panasonic GF3 feels small and cheap. The Nikon has no good grip on its front. The price for a dedicated Nikon grip on the V1 is ridiculous (> $/€100 for a piece of plastic/rubber with a screw that should be included standard for this price setting anyway…). A glue-able alternative is available at flipbac, or at kleptography (for the J1).
Most of these cameras are made with P&S in mind, and hence should be easy to use for everyone out-of-the-box. This is certainly the case for the V1, you set the language, date and time, and off you go! Nikon restricted control of the camera by means of buttons or wheels, e.g. there is no PASM dial on the V1, you have to enter the menu for those. Manual focus also requires a lot of clicking.
The menu structure of the EP3 seems to be complex and cumbersome. When looking at EP3 pictures on the web, all of them seem to have a magenta cast which requires manual interference to white balance them correctly.
The GF3 has almost no dials and buttons, and will be mostly menu driven. The new GX1 does have more dials and buttons.
The Nikon is uses a 10 Mpix CX-sized sensor (116mm^2, pixel pitch 3.4um) from Aptina. The Olympus and Panasonic use a 12 Mpix sensor (225mm^2, pixel pitch 4.2um). Although the 4/3 sensor is almost twice as big as the CX sensor, the pixel pitch is only 33% higher.
DXOmark reports the intrinsic sensor performance, and excludes differences due to in-camera processing. The GF3 is not reported, but because the GF2 contains the same sensor I took those figures. As you can see that despite the smaller sensor in the V1, the sensor quality is the same when appropriate processing is applied. So, with the right processing the V1 can achieve the same or better results as the micro-4/3s. Don’t let yourself being fooled by the tests on the web that show blurry or noisy pictures on the web. You’re looking at differences in (in-camera or RAW) processing, and not at the intrinsic possibilities of the sensor!
Some graphs are shown below (see DXOmark for a flash version – did I say iPad?). Basically these are all overlapping, besides the dynamic range of the V1 being somewhat better. Quite surprising to get these results out of such a small sensor, and definitely a good result.
If a system camera is going to be your only camera, and sensor quality is of big concern for tough situations, the Sony NEX has a clear advantage (77/23.6/12.7/1079). In my specific case, having a Nikon D700 (80/23.5/12.2/2303) or a D7000 (80/23.5/13.9/1167) at hand makes absolute sensor quality less important, there is always a backup. Besides, it is not only the sensor that matters, a system is as weak as its weakest component, so the lenses are as important as well. And this is a situation where Sony is not so very strong as can be seen in the next section…
In line with the size comparison, we see that the Canon S90 – Nikon V1 – Nikon D7000 have one stop advantage noise wise. The main advantage of the V1 is the Dynamic Range that is preserved better in higher ISO levels.
Last but not least, the SNR graph to compare the V1 to the D700 and D7000. As you can see, it is about 1 stop away from the D7000, and 3 stops away from the D700 in terms of noise. When compared to a D70 or D200, it is comparable or a maximum of 1 stop off. That’s what I would call innovation, almost SLR performance of 7 years ago in a small body!
No matter how good a sensor is, a bad lens will never provide sufficient sharpness, contrast and texture. Look at for instance the Sony 18-55mm (effectively 27-83.5mm) and the Olympus 14-42 (effectively 28-84mm) measured by slrgear. Although the Sony has a better APS-C sized sensor, the lens of the Olympus outperforms the lens of the Sony. Although in general more noisy, the Olympus pictures are more crisp and appealing in real life situations.
Also the Panasonic lenses show good results, though in general less good than Olympus. Some excellent lenses in the 4/3 camp are the Olympus 12mm/2.0 (€749, very expensive), the Olympus 45mm/1.8 (€299) (look how incredible it is here), and the Panasonic 20mm/1.7 (€349)
There are no slrgear tests of the Nikon lenses reported yet. There are some measurements reported on photozone and dxomark, but none of them can be compared to the 4/3rd lens measurements. There are also some private tests on dpreview. Below are the measurements of the Nikon 10mm/2.8 and the Nikon 10-30mm/3.5-5.6 of photo zone (NEW: Nikon 30-110mm, and a nice comparison of these lenses at Mansurovs):
On the web cameras are rated based on comparing images of a given object. Although these tests are valid on themselves, absolute ratings based on these tests can give a wrong view of the possibilities of a camera. It depends largely on the settings inside the camera, and how you utilize the camera features.
As an example, I’d like to show the comparison images shown on dpreview and imaging-resource. On dpreview you can see that the point of sharpness is different from all other cameras. In some parts of the scene the V1 is clearly less sharp, in other areas is looks sharper than all other cameras. You can see confusion occurring in blogs, where people refer to just a part of the scene, and make absolute statements about sharpness. Depending on the selected part, the outcome favors one camera above the other, giving an inconsistent view. Furthermore, for the Olympus a prime lens is used, for the V1 a zoom lens. Creating such differences in a test like this make it difficult to draw general discussion.
Even more extreme is the comparison on imaging-resource. The Nikon V1 picture @ISO1600 in the comparison looked so bad, that I considered viewfinder and handling of the V1 not enough good reasons to prefer it above an Olympus EP3. However, when investigating the root cause for the unsharp image (default noise reduction of V1 very aggressive, point of sharpness different e.g. look at the watch), and the difference in brightness (Olympus uses more contrast, the setup of the doll and light is different for both pictures) the outcome became different:
RAW after a bit of post processing (noise reduction, contrast & brightness) for both:
Essentially, it shows that intrinsically both cameras are equal in terms of reachable image quality.
All of this explains why I prefer the tests of DxOmark, as it measures the intrinsic sensor performance excluding internal camera processing, and gives a better idea of what you can accomplish by applying some post processing.
Last but not least, I did some comparisons with my own cameras. The results are default processed RAWs out-of-the camera, all at ISO1600, taken on a tripod. The full picture is shown below:
The excerpts are
As you can see, if out-of-the-camera noise is your main concern, buy a Nikon D700.
The V1 noise is easy to reduce using post processing (pixel peeping):
Real-life experiences and Pictures
My own stuff (all ISO3200, 1/10s, applied little noise reduction and 70s-style):
Take a look at the following pages to form your own opinion (M = measurements, C = comparison pictures, E = example pictures)
– DXOmark (M)
– Photoreview Australia (M)
– Techriot (M, E)
– ColorFoto (C, M)
– Focus Numerique (M)
– Techradar (M, E)
– Steve Huff (E, C)
– Mansurovs (E, C)
– DPreview (E, C)
– Camera Labs (C, E)
– PhotographyBlog (C, E)
– Fotomagazin Test Images (C)
– FotoVideo (C)
– Les Numeriques (C)
– OpenBloom (E)
– Visual Science Lab (E)
– CNET (E)
– Pocketlint (E)
– Rob Galbraith (E)
My opinion is that the Nikon V1 is capable of making good quality pictures. If you want ultimate sharpness, a lot of resolution, or noise-free pictures out-of-the camera, stick to a full-framed SLR with some good prime lenses. If you want a camera that you can easily take anywhere you want, buy a Nikon V1 (or another brand), and have fun! I see this opinion is shared by those who take the camera out of the lab, and in to the field, they are happily surprised about its performance.
+ Very nice viewfinder.
+ Easy operation, nice handling.
+ Good quality pictures.
– Bad grip.
– High price.
– Very high prices for accessories.
The Olympus has a small resolution advantage, and currently better lenses in its portfolio. It is to be seen what Nikon (and others) will bring in the near future. I however prefer the viewfinder of the V1, and assume Nikon (and others) will surprise us with nice lenses in the near future.