Greets, Ed.

Another lens I didn't think of yesterday was the Olympus 7-14 4/3 lens. That lens also seems to be rather sharp.

Greets, Ed.

The problem with these test however is that a lens (especially a zoom) can have a very defined sweet spot. So this baby works best at 16 (near the lower end of the zoom range), and it will have some small drawbacks (not as sharp in the edges as a prime) that stem from the zoom design (almost all zooms suffer from corner problems, as do a lot of ultra wide primes as well, I had a Nikon 14 that did not shine in that department).

P.S. the best UW zoom I found was a 7-14 panasonic for micro 4/3.....but that does not get quite as high in detail on the DxO score (only tested on older pana sensors ot on the newest generation and that makes quite a difference).

Greets, Ed.

Personally, the only reason why I'd buy a D800E would be for use with that lens. Of course, if I had a D800E I'd certainly buy more lenses that that, but the OM-D gives me everything I want except for at the longest and shortest focal lengths. The 9-18 and the 100-300 are the only lenses I have that I'm slightly disappointed with. I ought to give the 7-14 a good try to see if it is significantly sharper than the 9-18 (tests seems to indicate that they have rather similar performance) and if the flare issue is manageable. But the 14-24 seems tempting in spite of it's huge price tag and weight.

Greets, Ed.

The real difference comes now from the depth of that pixel well. OM-D saturates at 25.000 photons while D4 does it at 120.000 photons. This means that D4 is almost 5X better instrument at measuring differences in tones. At highlight there is no practical difference because both are good enough, but in shadows when the changes are subtle D4 can see these changes a lot better and further into blackness. This corresponds to higher dynamic range and less noise at high ISOs.
-p-

Great thanks for your thorough explanation on the variety of "parameters" having effect on the final result. I agree that sensor is not itself the thing. This was just wondering about what is really the role of sensor size in this game. I have OM-D and I know that it really fulfills my needs very well - using mostly 12/2 lens.
With Very Best Regards
Juha

I do not take pictures with a sensor but with a camera. That is a sensor and a lens. As a classic photographer a picture to me is composed of three things:

1) Shuttertime (fast enough to freeze motion, even when shooting landscape since tree do move in the wind, but who am I to explain that to a Fin :-) being Dutch and all).

2) Aperture (small enough to give enough DOF unless you are a shallow DOF junkie), but in classic photography we fight for DOF. But also large enough to be not blinded by the dradfull difraction monster.

Added to the party is recently:

4) ISO behaviour (as ISO rises noise increases but also dynamic range drops (in some cases like a stone in vacuum)).

5) When using a nose and two hands as a tripod also IS ads to the fun. Well endowed cameras like the OM-D (5 axis IBIS is "as good as it gets" without calling Helen Hunt and Jack Nicholson) outperform less well endowed cameras like the Nikon or Canon.

6) Sensor size but that cuts both ways (something FF addicts tend to, well, forget). Bigger sensor means better noise performance and less risk on diffraction (bigger sensor wells means bigger airy disks are acceptable). Smaller sensor means more inate DOF therefore (using the same shuttertime), an OM-D can use 8 where a Nikon D800 has to blink through aperture 16 to get the same DOF. That translates to lets say an OM-D at 200 has to be compared to a FF at 800 (something most camera testers also seem to forget).

7) Lens performance. Lenses for smaller sensors have a smaller image circle and the smaller the image circle the easier a lens is to build. So if you compare a Nikon 35 with an Olympus 17 or Sigma 19 on ephotozine and check the IMATEST results, you'll find out that the overall performance of those pesky micro 4/3 lenses isn't half bad, to say the least. In short 36 Mpixels of blur translate to well, blur that you can magnify a lot. Now that is an exageration in fact Nikon makes okay lenses, but not all of them are up to the task of servicing the 36 Mpixel machine.

8) Desired output size......if you enlarge to A3 (magazine spread) 16 Mpixels is enough to give 300 dpi.....more does Anna Wintour not need to be happy. If you however make a high def poster for a car stand or a large scale poster.....well then 16 Mpixels might not be enough. In this case sensor resolution beats image resolution. Allthough pixeling might do the trick as Ming Thein once proved when he uppixeled 16 Mpixel OM-D shots and mixed them with 24 Mpixel (downpixeled) D800 shots. Client was non the wiser.

Now when you add all this together you can device situations (like for instance Scarlet Johansson on a red carpet) where a micro 4/3 like the OM-D can outplay it's strength (fast single AF (and accurate), bigger DOF, great IBIS etc. and lower weight) against a Nikon D800 (BTW a Nikon D4 can do this also since it has better ISO performance).

When shooting architecture (using a tripod) and using optimal light (sunny 16 anyone, meanig 100th of second, using aperture 16 at 100 ISO) a D800e will pulverise a OM-D (no doubt about it) using good glass of course (for instance Zeiss glass performes quite well on Nikon). But a D800e is bit of a prima ballerina it's a medium format DSLR in disguise and needs to be shot that way. If you use it like a ordinary DSLR you will get nice pictures but they will never reach their full potential (I know of photographer of the professional kind that cancelled D800 orders when the D600 came out, just because of this reason).

Now these two links to a German magezine are quite revealing, because here two camera's are parred.....a 16 Mpixel D4 versus a 16 Mpixel OM-D. The game is resolution.....and guess what....

http://www.colorfoto.de/produkte/nikon-d4-1277399.html (the Nikon D4)

http://www.colorfoto.de/produkte/olympus-om-d-e-m5-1273871.html

Now "Auflösung" is translated with resolution and higher is better.....and indeed the OM-D wins. And why, well I forgot one number in the game.

9) AA filter. As Pekka pointed out (and I had already noticed) the OM-D uses a AA (anti aliasing) filter of the flimsy kind. I would call it non existent either. Now AA is nothing more or nothing less then a build in blur filter over the sensor. The Nikon D800e does not use one, neither does the X-Pro 1 and the OM-D followed that route. The Nikon D4 however does use a standard (old school) AA filter, cripling it.....

So stop reading about sensor data. My first digital camera was a Konica Minolta Z3.....3.2 megapixel and even that could (thanks to some great optics) produce convincing A2 pictures. From that little camera I learned one thing.....a digital camera is system and a system has use cases. And you have to keep that in mind, all day and every day. No camera is a jack of all trades. A guy from Luminous Landscape compares cameras to cars and I parrafrase here:

"You can lust for a Ferrari but when you have to take 6 kids to a baseball game a minivan is a better solution".

Hope I made things clear to you.

Greets, Ed.