Camera Labs

Quoted verbatim from the new DxO review here.
So I was naturally inclined to have a look at this 80MP medium format monster and learn how it dethroned the best sensor so far: the Nikon D3x.
And mind you, I didn't expect much of a battle there because the Phase one sensor has 40.4 x 53.7 mm in size that is 2.5x the size of a full-frame sensor and the back alone costs around 40k$
BUT...
1. a look at the ISO values indicated that something is seriously wrong with the Phase-One: the measured ISO is only 1/3 of the manufacturer ISO: So ISO 100 is in reality only ISO 29
2. On a per pixel basis (DxO calls this "screen") the values of the Phase-One for SNR, dynamic range, tonal range and color sensitivity are always below the D3x even at 29 ISO! You have to take into account that the pixel-pitch at 5.17 is a little smaller than the 5.9 of the D3x. A comparable FF/FX sensor would have 32MP at the same pixel-pitch.
3. On a comparable output size (DxO calls this "print") the Phase-One is better at SNR, tonal range and color sensitivity only at ISO 60 and ISO 29. But it cannot top the D3x at dynamic range at any settings. If you use comparable ISOs the Phase-One is 1-1.5EV worse at dynamic range than the D3x and only comparable in the other categories.

Conclusion:
There's only one real advantage of the best FF/FX-sensors today: detail! That is if the lenses you put in front of that sensor can resolve as good as the best film-lenses. And as the pixel pitch is indeed smaller than that of a D3x you'd even get diffraction limited faster and will see any lens-aberrations like CA "better".
Only when working at ISO 60 and ISO 30 will the "King of all sensors" have any measurable IQ advantage over a D3x. And if you think that such a large sensor will isolate your subjects even better than a FF/FX body, think again: If you're doing portraiture with a 85/1.4 on a D3x you'd need a 135/2.2 to achieve the same dof. Or if you want something similar to the Nikon 200/2.0 you would have to get a 300/2.8 to be at least on a par.


So if you have enough light in your studio to work at ISO 30 and stop down your medium-format lenses to live up to the narrow pixel-pitch you can expect images that have 2.5x the details at lower noise and better tonal and color-fidelity than you can produce with the best film-format cameras today.
If...

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Ok, that's a tough one to compare!

On the question of the low measured vs nominal ISO settings, I'm not sure that matters providing there's enough headroom on either end of the scale. Practically, at worst you have a bit more shadow noise but more highlight headroom.

The biggest weakness of the sensor is that it is a CCD. Compared to CMOS they just don't do high ISO well, but my understanding is they're easier to scale in size than CMOS. Look at the digital medium format offerings, I think they all use CCD.

Overall, it may be a bit niche, but if you really need it, you need it. Certainly everything below medium format is good enough for well over 99.9% of people.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi Thomas,

Interesting. I'm struggling to understand what's in it for Phase One when, for example, the measured ISO is 120 when they claim 400. The D3X is a little more honest with a measured ISO of 337 but that's still some way off.

Might the inability of the Phase One sensor to match the ISO performance of the D3X be down to much less effective microlenses and even a lesser quantum efficiency?

I suppose the camera manufacturers will claim that we don't actually need to know what the base ISO of a sensor is so long as the camera manages to produce a correctly exposed image if we were to go "old school" and set the aperture and shutter speed as indicated by an external light meter for the selected ISO. The rest is all smoke and mirrors with the only external sign that there has been tinkering by the camera or external RAW converter being image noise, or lack of it.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Headroom or not: The dynamic range measures just that and it shows that the Phase-One has no extra Headroom.
The effective ISO simply shows what exposure is needed to produce a predefined level of brightness in the RAW data.
So in the end methinks something is limiting the theoretical performance-potential of the Phase-One sensor be it quantum efficiency, read-out noise etc.
But at that price it's really a pity that it didn't trash the competition at normal ISOs

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses