The nice folks at Lensrentals sent me a Hasselblad X2D-100CII and the 35-100mm XCD lens to test. This is the first in what will probably be about a dozen posts on the camera and the lens. One of the things that I do with a new camera to suss out how much spatial filtering exists is run a dark-field series at ISO 1000 at all timed shutter speeds. I did that today with the X2D, using single shot shutter mode, with mechanical and electronic shutter at 14 and 16 bit precision. I’ll present the result as engineering dynamic range (EDR) for each of the four raw channels. In this case, EDR is defined as full scale over read noise.
We can see that this is a dual conversion gain sensor, with the gain increase occuring at ISO 200.
At 16 bit precision, there is a small improvement. We’ll look at that more closely later.
Now let’s compare 14 to 16 bit precision, averaging all the raw channels.
About the same.
Also about the same.
Now we’ll compare ES to MS.
It doesn’t make much difference which shutter type you use.
These numbers do not represent the kind of improvement in EDR you’d expect to see when going from the X2D with base ISO of 64 and the X2DII with base ISo of 50. In fact, EDR at base ISO is a hair worse with the Mark II version of the camera. The flattening of the curve at high ISO is usually an indication that the camera has stopped applying gain to the raw files, and is counting on the raw developer doing that. This was done in the early versions of the GFX 50 cameras, and in my opinion is a good way to handle ISOless portions of the camera’s response to photons.
If, instead of looking at EDR, we look at input referred read noise in electrons, we get this:
I won’t bore you with the other precision and shutter mode settings. There isn’t much difference among them. The sag at the high ISOs is not real; it’s the result of the camera’s flattening of the gain applied at those settings. This camera is esssentially ISOless from ISO 200 on up.
Leave a Reply