Hasselblad X2D II summary

The nice folks at LensRentals sent me a Hasselblad X2D II and a 35-100 XCD lens to test. Spoiler alert: I like them both. I’ll first talk about the camera, and then, in my next post, the lens.

The X2D II is the camera the original X2D should have been. It’s fixed most of the maddening things about the Mark I and retains almost all the good things. The autofocus, which before was so persnickety that I didn’t even try to use it in social situations, is now much improved. AF-C, which was MIA in the first version, is now fully present, although with some vicissitudes. The full well capacity is improved with virtually no highlight nonlinearities. Wonder of wonders, there’s a joystick.  Even the door to the flash card, which on my X2D Mk I was so stiff that I had to slick gaffer tape on it to make it grippy enough to open it, is now perfect: easy to open, but not so loose that it can open when you want it closed.

The menu structure remains lean and clean. The camera remains physically lovely. The top panel is even more understated, with the front and top logos unpainted. That’s a nice look, but it has the unfortunate effect of making the bright orange shutter release button look even more out of place. The sharp edges milled into the top panel have the usual issues associated with such small radii, which the anodizing having a tendency to come loose there, revealing the silver-colored body material. Not much shame in that, Hasselblad; even Apple has had the same problem with some of their iPhones.

I’ll get to the technical details, but let’s start with one of the biggest improvements, even though I’ve not come up with quantitative tests. The autofocus accuracy has not improved in static situations, but the camera can acquire focus in dim light and backlit situations much more consistently. In AF-S mode with the 35-100 XCD, the camera is almost as speedy as the GFX 100 II with the 45-100 GF. Eye detection works. You won’t confuse the responsiveness with the AF of, say, the Nikon Z8, but you won’t be wanting to throw your camera at the nearest wall in situations where there’s moderate movement and the light is not great.

AF-C is more like the dancing bear. The marvel is that it dances, and it’s rude to criticize its choreography. It is indeed AF-C, and the accuracy is pretty good, but there are issues. When you first half-press the shutter release, there is a multi-second delay where the camera says to itself, “AF-C? AF-C? Oh, yeah; I remember how to do that.” After the first shot with the shutter in continuous mode, it’s fine, but the lag is initially shocking. Continuous mode exposes the underbelly of 33×44 MP cameras. It takes so long to read out the sensor that the viewfinder just shows you the last picture you took, which is somewhat of a pain in the thighbone.

The X2D II uses lidar for focusing assistance. Radar stands for radio detection and ranging. Lidar stands for light detection and ranging.  In this case, you can’t see the light. It’s infrared. There are unconfirmed reports that the lidar AF is effective up to about 5 meters. However, I haven’t noticed any improvement in system response for subjects closer than about 16 feet, so I can’t say how much of the improvement in AF performance is due to the lidar, and how much stems from other changes.

The rear display is brighter, which is nice. I didn’t notice the extended color gamut.

Hasselblad claims a 10 stop IBIS rating using the CIPA test conditions, a 3-stop improvement over the Mark I camera. There are lies, damn lies, and benchmarks, and I think the CIPA rating system has now diverged so far from actual shooting experience that the numbers are essentially useless. Without IBIS and with a 100 mm lens on the camera, I would normally use shutter speeds of 1/200 second or shorter to consistently obtain something critical sharpness. With the 35-100 set to maximum focal length and IBIS on, 1/20 is safe. That’s 3.3 stops. Ten stops would be 5 seconds; that’s not happening, at least with me holding the camera. I will say that the IBIS on the X2D II seems to be better than the IBIS on the X2D Mark I.

I don’t shoot JPEGs, so I didn’t test and will not comment on the HDR features of the X2D II.

Now for some technical details:

Here’s a photon transfer curve at base ISO.

The normalization is the same as performed by Bill Claff for his photographic dynamic range calculations. The horizontal line indicates the threshold the Bill uses for PDR calculations and show that the X2D II has a PDR of about 12.5 stops. Bill got 12.46 stops using a different method. The above is for 16-bit precision. The 14-bit precision curves are virtually identical.

Here’s what I got for full well capacity and input-referred read noise at base ISO.

The FWC is more than a third of a stop improved over the X2D Mark I.

Here is the engineering dynamic range with a 0 SNR threshold as a function of ISO setting at 16-bit precision.

The drop at ISO 200 is due to the change in conversion gain at that ISO setting. The drop off after ISO 6400 is not real. It’s the result of the camera not increasing gain beofre writing the files, but instead sending instructions to the raw developer in the metadata to push those files. The 14-bit curves are pretty much the same. Like the X2D Mark I, there is little reason to use the 16-bit precision raw settings with this camera.

As is true of most cameras, longer exposures adversely affect read noise:

14-bit precision produces similar curves.

All in all, this is quite a capable camera, and one whose envelope of performance is a large step up from its predecessor.