Is autofocus on 120mm XCD macro improved on X2d compared to X1D?

[MGrayson]

I just received the 120 macro from B and H. I must be dreaming, but the autofocus on my X2D with the new firmware is incredibly fast! From a subject 1 meter away to infinity, lightning fast (maybe a little hyperbole, but for Hasselblad ok). No hunting, just nails it. I was expecting much worse, but admittedly this is in good light. Anyway, very happy.

Wayne

I just measured it with high speed video. It took my X2D and XCD 120 0.9 seconds to go from 2 meters to infinity focus in either direction. About the same from 12 inches to 2 meters, although I didn't time it. Does that sound right?

Matt

[wayne1]

Not quantitative, but moving from 1 meter to infinity or the other way, around 1-1.5 sec. What makes it so much better than previous reports suggest? the phase autofocus? Something in the 3.1.0 FW? Some tweaking of new copies of the lens?
Also, the images are remarkably good at f22; a bit softer than f8 of course, but very useable. Even f45 would do in a pinch if you really needed DOF. I wonder if there is a sharpening algorithm tuned to mitigating the effect of diffraction?

[wayne1]

In regards to sharpening in software to get rid of diffraction, I wonder if one could deconvolute with an Airy disc function appropriate to the lens opening.

[wayne1]

I just found that Capture One has a deconvolution algorithm to remove some effect of diffraction. I don't have Capture One. Can you use Hasselblad raw files?  Has anyone used this? It would be great to extend the DOF with f45, since it depends on the inverse square of the f number.

Wayne

[MGrayson]

In regards to sharpening in software to get rid of diffraction, I wonder if one could deconvolute with an Airy disc function appropriate to the lens opening.

Yes. The problem is that deconvolution requires *dividing* by the kernel transform and that has zeros in it. You can do a lot to repair diffraction, but (as you'd expect from the fact that we don't have infinite resolution optics) you can't get rid of it.

[hcubell]

I don't know the science behind the sharpening algorithms in Focus Magic and Topaz Sharpen AI, but I find the output for X2D files to be extraordinary.

[wayne1]

Thanks!
In fact I just purchased Topaz AI to try out the sharpening algorithms! I already used the denoise earlier, and found it quite good.

Last night I did some shooting in a dimly lit room with the 120 macro, and now I can see the issues that JCM and others had with slow and hunting focus. This will not be an issue for me, since my use will be outside in decent light where the autofocus works extremely well, but I can see for studio work it could be annoying.

Wayne

[wayne1]

hcubell

Do you import tiff files already processed in Phocus into Topaz AI? I see that it does support Hasselblad raw files so one sharpen first in AI, but then the output will be either a DNG or TIFF I guess? In that mode, you lose the lens corrections and color space in Phocus, so I guess I will process in Phocus, export as 16bit tiff and finally sharpen in AI, if needed. Actually, I generally need almost no processing in Phocus on files from the X2D prior to printing from LRC. That's one of the reasons I love this system. However, with the 120 macro I can see that one might want to use small apertures (large f) for DOF, and if one could sharpen diffraction limited images, that would be nice. Macro is new territory for me, and I am learning.

Wayne