fat pixel digital backs

[jng]

I wonder if the beauty of the fat pixels you see can be attributed to the drawing of the older lenses instead. I see an "improvement" when shooting with older V instead of new XCD lenses.

Very interesting. Sometimes tricky to correctly ascribe causation when two variables change simultaneously.

Cue run-up in prices on older V lenses…

Welp, these latest posts got me thinking (dangerous, I know)... Having convinced myself that the legacy V system lenses render less clinically than the modern lenses (that old "Zeiss magic" thing), I decided to put my assumptions to the test, holding as many parameters constant as possible. So...

Lenses tested: XCD 3.2/90 vs Zeiss CF 3.5/100 Planar
Subjects: two of my old V system lenses, shot at an approximate distance of 1 meter (adjusted to give comparable fields of view), focused on "f=120mm" of the front of the S-Planar
Camera/sensor: Hasselblad 907x-CFV100C
Camera settings: aperture priority automatic exposure; same white balance settings for all; electronic shutter
Post processing: Phocus using default settings (w/default lens corrections for XCD90 including CA corrections; no lens corrections applied to 100 Planar); cropped square and exported as jpegs without any additional adjustments

Each lens was tested at f/3.4~3.5, f/4, f/5.6, f/8, f11, and f/16. Below are sample images shot at f/3.4~3.5 and f/8. The entire series can be found here.

I'll leave it to folks here to draw their own conclusions, but with the exception of focus roll-off (seems steeper with the XCD90), to my eyes the old and new glass render more similarly than I had expected, at least under these conditions. It's very possible that more obvious differences might be found with different lenses (the 3.5/100 Planar is touted to be one of the sharpest lenses in the lineup), but I'll leave it to others to go down that rabbit hole. Suffice it to say that the old V system lenses more than hold their own - I'm happy to keep them in my kit alongside the XCD lenses.

John

 

[SrMphoto]

Welp, these latest posts got me thinking (dangerous, I know)... Having convinced myself that the legacy V system lenses render less clinically than the modern lenses (that old "Zeiss magic" thing), I decided to put my assumptions to the test, holding as many parameters constant as possible. So...

Lenses tested: XCD 3.2/90 vs Zeiss CF 3.5/100 Planar
Subjects: two of my old V system lenses, shot at an approximate distance of 1 meter (adjusted to give comparable fields of view), focused on "f=120mm" of the front of the S-Planar
Camera/sensor: Hasselblad 907x-CFV100C
Camera settings: aperture priority automatic exposure; same white balance settings for all; electronic shutter
Post processing: Phocus using default settings (w/default lens corrections for XCD90 including CA corrections; no lens corrections applied to 100 Planar); cropped square and exported as jpegs without any additional adjustments

Each lens was tested at f/3.4~3.5, f/4, f/5.6, f/8, f11, and f/16. Below are sample images shot at f/3.4~3.5 and f/8. The entire series can be found here.

I'll leave it to folks here to draw their own conclusions, but with the exception of focus roll-off (seems steeper with the XCD90), to my eyes the old and new glass render more similarly than I had expected, at least under these conditions. It's very possible that more obvious differences might be found with different lenses (the 3.5/100 Planar is touted to be one of the sharpest lenses in the lineup), but I'll leave it to others to go down that rabbit hole. Suffice it to say that the old V system lenses more than hold their own - I'm happy to keep them in my kit alongside the XCD lenses.

John

The fact that you do not see a difference in this subject/test does not mean you will or will not see a difference in a different subject or scene.
I was doing some shots with SuperAchromat 250 and felt that the images looked different (better?) than when shot with XCD lenses. However, I did not compare the same scene when shot with an XCD lens.
I agree that the V lenses fare very well.

 

[jng]

The fact that you do not see a difference in this subject/test does not mean you will or will not see a difference in a different subject or scene.
I was doing some shots with SuperAchromat 250 and felt that the images looked different (better?) than when shot with XCD lenses. However, I did not compare the same scene when shot with an XCD lens.
I agree that the V lenses fare very well.

Agreed, which is why I was very specific about the parameters of the test and my qualification of "at least under these conditions" to which I might underscore that another pair of lenses might yield very different conclusions. I should also add that the lenses might behave very differently at infinity focus compared to the relatively close ~1m focus distance. The bottom line for me is that the differences were less than I had expected based on a previous comparison I did testing an XCD lens on the X1D against a Zeiss lens on the Phase One. The colors and tones were very, very different, but I couldn't discern what was due to the lenses, sensor, or raw processing. With the current, more controlled test, my money is on the latter.

Whatever the case, if the more gentle roll off from in-focus to out-of-focus holds true here and for other legacy V system lenses, this could explain the perception that they render less clinically than the more modern glass (although the 250 Superachromat seems a bit less forgiving when it comes to focusing). I think that @MGrayson suggested this a while back (hi Matt!).

I've not had a chance to try the XCD135 + 1.7x extender, which would be the relevant comparison to the 250 SA. Nor will I likely bother. The 250 SA is simply fabulous and moreover I already own it and most of what I do involves static subjects. Ditto for the 350 SA +/- APO 1.4XE, and there's nothing in the XCD lineup in that focal length(s) anyway. Moreover, all of my legacy V system glass performs well on the larger IQ4 150 sensor, which is where they find the most use. YMMV, of course!

John