Side-by-side comparison of Hasselblad 907x + XCD lenses vs. Phase One IQ4 + Hasselblad V CF lenses

[onasj]

Here's a tripod-mounted comparison of the same outdoor scene at f/8, ISO 100-200 between:

907x-50c + XCD 80/1.9 @ f/8
vs.
IQ4 + 100/3.5 CF @ f/8

and

907x-50c + XCD 135/2.8 + 1.7xTC (230/4.8) @ f/8
vs.
IQ4 + 250/5.6 SA CF @ f/8

Captured in RAW, auto-adjusted in C1 and PS and converted to highest quality JPGs. Complete, full-resolution files:

907x-50c + XCD 80/1.9 @ f/8: https://www.dropbox.com/s/5jybqhwis7fndks/907 + XCD 80 + CFV-II-50c 1-500s, ISO200.jpg?dl=0
IQ4 + 100/3.5 CF @ f/8: https://www.dropbox.com/s/klydaal8hepkxcl/ALPA TC + 100 CF + IQ4 1-700s, ISO200 P0003169.jpg?dl=0

907x-50c + XCD 135/2.8 + 1.7xTC (230/4.8) @ f/8: https://www.dropbox.com/s/c83k8acexeu5ug3/907 + XCD 135 + 1.7xTC + CFV-II-50c 1-640s, ISO200.jpg?dl=0
IQ4 + 250/5.6 SA CF @ f/8: https://www.dropbox.com/s/mezney9342b8b74/ALPA TC + 250 CF SA + IQ4 1-700s, ISO100 P0003158.jpg?dl=0

You can see that each pair of images covers approximately the same field of view. Here's the uncropped scene from the 100/3.5:


And here are 2600-pixel wide crops from mid-field (~1/3 to 2/5 of the frame width from the edge):

907x-50c + XCD 80/1.9 @ f/8:


IQ4 + 100/3.5 CF @ f/8:


907x-50c + XCD 135/2.8 + 1.7xTC (230/4.8) @ f/8:


IQ4 + 250/5.6 SA CF @ f/8 (yes, that's a jet in the distance photobombing my test shot):


Both cameras and lenses can take excellent photos under these conditions.

In terms of lenses, the decades-old Hasselblad V 100/3.5 and 250/5.6 superachromatic are outstanding performers. In general, XCD lenses are sharper, more contrasty (and more auto-focus-y than the V lenses, but these two V lenses are considered among the very best of the V lineup, and it shows. Both XCD lenses are also outstanding, though I was surprised to see noticeable CA from the 80/1.9 at f/8. The XCD 135+1.7x TC performs like a premium 230-mm prime lens. Purely on image quality I would rate the 250/5.6 SA as the most optically impressive lens of this excellent bunch, because it is virtually free of any color fringing even in this very high-contrast scene, and captures a stunning amount of detail edge-to-edge. Of course if you need autofocus, none of that will help you.

When examining the 100% crops, the 3x higher resolution of the IQ4's 151-MP IMX411 sensor is apparent, as well as the higher light sensitivity of its BSI architecture, with the IQ4 needing 1/2 to 1 stop of less shutter-open time or lower ISO to achieve the same overall exposure.

The substantial performance benefits of the IMX411 BSI sensor in terms of light sensitivity and detail capture is a reminder that a 44x33 mm IMX461 sensor, the cropped version of the 54x40 mm IMX411 sensor, in a future CFV-III-100c would offer a very nice step-up in image quality (hint hint, Hasselblad!).

Overall, kudos to Hasselblad for more than half a century of outstanding lenses and cameras, and to Phase One's IQ4 back for continuing to offer a sensor of unparalleled image quality.

 

[Paratom]

Thanks for taking the time for this comparison. Did you also print?
To me the most appearent difference in the posted images is a little more saturation in the 907x image, and a little more neutral looking in the Phase one.
The second is different because there is sun in one and cloudy in the other shot.
The phase definatly managed better to capture the airplane
WHat are your findings in day to day photography ? Which do you use for which?

 

[spb]

The phase definatly managed better to capture the airplane

I did not see the 907 capture of the airplane so cannot judge that Phase managed it better.

 

[onasj]

Thanks for taking the time for this comparison. Did you also print?
To me the most appearent difference in the posted images is a little more saturation in the 907x image, and a little more neutral looking in the Phase one.
The second is different because there is sun in one and cloudy in the other shot.
The phase definatly managed better to capture the airplane
WHat are your findings in day to day photography ? Which do you use for which?

For street or other moving subject photography, shooting with V lenses is a challenge. It can be a fun challenge like using a Leica to shoot moving subjects can be, but the long focusing throw of V lenses makes it difficult to know you will get the shot for sure. But the image quality of the IQ4 back is still the best available in the world now, and XCD lenses can’t cover a 54x40 mm sensor. So if you want the highest possible quality landscape photos, the IQ4 and top lenses that can cover a 67 mm diameter image circle, such as those by Rodenstock, Schneider, Phase/Mamiya, or Hasselblad V, would be my choice. For general medium format photography where large prints or heavy cropping aren’t involved, XCD lenses with the 907x-50c or the X1Dm2 would be my choice. I haven’t tried a Fuji GFX system yet beyond a couple minutes in a store, but from what I’ve read I’m sure it can also offer outstanding 44x33 mm (crop) medium format quality.

But for grabbing a body and a lens or two to go, it’s hard to beat the 907x-50c. Plus, it’s gorgeous.

 

[nathantw]

Thanks for taking the time to do this comparison. It's always helpful to see what lies "on the other side of the tracks."

That said, I know you were basically testing the sensors so lens performance wasn't a priority and almost all lenses perform best 2 or 3 stops down from maximum so it didn't surprise me you found they were a bit similar. If you wanted to see a difference with the lenses then setting the aperture at their maximum aperture (i.e. f/4) probably would have been more illuminating.

 

[jng]

In my experience on the X1D, IQ160 and IQ3100 (and born out by looking at the MTF charts - cf http://www.hasselbladhistorical.eu/HW/HWLds.aspx), the 250 SA is sharp wide open at f/5.6 with perhaps some discernible improvement in sharpness/contrast at the edges by stopping down. I generally stop down to f/11 for depth of field and/or to compensate for any inaccuracies in focus (to f/8 if I'm feeling lucky) - this lens is not at all forgiving. I have similar experience with the 100/3.5. Both lenses are diffraction limited when it comes to resolution and are gems in the way they render.

John

Thanks for taking the time to do this comparison. It's always helpful to see what lies "on the other side of the tracks."

That said, I know you were basically testing the sensors so lens performance wasn't a priority and almost all lenses perform best 2 or 3 stops down from maximum so it didn't surprise me you found they were a bit similar. If you wanted to see a difference with the lenses then setting the aperture at their maximum aperture (i.e. f/4) probably would have been more illuminating.

 

[FloatingLens]

... the 250 SA is sharp wide open at f/5.6 with perhaps some discernible improvement in sharpness/contrast at the edges by stopping down. I generally stop down to f/11 for depth of field and/or to compensate for any inaccuracies in focus (to f/8 if I'm feeling lucky) - this lens is not at all forgiving.

My copy of the SA is from early 1972, so possibly one of the first few ever made. My experience is very similar. Resolution is extremely high at full aperture, contrast not so much. Micro-contrast peaks at f8 or f11. At pixel level, the focus plane at f5.6 is so shallow, it is almost impractical to use. That being said, irrespective of the chosen aperture, the overall image looks great and focussing errors are secondary at normal magnification level.

I still have to take some practical images with the Planar 100 and the new CFV to follow-up your statements. So far, I used the Planar 110 more (electronic shutter to the rescue) and I really like the results stopped down a lot. The files it yields are really impressive in their own regard...

 

[jng]

I hope people here will indulge me as I use this thread as an excuse for posting the following image, taken as I was testing the new Cambo HVSA adapter using my Zeiss 120/5.6 S-Planar, which therefore is missing from the image. Together with the 100 CF, 150 C and 250 SA, these old Zeiss lenses have earned a more or less permanent place in my kit.


WRS1250 | IQ3100 | Zeiss 120/5.6 S-Planar | f/5.6 | 21-image focus stack

Note regarding the HVSA adapter: it's well-machined and a nice piece of kit! The shutter mechanism works as advertised although I used the back's electronic shutter for this 21-image focus stack.

John

 

[darr]

I hope people here will indulge me as I use this thread as an excuse for posting the following image, taken as I was testing the new Cambo HVSA adapter using my Zeiss 120/5.6 S-Planar, which therefore is missing from the image. Together with the 100 CF, 150 C and 250 SA, these old Zeiss lenses have earned a more or less permanent place in my kit.

View attachment 176667
WRS1250 | IQ3100 | Zeiss 120/5.6 S-Planar | f/5.6 | 21-image focus stack

Note regarding the HVSA adapter: it's well-machined and a nice piece of kit! The shutter mechanism works as advertised although I used the back's electronic shutter for this 21-image focus stack.

John

John,
Nice image and collection of lenses.

Quick question, how does this lens adapter cock the lens?
The adapter made for ALPA by a Korean manufacturer, uses the same type of mechanism from the FlexBody, but here I see no knob to cock the lens.

Thanks,
Darr

 

[jng]

Hi Darr,

The shiny tab at 11 o’clock is attached to the cocking mechanism. Upon releasing the shutter the tab travels clockwise around the circumference of the barrel; pulling it back around to the position shown re-cocks the shutter. It’s basically a different implementation of the Flexbody‘s mechanism, using the tab instead of the little dial to rewind the screw.

John

 

[darr]

Hi Darr,

The shiny tab at 11 o’clock is attached to the cocking mechanism. Upon releasing the shutter the tab travels clockwise around the circumference of the barrel; pulling it back around to the position shown re-cocks the shutter. It’s basically a different implementation of the Flexbody‘s mechanism, using the tab instead of the little dial to rewind the screw.

John

Thanks John.

The FlexBody knob can be difficult to turn depending upon lens configuration.
For example, when I use the CFi 120/4 alone, no problem turning the knob. But when I use it in combination with two extension tubes, it is harder to turn.
Any difficulties wit your adapter?

 

[jng]

Hi Darr,

I just checked. There is indeed increased resistance with added extension tubes. I find this to be the case on my 500 series bodies as well (the autobellows really adds resistance). It's still pretty easy to cock the shutter (just takes a bit more scrunching up of the nose), although after a few cycles I did notice dents on my thumb and forefinger. My dealer told me that the final decision on the size of the tabs was a tradeoff between snaggability (if that's a word) and ease of use.

One thing I noticed is that the shutter is a bit reluctant to close when set to bulb mode, often requiring a little flick to the tab after releasing the cable release's plunger. This is essentially a non-issue for me as I would just use the back's electronic shutter for long exposures, but not all MFDBs provide that option.

John

Thanks John.

The FlexBody knob can be difficult to turn depending upon lens configuration.
For example, when I use the CFi 120/4 alone, no problem turning the knob. But when I use it in combination with two extension tubes, it is harder to turn.
Any difficulties wit your adapter?

 

[FloatingLens]

The FlexBody knob can be difficult to turn depending upon lens configuration.

Not surprising as each accessory is spring-loaded.

 

[FloatingLens]

I hope people here will indulge me as I use this thread as an excuse for posting the following image, taken as I was testing the new Cambo HVSA adapter using my Zeiss 120/5.6 S-Planar, which therefore is missing from the image. Together with the 100 CF, 150 C and 250 SA, these old Zeiss lenses have earned a more or less permanent place in my kit.

WRS1250 | IQ3100 | Zeiss 120/5.6 S-Planar | f/5.6 | 21-image focus stack

Impeccable trio! I like how all settings are right aligned and at the infinity mark – nice touch.

Whats your focus stacking SW solution btw?

I've got my first XCD lens and now all is set to try that myself with the 907X.

 

[Godfrey]

Nice lens kit.

In the Hasselblad webinar centered on Focus Stacking, they highlighted Helicon Focus and Photoshop. I'm going to give Helicon Focus a try.
The webinar is available on Hasselblad's YouTube channel @

G

 

[jng]

Impeccable trio! I like how all settings are right aligned and at the infinity mark – nice touch.

Whats your focus stacking SW solution btw?

I've got my first XCD lens and now all is set to try that myself with the 907X.

I use Helicon Focus for stacking tiffs exported from my raw processor(s) and Photoshop for final edits.

Nice lens kit.

In the Hasselblad webinar centered on Focus Stacking, they highlighted Helicon Focus and Photoshop. I'm going to give Helicon Focus a try.
The webinar is available on Hasselblad's YouTube channel @

G

I've toyed with the X1D's focusing stacking function as well. I don't use it much so it usually takes me a while to find it in the menu system. Since Hasselblad's implemention only allows one to specify a starting point (vs. a starting and end point) plus number of frames and steps, it's a bit of a trial and error process to get the number and size of steps right. I imagine that this all becomes more intuitive with practice. Anyway here's a 25-shot focus stack of all four lenses using the X1D with XCD 90/3.2 shot wide open, starting with near point focus and (I think) medium steps. I didn't think to shoot a grey card so I could match the color balance between this and the previous image I posted using the 120 S-Planar/IQ3100 combo (which was processed through Capture One instead of Phocus before processing with Helicon and PS), so I just seasoned to taste in each case. Nonetheless the two imaging pipelines do seem to give a different look over all.

John

Zeiss Magic

 

[spb]

Nice looking lenses

 

[SrMphoto]

I use Helicon Focus for stacking tiffs exported from my raw processor(s) and Photoshop for final edits.



I've toyed with the X1D's focusing stacking function as well. I don't use it much so it usually takes me a while to find it in the menu system. Since Hasselblad's implemention only allows one to specify a starting point (vs. a starting and end point) plus number of frames and steps, it's a bit of a trial and error process to get the number and size of steps right. I imagine that this all becomes more intuitive with practice. Anyway here's a 25-shot focus stack of all four lenses using the X1D with XCD 90/3.2 shot wide open, starting with near point focus and (I think) medium steps. I didn't think to shoot a grey card so I could match the color balance between this and the previous image I posted using the 120 S-Planar/IQ3100 combo (which was processed through Capture One instead of Phocus before processing with Helicon and PS), so I just seasoned to taste in each case. Nonetheless the two imaging pipelines do seem to give a different look over all.

John

<snip>

If I don't need significant lens corrections, I export DNGs to Helicon Focus and get a DNG back. The advantage is that I can process the resulting image after Helicon Focus has stacked them.
An advantage of Hasselblad's implementation is that you can specify the center of focus stack, instead of only the start or end, as the limits are sometimes hard to identify. I wish Hasselblad would implement a feature similar to Fuji: allow specifying start/end, and the camera determines the number of shots needed.

 

[FloatingLens]

I wish Hasselblad would implement a feature similar to Fuji: allow specifying start/end, and the camera determines the number of shots needed.

I do, too.
Hasselblad's philosophy seems different, based on CoC at a given aperture and distance.
Actually, their approach is rather technical looking at the user's manual.

 

[SrMphoto]

I do, too.
Hasselblad's philosophy seems different, based on CoC at a given aperture and distance.
Actually, their approach is rather technical looking at the user's manual.

Fuji's and Nikon's approach is similar to Hasselblad's implmentation. The difference is that only Hasselblad explains it properly.