Comparison of Hasselblad photos / lens quality with XCD lenses versus Zeiss V system lenses?

Discussion in 'Medium Format' started by 10961137, Oct 26, 2019.

  1. Hi all,

    One can find several comparisons between the XCD lenses for the Hasselblad X1D (II) mirrorless, medium-format digital camera and the full-frame (35mm) Otus lenses made by Zeiss on the same camera (using a suitable adapter, of course), as e.g. done by Usman Dawood:

    Hasselblad X1D and the Zeiss 85mm Otus: The Ultimate Combination?

    I'm rather curious about how the XCD lens lineup match up against the classic V system lenses made for the Hasselblad 500-series film cameras by Zeiss (C, CF, etc.) using Hasselblad's XV adapter for the X1D (II) as far as the photo quality / characteristics are concerned, but I couldn't find such a comparison so far.

    Of course, I'm not considering features such as autofocus capabilities (which the V system lenses obviously don't have), I'm really trying to make an apples-to-apples comparison here. Just looking at the spec sheets provided by Hasselblad in its website and e.g. the Zeiss spec sheets available at the Hasselblad Historical web page, one can see that (at least on paper) the XCD lenses are faster and are supposed to have better color rendition since they are apochromatic lenses. However, how do they compare against old flagships such as the super-low-distortion 100mm Planar in a "real life" situation - that is, on the photos?

    I mean, my point is: apart from autofocus and lower f-stops, is there any other reason for which the XCD lenses are objectively superior to the old Zeiss V system lenses as far as the final result is concerned, and therefore justify Hasselblad's marketing and price tag? I guess color rendition should be better due to apochromaticity, but that is on paper... How about sharpness / bokeh / etc.?

    Any links to such analyses / photos are welcome!
  2. You aren't seeing much info about this combo simply because it isn't practical yet. Unlike the adapter for electronic H lenses, the adapter for mechanical V lenses has no ability to operate their leaf shutter. You must fall back on the X1D sensor-based electronic shutter, which in these large MF sizes isn't as effective or usable as it is in smaller-sensor cameras. It constrains your choice of shooting conditions and subjects, and doesn't make the best use of the glass in front of it.

    Reports are mixed from those who have tried the V lenses on the Fuji GFX (which has the same sensor as X1D but a more versatile standard focal plane shutter) or Hasselblad H bodies with the same sensor. The crop factor is pretty strong, and some of the V lenses aren't quite up to snuff on the demanding 5.3 micron pixels. It depends what qualities of the V lenses you prize most: you'll get the color rendering and other factors, but absolute resolution won't seem as good as it was on 6x6 film or the older, larger CCD sensors that were commonly used with V lenses a decade ago. The cream of the bunch like 100mm Planar and 250mm Super Achromat still hit the mark on 50MP 33x44 sensors, but many popular focal lengths long considered superb on film (like the lovely 60mm Distagon) can disappoint. Roughly speaking, bokeh is comparable between X and V assuming similar focal lengths: both are hamstrung by pentagonal diaphragms, but the different optics have some effect.

    The V lenses are a nice supplement to an X1D kit if you already own them, but it probably isn't worthwhile to buy them instead of dedicated X lenses unless most of your work is very static, and a comparable X lens is too expensive. The X lenses seem expensive because they are current "boutique" items with costly high-speed leaf shutters and digital-optimized glass. The V lenses were designed decades ago with the somewhat lower resolution required by a film frame two or three times larger (the same way fine lenses for 4x5 view cameras are "lower resolution" than XCD lenses). So the current Otus lenses would be a more apples-to-apples Zeiss/Hassy X comparison. Also one needs to remember the bargain price of used V lenses today bears no relationship whatsoever to their historical value: 25 years ago the 50mm Distagon V sold new for more than the current price of the XCD 45mm, even adjusted for inflation.
  3. FWIW, over the years I've occasionally been able to use my Zeiss V lenses with several digital backs on my 500cm and elx bodies (50, 60, 80, 100, 120, 150, 250). They performed rather nicely on the old Phase One P25+, a CCD back with large sensor much closer to 645 film size than the X1D. They performed less predictably with the newer Hasselblad CFV-50c back, which for all practical purposes is the same Sony sensor as the X1D. Some lenses like the Distagons and 150/250 Sonnar faltered a bit on the smaller denser 33x44 sensor. Not enough to bother me, but perfectionist pixel peepers would be non-plussed. Possibly this is down to the greater precision needed for focus, which the old film bodies are hard pressed to deliver with digital back attached.

    I've also enjoyed my V lenses on 35mm-format (24x36) sensor cameras like Nikon D700, D850, Canon 6D and Sony A7II. The 50mm Distagon, 100mm Planar and 250mm Sonnar are especially sweet on 24x36. If they weren't so large and clumsy, I'd replace some of my Nikkors with my V glass. In the end, how you feel about various performance characteristics will determine your satisfaction with V glass on the X1D (that, and whether the electronic sensor shutter limitation lends itself to your work). Some photographers lose their minds at the slightest hint of CA, I'm much less picky about that and other flaws. The distinctive way the V lenses draw an image is something I really like, that often transcends aberrations. Very subjective.
  4. Hmm... Good point, I didn't pay attention to the fact that the XV adapter cannot give control of the V lenses' leaf shutter to the X1D body. This is certainly a handicap for the use of V lenses on this camera, considering the fact one has to use the on-sensor electronic shutter (at least for now... Until Hasselblad or other vendor comes up with a more functional adapter than the XV).

    Ok, this is the kind of info I was looking for, but I'd like to see this on photos somewhere... I'l try to look for usage of V lenses on the Fuji GFX to have a better impression.
  5. Interesting... In fact, I'm curious (as many out there) about how the upcoming CFV II 50c attached to a 500-series mechanical body will fare in this respect. Given that sensor size is a factor and that won't change from the previous iteration, it seems that the problems you pointed will remain. As a matter of fact, I'm more interested in the CFV II digital back for use in my 503cx than in the X1D itself or on the CFV II companion (?) 907X. It's a pity that Phase One no longer makes backs for other than their own cameras.
  6. I guess so, because the crop factor is so large you are close enough to the lens axis to get a significantly lower distortion (depending on the lens), but these lenses are really large and heavy compared to full-frame 35mm lenses of the same focal length, so it's not something one would want to use on a 35mm camera most of the time. Color rendition and contrast are of course highly subjective - that's the reason I love classic German glass and loathe Canon glass as it shows on photos, for instance.
  7. In due time: regarding manual focusing precision, I reckon that the 100mm Planar's success on the Fuji GFX (and on MF digital cameras on the same sensor) has to do with the fact that, in addition to its extremely low distortion, its focusing mechanism is reputedly very precise, so it should be up to digital standards. The only problem is that, with such a crop factor, its effective focal length gets a bit long and therefore excludes some of its applications. Pity that the Distagons don't seem perform so well in that respect, according to your finds.

    I still couldn't find any reports on how well Hasselblad / Zeiss V system lenses fare on the Fuji GFX, but I did find some photos on Flickr using the CFV-50c back. Not many using the 100mm Planar, though, but I could find several nice examples using the 120mm Makro Planar. Unfortunately, that setup kind of precludes the kind of comparison I asked for, at least until one gets the 907X / CFV II combo to make such a test.
  8. I stand corrected - actually, the IQ3 digital backs from Phase One can be used on Hasselblad H- and V-series camera bodies (but not the IQ4). The problem actually is the car-grade price of those...
    Last edited: Oct 29, 2019
  9. Prices on the more desirable digital backs for Hasselblad V series have remained stubbornly high: lots of people still love the body and Zeiss lenses. You can get excellent Imacon, Sinar and Phase backs for less than $1200, but those are the oldest "blind" backs with no screen that require tethering to a computer. The least expensive back I would consider practical is the Phase One P25, which at least has a large (near-645) sensor, rudimentary screen, and ability to save files on a CF card. Those go for about $3000 still.

    I liked the Hasselblad CFV-50s, except for the idiotic, ugly, un-ergonomic external battery. The price was a bargain next to a comparable Phase One with the same sensor, but still way out of range for most Hasselblad enthusiasts (including me). Relief may be coming, however: pricing for the CFV-50c mkII appears to be dramatically less. The original was introduced at $15K, eventually dropping to about $10K. The new version is available for pre-order right now in a package deal including the new 907 body for just $7500. Assuming that remains the base price, the back alone without the 907 body will likely be $5995 or less: much more accessible to many more Haselblad owners.

    Now if only Haselblad would get its finances in order and actually mass produce the thing...
  10. Will the CFV II be available as digital back only? Or it will be available only coup,ed with the 907X? The Hasselblad WEB site it's not clear on this matter and I suspect they will sell the back only coupled with the 907X (so this combo could lead the custoner to buy X lenses).
  11. The potential market for the 907x is so small its a rounding error. After the initial rush of die-hard SWC fans buy it to finally free themselves from cropping pain and lens cast issues, thats it. Hasselblad needs every last penny it can shake out from the sofa cushions of its former legion of V enthusiasts: theres no way the CFV-50CII doesn't eventually become an independent product. The model name itself still tags it as a V-oriented item: if there wasn't an intent to make a play for the V users with an affordable modern back option, the 907x would have been easier and cheaper to produce as an integrated standalone camera.

    Of course at this point Hasselblad management is sharing the same gin-soaked Ouija board that Nikon relies on for critical decision making. I wouldn't put it past them in the least to delude themselves the 907x would be a coherent product with mass appeal, and only offer the back with it as a set. Doesn't really matter as long as they actually produce them in some plausible quantity: at $7500 for the set its still half the price of the first CFV-50c, with better live view and neater internal battery. Many happy V users would buy at that price and dump the unwanted 907x on eBay, much as Canon 6D and Nikon D660/D750 buyers dumped their subsidized 24-105 kit lenses.
    10961137 likes this.
  12. Well, here we are... the 907X has been officially released at $6399:

    907X 50C - Hasselblad Store

    The best review I've found on it, especially regarding comparison between Zeiss V glass and XCD lenses, was made by Jay P. Morgan on his The Slanted Lens site, reposted at PetaPixel:

    Hasselblad 907X Hands On Camera Review - The Slanted Lens

    Hasselblad 907X Hands-On Review and Vintage Lens Test

    As usual when using vintage glass on modern digital cameras, manual focus precision becomes more critical than on film. This is also emphasized in the CFV II 50C manual. It seems to me that Zeiss glass holds up pretty well in that respect, though (albeit maybe not up to XCD autofocus precision standards), probably thanks to these lenses' superb microcontrast.

    Mike Aubrey commented above on PetaPixel regarding the best choices of V glass for the CFV II 50c (in his opinion):

    Some of his recommendations - both best (Planar f/3.5 100mm, Sonnar-Superachromat f/5.6 250mm) and perhaps not so great (Distagon f/3.5 60mm) - happen to coincide with orsetto's made above. It seems to me, though, that he based his judgement on measured lens sharpness (MTF charts, etc.) rather than actual experience with the combinations. What do you think?
  13. Just a side note to this very interesting conversation. I've seen it written more than once that focus is critical when using a digital back on a V System body. The explanations I have seen many times mention the thickness of film emulsion being more forgiving than the relatively flat surface a of a CCD. Apparently film "absorbs" focus error that a CCD can not because of the emulsion thickness. I surely don't know enough to argue with this logic - it makes sense to me.

    Where I can add to this is mention of the condition of mirror pads in these sometimes 60 year old machines. Hasselblad placed the mirror on three pads within the mirror frame. They are arranged pretty much like a triangle with one pad at the center top of the mirror (the pivoting end of the mirror frame) and two in the corners at the bottom (closest to the lens). These pads "perish" as the Brits like to say. This allows the mirror to move up and down in its frame and makes it unlikely the mirror is restoring to the same point each time the advance knob resets the mirror. This situation is a contributing cause of images that appear in focus on the ground glass but that are not in focus on the film plane.

    Why Hasselblad chose to use only three pads rather than one in each corner of the mirror is a question I would love to know the answer to. Regardless though, its a safe bet that any V-Series body (including 1600F and 1000F) that has spent the last 20 years in a closet, will need new mirror pads. Does Hasselblad in NJ replace the pads when a body comes in for service? I would hope so. Do other specialists out there do this as well when they service a body? Again, I would hope so but I don't really know. What I do know is that focus errors when using a digital back will only be made worse if the mirror pads are shot and your mirror is moving.

    When the pads are shot its fairly easy to detect. Remove the lens and rotate the body 180 degrees so that the tripod mount is facing upward. With a toothpick or something that won't scratch the mirror (eraser end of a pencil, a q-tip, whatever) lightly put pressure on the mirror glass to see if it moves within its metal frame. Release the pressure and see if the mirror glass moves back to the original position. Any movement indicates the mirror is not being supported tightly and the pads may need replacing.
  14. Where have you been Andy?
    CCD sensor technology gave way to CMOS at least a decade ago.

    WRT emulsion thickness; it most certainly does limit sharpness.
    Colour film emulsions may have up to 20 layers, apparently, and a conservative estimate of that emulsion thickness might be 20 microns - possibly more. (Film manufacturers are seemingly a bit shy about putting hard numbers on the thickness of their emulsions.)

    Anyhoo. Let's take a figure of 20 microns thickness and calculate the point spread between top and bottom of such an emulsion. With an 80mm lens and an aperture of f/2.8, it works out that there's a point spread of about 7 microns, meaning that theoretically only 70 line-pairs/mm would be resolvable. And that's without taking account of any diffusion of the light through turbidity of the emulsion.

    That figure improves as the aperture gets smaller, but there's quite a small window of opportunity before diffraction starts to take resolution away again.

    So any lens resolution much in excess of 70 lppmm would be largely wasted on film.

    FWIW, the best 300mm telephoto lens I own is a Mamiya Sekor C, designed for the M645 system. It totally outshines anything I've yet to come across designed specifically for a 35mm system.

    Most of the rest of my Mamiya 645 lenses also perform exceptionally well on a high-res full-frame digital sensor, and generally better than big-name 35mm system lenses from the film era.

    I guess Canon, Nikon and the like made most of their lenses 'good enough' for the 35mm film of the time, and not much better.
    Last edited: Nov 3, 2020
  15. LOL. Well I certainly haven't been at the front of the line. My archaic P45+ has CCD technology. Rather than step into the Roaring 20s, I'm happy where I am and instead, put a new roof on my house. Thanks for the theory lesson though.
  16. Sorry! Didn't mean to lecture Andy.... but the question was asked.

    Keeping the rain off is obviously quite important. Although if someone offered me a 400mm f/2.8 lens to live in a tent for a month, I might be tempted.
    No. On second thoughts, I'd definitely be tempted.
  17. Thanks for all the comments. I think I've been able to get a decent overall picture of the situation concerning using the CFV II 50c back with a V-system camera and lenses, which to some extent also applies to using adapted V-system lenses on Fujifilm GFX cameras. To summarize:

    • The small pixel size of the CFV II 50c makes accurate focusing critical. It baffles me that, unlike the original CFV back, the 907X kit doesn't include the split focusing screen with the X1D / 907X form factor framelines, just a dinky focusing screen mask. The split Fresnel at the middle helps enormously with fine tuning focus, if my previous experience with 35mm film cameras (Leica R6, Minolta SRT-101, Kodak Retina Reflex III) has any worth. You can still buy the screen separately, of course, but it's a pain. Alternatively, you can also lift the mirror and use e.g. focus peaking with live view.
    • Only a few of the V-system lenses are up to the more exacting digital standards of today - most notably, it seems, the 40mm f/4 FLE Distagon, the 100mm f/3.5 Planar and the 250mm f/5.6 Superachromat Sonnar (particularly the latter being an apochromatic lens, just like the XCD lenses). The 120mm f/4 Makro-Planar and 180mm f/4 Sonnar also seem to deliver on the CFV backs, judging from photo samples. For me, the 100mm Planar + 120mm Makro-Planar combo seems especially compelling to that end, since these lenses nicely complement each other (also on film, obviously) - both have similar lengths and perform exceptionally well on the whole aperture range, with the 100mm being the sharpest near infinity (with its famed ultra-low distortion there) and the 120mm being the sharpest at close range (being a lens optimized for close-ups). Moreover, both lenses have a very precise focusing ring.
    • Even though the 907X opens access to modern XCD glass, it only makes sense to buy it over (say) the X1D II if you are already invested into the V system, especially considering that the X1D II costs about USD 500 - 600 less than the 907X and has more features. Of course, new XCD glass, although arguably superior and more adequate for digital photography, is much more expensive than used V glass (especially if you already own it ;) ), but one may also argue that if you have USD 5800 - 6400 to spend on a camera body, you probably can also afford to buy at least one or two XCD lenses instead of a whole V-system setup. On the other hand, that's not what the CFV II 50c back is all about - it's about the whole experience of shooting with a V system you already own being transported into the digital age, which is great provided one watches out for the above provisos. That's clearly the market Hasselblad is aiming at with the 907X, with the possibility of inviting V-system users to try the XCD lenses in time.
    Last edited: Jun 25, 2021
  18. Nitpicking, but...
    The Sonnar Superachromat is not an Apochromat, but a Superachromat. Superachromatic correction (At least four colours in focus, with a secondary spectrum so minimal that it doesn't show itself) is a class above apochromatic correction (three colours in focus, the rest close but probably not). Superachromats can focus near IR in the same plane as visual light. Not many (if any) apochromats can.
  19. While it seems rather strange that anyone would "spend $7-10K to use old obsolete lenses", the situation isn't really that cut and dried. Most potential purchasers of the CFV50cII/907x, X1D series or Fuji GFX series aren't buying these expensive bodies to use only with old Zeiss V lenses, nor are they buying the bodies and then going out to buy a set of old V lenses from scratch instead of the native digital lenses. The typical buyer of these cameras either already owns several V lenses, or plans to perhaps buy one or two as supplements to a primarily current lens kit. IOW, when the V lenses are essentially "free of cost," using them with these new cameras is less of a financial commitment than an artistic or sentimental choice.

    The ergonomics of the specific camera body come into play as well. Despite the advantages of their integrated fully-digital mirrorless design for manual focus lenses, the X1D and GFX are almost absurdly clumsy with V lenses (seriously: slap a mount adapter and the 40mm Zeiss IF lens on the X1D and tell me you feel inspired and motivated by the handling of that combo). The X1D was clearly (and exquisitely) designed to be the ultimate medium format digital hand camera, with coordinated AF lenses of fairly compact size. Ditto the Fuji GFX50R, tho the other GFX bodies are a bit larger (as the Fuji lenses are larger than XCD glass).

    Admittedly only a few of the shorter XCD and GX focal lengths could really be called compact, but the addition of AF and AE compensates greatly. Adding a bulky adapter and manual focus glass from Zeiss, Pentax or Mamiya upends that compromise: the lens hangs that much farther out in front of the body, lack of electronic interface means a lot more fiddling with controls, and if your V lens isn't the very latest CFi version the dog slow focus ring can be a real buzz kill. You'd likely only want to deal with these tradeoffs in certain shooting situations, or to use one particular favorite lens. Mostly you'd be using native lenses on the X1D or GFX cameras.

    The Hasselblad 907x/CFV50cII is a different animal altogether. The digital back was expressly designed for use with Hasselblad V camera bodies and lenses, similar to the old Phase and Leaf backs from when Hasselblad dominated early digital studio photography (but with modern sensor and features at 1/8th the retail price). It is the perfect digital conversion accessory for diehard V system enthusiasts who can seamlessly switch between film back and digital back using the same familiar V cameras and lenses. While there can be issues with the aging cameras and manual focus that make them lag behind integrated modern digital bodies/lenses in technical perfection, if you love the V experience and can afford $6K the CFV50cII is a win-win.

    Hasselblad evidently decided not to offer the current mkII on its own directly to holdout users of the V system, instead preferring to bundle it with the unusual 907x body in an effort to boost market penetration of the XCD lens mount. When used with native 21mm or 30mm wide lenses, the 907x + CFV50cII provides handling comparable to the legendary SWC film camera (which unfortunately does not work as well with a current digital back vs film). Since lack of ultrawide options is a drawback of the V digital platform, bundling the 907x with CFV50cII at an attractive price point makes it a "free" SWC replacement (while promoting sales of the XCD 21mm and 30mm to complete it).

    Using V lenses on the 907x would require the same adapter as for the X1D. The the waist-level ergonomics of 907x might seem to b an improvement over X1D for V glass, but the very limited body controls of the 907x make it less practical in that mode. And you have the same difficulty of no physical shutter mechanism the X1D entails. Its really a dedicated novelty body for fully coupled XCD lenses: a supplement to to fill in the "digital SWC gap" for V shooters.
    Last edited: Jun 25, 2021
  20. My feelings exactly orsetto, apart from the "SWC factor" which I hadn't thought of. Indeed, if you think about it, the optional grip and viewfinder for the 907X are both strongly remniscent of the SWC. I had no idea that the CFV backs wouldn't work so well on the SWC - my impression was otherwise.

    Anyhow, as I said above, I also think that providing the CFV II 50c as part of the 907X is mostly a kind of invitation to the V-system users for entering the XCD lens ecosystem.

    Another (maybe silly) thought that occurred to me is that the V system cameras coupled with the CFV II 50c, the focusing bellows extension (using, say, the 135mm Makro-Planar lens) and the negative holder accessory for the latter may be an interesting platform for scanning medium-format negatives, similar to the Nikon ES-2 adapter for scanning 35mm negatives. A quick calculation shows that such a scanning rig may achieve almost 2800 dpi of optical scanning resolution with the CFV II 50c (taking into account the 37.5MP resolution of the 32.9mm x 32.9mm square crop of the CFV II 50c sensor and the resulting 0,5875x crop factor as compared to 6x6 = 56mm x 56mm). I know, the 907X alone is more expensive than, say, a 4000-dpi Nikon Super Coolscan 9000 ED, but if you already have the 907X for taking pictures... The focusing bellows extension, the 135mm Makro-Planar and the negative holder accessory can be considerably more affordable in the used market.

Share This Page