907x 50c now available

VBlad said:

I remember the first commercial use Kodak digital camera was priced at $20K if my memory serves right. Its' resolution was about 1MB. So I cross my fingers that I will see a 6x6 format digital back with accepting price in my life (I will be at 60 in three years).
..

Click to expand...

There are two ways for the price of a sensor to come down: 1) the demand is so high that they can make it up in volume (only if it actually MAKES money, don't laugh, this is in contrast with the current "market-leader at all cost, including losses" business model), 2) the manufacturer cost comes down because of better technology or other advancements.

So never says never, but there is not enough demand for the high cost to come down.

As for lower manufacturing cost, that won't happen either. Basically, when the sensor (wafer) gets bigger, the prices do not come down because they are pretty much at the limit of their yield. Imagine they have on average 1 defect per 1 square millimeter (I have no idea what the actual numbers are), that figure will not change much in the future. The larger the sensor, the larger number of average defects. They can mask out the effect of some bad bits, but again, that figure will not change much in the future.

VBlad said:

I remember the first commercial use Kodak digital camera was priced at $20K if my memory serves right. Its' resolution was about 1MB. So I cross my fingers that I will see a 6x6 format digital back with accepting price in my life (I will be at 60 in three years).

Click to expand...

I do not think that Hasselblad V lenses are precise enough for digital. My last kit was 503cx, 500c/m, 903 SWC/M, 50 CFE, 100CF, 180CF and my Leica S images blow them away - both in tonality and detail.....

john

John_McMaster said:

I do not think that Hasselblad V lenses are precise enough for digital. My last kit was 503cx, 500c/m, 903 SWC/M, 50 CFE, 100CF, 180CF and my Leica S images blow them away - both in tonality and detail.....

Click to expand...

What do you mean "precise enough"? I'm getting some amazing detail and tonal qualities out of my C Planar 80mm f/2.8 and CF Makro-Planar 120mm f/4 lenses on the CFVII 50c sensor. Cropping the 907x files to a 6000x4000 FF pixel space, the same photos look virtually identical to similar test photos I made with the Leica SL + SL24-90mm lens but have additional tonality.

I'd have to see some careful test images to make a judgment so grand as to say that the performance of the Leica S "blows them away." I'm sure the Leica S lenses and bodies are superb quality, but that seems too much of a generalization to be considered objective.

G

Sorry I was talking about film, sold my 'Blad gear a long while ago (except some sentimental stuff) so cannot compare with digital backs. But I was going by fine grain monochrome output which is what I used.

john

Godfrey said:

I can (and do) the same, in addition to obtaining digital image capture with the CFVII 50c, although my current scanning solution is to use the 907x fitted with a lens adapter and the Leitz Macro-Elmarit-R 60mm f/2.8 macro lens as the capture instrument. The resulting 39 Mpixel file has enough quality to satisfy my aesthetic and practical needs, and nets the original qualities of the Hasselblad lenses imaged onto film as they were originally intended. The 16-bit raw files from the CFVII 50c provide even better tonal range and adjustment capabilities than my old Nikon Super Coolscan 9000ED, although not quite as much pixel resolution.

I doubt I'll see a 56x56mm sensor in a digital V system back at under $10,000 standalone price anytime soon. Maybe within my lifetime if such a niche product becomes popular enough ... I'm 66 next week, and I suspect I need it to happen within about 20 years. Could happen....! Of course, the question of whether I'm still in the market for a high end digital camera like that at that time is a bigger question mark! LOL!

G

Click to expand...

Curiously nobody thinks that we could potentially be able to put two sensors next to each other and achieve bigger sizes by just doing so.
Would that be so difficult? I don't know.
Most probably we would miss some pixel columns (or rows) in between the two sensors, which must be somehow reconstructed by interpolation. But it seems doable to me.

John_McMaster said:

I do not think that Hasselblad V lenses are precise enough for digital. My last kit was 503cx, 500c/m, 903 SWC/M, 50 CFE, 100CF, 180CF and my Leica S images blow them away - both in tonality and detail.....

john

Click to expand...

Sharpness, tonality, etc are really personal tastes by my opinion if you are not selling your art for others' taste.

Several years back, I already suspected hassy glass might not be up to full frame sensor's challenges. At that time, I had a chance to meet a Hasselblad representative. I asked him if Hasselblad would produce a full frame digital back for V system in the future. He was elusive for answering my question. Nevertheless, who would thought where we are now after 30 years from a merely 1MB digital camera with a price tag of $20K? I believe the technology, and I want to think outside the box - my two cents.

VBlad said:

Sharpness, tonality, etc are really personal tastes by my opinion if you are not selling your art for others' taste.

Several years back, I already suspected hassy glass might not be up to full frame sensor's challenges. At that time, I had a chance to meet a Hasselblad representative. I asked him if Hasselblad would produce a full frame digital back for V system in the future. He was elusive for answering my question. Nevertheless, who would thought where we are now after 30 years from a merely 1MB digital camera with a price tag of $20K? I believe the technology, and I want to think outside the box - my two cents.

Click to expand...

From testing I've done and other test data I've seen, the only V Zeiss lenses that can handle the resolution of the new(er) sensors would be the 250/5.6 Sonnar SuperAchromat, 350/5.6 Tele-Superachromat, and the 300/2 APO TPP. Well, that and the 105/4.3 UV Sonnar. That was my favorite lens, so many really neat things that could be done with enough time...

The next closest would be the 180/4 CFE. It is sharp enough to hold it's own on the newer sensors stopped down a hair and with patience.

I had been hoping that the H cameras were going to get AF Zeiss lenses when they first came out, but we can thank Sony for screwing that one up. The rumour was that there was/is an exclusivity contract for Zeiss AF that prevented anybody from getting them except for Sony.

Actually some of these old Hasselblad/Zeiss lenses hold up remarkably well on the modern sensors, with the Superachromats being superb. From the published MTF data (and my experience using it on a 60 Mp sensor) the 40 IF CFE may be the sharpest of the lot. In my experience the 100 and 180 more than hold their own at 100 Mp and I would expect the same at 150 Mp, which represents only 22% increase in resolution. The 100 is not only crazy sharp across the field but shows minimal distortion and also gives a rendering that I can only describe as "that Zeiss magic." Ditto for the 120 S-Planar for macros. Where they fall short of the Superachromat (I have the 250) isn't so much in sharpness as in correction for longitudinal chromatic aberration, which can be a minor or insignificant issue depending on the conditions of the light. I'm happy to keep these old lenses in the bag next to my Rodies and SK 120 ASPH. YMMV, of course...

John

jng said:

Actually some of these old Hasselblad/Zeiss lenses hold up remarkably well on the modern sensors, with the Superachromats being superb. From the published MTF data (and my experience using it on a 60 Mp sensor) the 40 IF CFE may be the sharpest of the lot. In my experience the 100 and 180 more than hold their own at 100 Mp and I would expect the same at 150 Mp, which represents only 22% increase in resolution. The 100 is not only crazy sharp across the field but shows minimal distortion and also gives a rendering that I can only describe as "that Zeiss magic." Ditto for the 120 S-Planar for macros. Where they fall short of the Superachromat (I have the 250) isn't so much in sharpness as in correction for longitudinal chromatic aberration, which can be a minor or insignificant issue depending on the conditions of the light. I'm happy to keep these old lenses in the bag next to my Rodies and SK 120 ASPH. YMMV, of course...

John

Click to expand...

The 100 is a great lens, definitely with "Zeiss Magic", especially when the conditions are right. The HC100 is also very similar. I've owned a bunch of CF100s as I never liked/owned the 80 (well, I had one for a few days and got rid of it). However, I actually prefer the 110/2 FE more, as I'm more of a fan of the 203/205 than I am the 503 (and I've had a pile of 503CXi/CW bodies) and the 110 is just unique in it's rendering of images. The 180 is still my favorite long lens, with the 105 holding a special place especially when the conditions are right and the proper media is being exposed.

The SA lenses are just simply beautiful with great microcontrast and colors. Yes, they aren't the sharpest, but CA also affects the perception of sharpness in an image, so the "perceived sharpness" of the images is much greater. Along with my MF gear, I keep a 1DX kit as my "speed" small format and the Otus' are my favorite lenses. They remind me a lot of the 250SA in drawing, albeit much wider (the 55 was my favorite).

The 40IF is probably *the* sharpest V Zeiss, but it's also the newest and most advanced so one would hope that it has sharpness improvements. Yet, I prefer the drawing of the Biogons instead of the retrofocus design. Funny thing is that my overall favorite V lens is the 50CFi FLE; that lens and the 110/2 lived on my 203 for a lot of trips and hikes (when it's for fun, not production work).

As far as macro goes, I used the 120 extensively in production along with the 135 S-Planar, but my favorite is still the 63 or 100 Luminar on a vario-tube (or even bellows)!

docholliday said:

The 40IF is probably *the* sharpest V Zeiss, but it's also the newest and most advanced so one would hope that it has sharpness improvements. Yet, I prefer the drawing of the Biogons instead of the retrofocus design.

Click to expand...

Agreed! If only there were a way to use my old SWC on the full frame BSI sensor. I may have to settle for the BSI crop sensor if Hasselblad ever releases the 100 Mp back, but the smaller format really defeats the purpose of using that amazing 38mm Biogon.

John_McMaster said:

Sorry I was talking about film, sold my 'Blad gear a long while ago (except some sentimental stuff) so cannot compare with digital backs. But I was going by fine grain monochrome output which is what I used.

john

Click to expand...

I don't feel, after trying for many years, that it is possible to rank performance across film and digital to the required level of specificity such that the ranking is credible without doing actual side by side testing. As example, some of my ancient Leica lenses produce remarkable results on the latest 47 Mpixel Leica bodies, even though by rights and design meme they shouldn't at all, and compete well with the very latest lenses on some axes of performance.

Film and digital capture, when at the limits of performance with gear at the level of Leica and Hasselblad, are simply such different recording media that making broad generalizations about lens performance is just as likely to be incorrect as correct. I have to say that the XCD 21 and 45 lenses I purchased with the 907x SE prove to have staggeringly good performance and are very easy to work with, given access to the full functionality of the camera. BUT, even with that, comparing images taken with my up to 52 year old lenses makes me wonder at how some lens designers and engineers could possibly have made them work so well on these modern sensors, without the first idea at that time of the nature of these sensors.

When it comes to camera equipment, we live in an extraordinary moment in time with gear that operates at such a high level of performance that it is nearly magical.

G

mristuccia said:

Curiously nobody thinks that we could potentially be able to put two sensors next to each other and achieve bigger sizes by just doing so.
Would that be so difficult? I don't know.
Most probably we would miss some pixel columns (or rows) in between the two sensors, which must be somehow reconstructed by interpolation. But it seems doable to me.

Click to expand...

I'm pretty sure that many engineers have considered this kind of digital capture system and that experiments in this direction have been performed. I'm pretty sure that some of those experiments have been used to construct industrial and scientific digital cameras for certain kinds of uses. But any such construction potentially has a large set of things which can go wrong, and so tends to be costly and difficult to build successfully in any consumer-priced device.

Who knows? It could happen. We can speculate on such things as much as we want. But at the end of the day I think it's more useful to consider what IS being made and how to use it to best advantage, and leave the pondering of future designs to the designers and engineers who have more insight into the creation of the technology.

G

mristuccia said:

Curiously nobody thinks that we could potentially be able to put two sensors next to each other and achieve bigger sizes by just doing so.
Would that be so difficult? I don't know.
Most probably we would miss some pixel columns (or rows) in between the two sensors, which must be somehow reconstructed by interpolation. But it seems doable to me.

Click to expand...

I believe that the old CCD sensors where essentially stitched together in some way, but I’m not sure if they are ever separate chips vs. just aligning the mask when making the chip in order to cut out a larger than the actual mask sized chip in the end.

I’ve though about the possibility of using deep learning to inpaint the space between two separate physical chips, but that would likely not really hold up beyond Instagram resolution. I think what might be a more interesting approach is use some optical trickery to split the image to to separate halves or quarters to be sent to smaller sensors. Maybe something like in figure 3a from this: https://www.photometrics.com/products/imaging-splitters/introduction-to-splitters
It wouldn’t be small though.

John_McMaster said:

I do not think that Hasselblad V lenses are precise enough for digital. My last kit was 503cx, 500c/m, 903 SWC/M, 50 CFE, 100CF, 180CF and my Leica S images blow them away - both in tonality and detail.....

john

Click to expand...

Hm the MTF curves of the 100CF are not really much worse than those of the Summicron-S 100 (or Hasselblad H lenses) - quite impressive actually for a lens designed in the 1960s (or 50s even?)

The 60CFi, 180CF and 350 CFE also look pretty competitive in terms of performance.

You can find the MTF data for the V-lenses here: http://www.hasselbladhistorical.info/HW/HWLds.aspx

The XCD lenses (and probably Fuji GF) are the ones that blow them away with their insane MTFs. But the older designs still seem „good enough“ for digital when stopped down

Without having done any scientific tests and having neither the newest digital back nor the Leica S system for comparison, I still am pretty happy with the results from the V system. The tech cam lenses seem to be sharper, but the V system is very fast to operate compared to a tech cam. Different tools for different purposes, so I have both. My maximum print size is 80x80cm.

From the lens line up, in my opinion there are no real 'bad' lenses in it. I upgraded the few which left room for improvement, notably the CF 40mm to 40mm IF (great lens!), and 250mm to 250mm SA.

Lenses I particularly like are the 40mm IF, 60mm, 100mm, 180mm, 250mm SA, 350mm SA. And the SWC/M, even though the LCC can be annoying at times.

PS: docholliday, I'm envious to your 105mm lens.
PPS: the luminar's were available for the V mount?

Godfrey said:

... and leave the pondering of future designs to the designers and engineers who have more insight into the creation of the technology.
G

Click to expand...

You know what? The problem of leaving such a pondering to the typical designers and engineers is that they are more constrained by financial revenues than by technical limitations.
A niche like the one we're talking about here will never be considered worthy for an investment.
We'd rather need a Kickstarter project made by some crazy and passionate people, like us. ,)

John_McMaster said:

I do not think that Hasselblad V lenses are precise enough for digital. My last kit was 503cx, 500c/m, 903 SWC/M, 50 CFE, 100CF, 180CF and my Leica S images blow them away - both in tonality and detail.....

john

Click to expand...

Not sure i understand this comparison, V lenses were made to resolve detail for film, they were never meant to resolve digital sensors...some hold up better then others, i generally find (most film lenses) have a certain look...it might not always have to be about sharpness and detail....
The Leica S system is hampered by 2 things, mediocre sensors and expensive lenses....the lenses might be great, although there are plenty of much less expensive lenses out there with similar technical excellence, maybe its the leica glow you are paying for...either way, for most professionals it does not make sense to spend that much on glass if the sensor is below standard...
The fuji and hasselblad systems show that you can have a great sensor and amazing glass for a lot less and both systems are much more flexible as well

buildbot said:

I believe that the old CCD sensors where essentially stitched together in some way, but I’m not sure if they are ever separate chips vs. just aligning the mask when making the chip in order to cut out a larger than the actual mask sized chip in the end.

I’ve though about the possibility of using deep learning to inpaint the space between two separate physical chips, but that would likely not really hold up beyond Instagram resolution. I think what might be a more interesting approach is use some optical trickery to split the image to to separate halves or quarters to be sent to smaller sensors. Maybe something like in figure 3a from this: https://www.photometrics.com/products/imaging-splitters/introduction-to-splitters
It wouldn’t be small though.

Click to expand...

It can be done...Arri has done it on a few models , notable the Alexa LF line and Alexa 65 line. To achieve larger sensor sizes and higher resolutions, they somehow managed to “stitch together” 2 or 3 of the existing sensors used in the rest of the Alexa line. Source.

anyone said:

Without having done any scientific tests and having neither the newest digital back nor the Leica S system for comparison, I still am pretty happy with the results from the V system. The tech cam lenses seem to be sharper, but the V system is very fast to operate compared to a tech cam. Different tools for different purposes, so I have both. My maximum print size is 80x80cm.

From the lens line up, in my opinion there are no real 'bad' lenses in it. I upgraded the few which left room for improvement, notably the CF 40mm to 40mm IF (great lens!), and 250mm to 250mm SA.

Lenses I particularly like are the 40mm IF, 60mm, 100mm, 180mm, 250mm SA, 350mm SA. And the SWC/M, even though the LCC can be annoying at times.

PS: docholliday, I'm envious to your 105mm lens.
PPS: the luminar's were available for the V mount?

Click to expand...

I don't have the 105 now...I had it for a while but sold it when I was thinning out my CF/i/e lenses. I didn't get to use it as often as I'd liked. The lighting had to be UV-strobe or visible light in the perfect conditions. It wasn't an expensive (though kinda-rare) lens when I had mine and I do miss 'er nowadays.

I think the universally considered "bad" lens in the V line was the 50 non-FLE. Not much love for that lens even though it wasn't *that* bad. Funny that you like the 60, as I found it and the 80 to be my least liked lenses. The 60 was not very dimensional to me. And then there's the 160CB that seems to be completely forgotten about...

My base set of lenses were 50CFi/FLE, 100CFi, 110FE, 120CFe/Makro, 150FE, 180CFe, 250CFe/SA, 63 Luminar, 100 Luminar. I never owned any 500/501 bodies but I had a pile of 503 CXi/CW and 203/205 bodies.

The Luminars were adapted using the 40037 RMS adapter or via another adapter V mount. You could focus it using the Vario-tube or a set of bellows. No shutter on 50x bodies, but I used them with 203/205 bodies with Vario-tube mostly. I've also used them on the 503 locking the shutter open with a cable release in a dark studio and firing the strobe manually. There's also a few adaptations to a Compur shutter and then adapting the shutter to V mount for the front of bellows.

docholliday said:

...
I think the universally considered "bad" lens in the V line was the 50 non-FLE. Not much love for that lens even though it wasn't *that* bad.
...

Click to expand...

Very amusing ... I presume you mean the Distagon C 50mm f/4 lens? I've had it for a while and was a little perplexed when I got it, it looked like new, and the seller said, "Eh, I gave you a really good price because no one wants this lens, it seems..." Then I went out and shot two rolls of film with it and found it made beautifully detailed, tonally rich negatives ... And on the 907x with the adapter tube when I focus it using the LCD screen and focus magnification, it produces very sharp and contrasty results, corner to corner.

What's not to like about it?

G