Weighing my Options for Hasselblad CFV-39 (Need Help)

Discussion in 'Medium Format' started by timlayton, Aug 6, 2011.

  1. Over the last year as I have been building out my 503CW system I have asked questions here and always received helpful and quality responses based on experience. I need your help in making sure I am considering all of the factors when thinking about getting a CFV-39 digital back. First, this is the equipment that I currently have: 503CW, Acute-Matt D screen, PME-45, WLF, 40mm CF, 50mm CFi, 80mm CF, 150mm CF, 250mm CF and two A12 backs. My style of photography is mostly landscape and nature fine art. I shoot black and white the majority of the time. I still do a lot in the darkroom and plan to continue that independent of any digital back. My workflow for digital is a scan of my negatives that are brought into Lightroom for cataloging and minor edits and Photoshop for any other edits.
    A new CFV-39 is about $14,000 US and that is a lot of money to me. I want to make sure I am thinking clearly before making this type of investment.
    Here are my current thoughts and questions at this point. Please add or correct as needed:
    1.) Is it reasonable to assume the CFV-39 will be viable 5 years from now? I have been using my film equipment for many years (in some cases 25+ years) without ever requiring an update or upgrade. I am concerned about spending $14k and not being able to get service or repair in 5 years. I am not worried about the quality issue because I am sure the CVF-39 will meet my quality and artistic standards indefinitely.
    2.) I use Lightroom and would like to leverage the 3FR and lens correction support there vs having to use Phocus software. Any experience or comments on this? If you recommend Phocus then what format do export from Phocus to Lightroom?
    3.) I am not replacing my film work, just adding digital as a medium. From my understanding there are no adapters required for the cfv-39 on the 500 series so I am assuming it is a reasonable idea to swap the digital back and A12 backs between exposures without too much of a hassle. Anyone with experience confirm my thinking?
    4.) According to the specifications there is a maximum of 64 second exposure for the cfv-39. That is a limitation for some of my work and will preclude me from being about to use the digital back for that. Does anyone have experience with long exposures on the cfv-39 and if so what do you find the practical limitations to be? I assume noise and quality issues will be the problem at some point.
    5.) Based on experience what do you find the optimal ISO range of the cfv-39? I know what the specifications say, but I was hoping to get some input based on real-world experience. Keep in mind I mostly photograph landscapes and nature.
    6.) Last, but not least, for $14,000 should I even try and make digital work on my 500 series and just consider another option? My initial thought is to leverage my current equipment as a base and continue to use it with my film work. I am not opposed to pursuing another system or option as long as it is not more money because the $14k is actually more than I want to go at this point. I would say that the vast majority of my work is wide and ultra wide angle if that matters in your opinion when considering cameras and lenses.
    If there are other points I should consider please add them here. I appreciate your time and thoughts.
    Tim
     
  2. The thing that makes the back so viable is that it is 16bit, not the 14bit you find in high end DSLRs. That's a heck of a difference in real life use. I think the CFV-39 is a tremendous value, when you consider its size and the ability to couple it with the rest of the V system.
    As for service and repair, Hasselblad just recently changed hands again. But I think the digital equipment is relatively safe in the ability to be repaired. I use digital H systems from time to time (renting them), and often think about buying the CFV. For that much money, it could be obsolete in 5 years. While I would love to buy one, I don't think I will make enough use of it to make it pay for itself (at least in comparison to shooting film and scanning it). So I compromised by sticking with film, and buying a higher end, drum scanner.
    I have borrowed a friends CFV-39 and love it. The images are breathtaking, especially for color, and the noise is well controlled. The Hasselblad Phocus software is very good at getting images ready for manipulation in Photoshop/Lightroom. I have not shot exposures as long as you are talking about. The longest was about 4 seconds, and it looked great.
    As for optimum ISO, I usually leave everything at 100 because it makes calculations pretty easy. But then I was using it for product shots, not landscapes, and I was using Broncolor lighting with tons of power. As for swapping backs, I was using my 555ELD camera with bus contacts, so you should have a similar scenario with your 503CW. Removing it is easy, but you have to take care in storing it. That's where your biggest time would be lost. But yes, it is pretty quick.
    As for $14k: You could buy a Pentax or Mamiya, with less MP (I believe), but with all your lenses, you would have a hard time replacing them with the other systems, and I do believe that Zeiss lenses have a character that is hard to replace at all. Most other systems in the 39mp range are far more money anyway.
    To me, it comes down to what are you going to get for your money, and how long would it take to pay off. I believe I would have a hard time justifying the money, as most of my personal work is done with V system, and professional work is either with an H system or 4x5. But I have considered whether to stop renting H systems, and simply start using a V system and CFV back to ultimately save money. It's just that I prefer the ergonomics of the H. If $14k isn't that big of deal to you, I'd go for it. I would go for it if the economy was better, and I was willing to throw fate into the wind, but I've been on a Dave Ramsey reduction plan in recent years, and rather enjoy not having any debt.
     
  3. Tim:
    You have good questions here and I can answer a few of them.
    1) No problem. I just brought a 7 year old Hasselblad back in for service and when I mentioned that I planned to get another five or ten years out of it, the rep said "no problem." Indeed, there were older backs being sent in.
    3) Digital backs attach to the V Series bodies with a setscrew, so you'll need to carry the special screwdriver with you. Otherwise, no problem.
    4) I suspect this number is optimistic and would recommend sticking with shorter times. Older Phase 1 backs (which can be bought for relatively low prices second hand from dealers) claim to allow very long exposures. Check this out for yourself before plunking down money.
    6) If you're presently scanning film, you're going to be quite surprised by the quality of the back. Of course, even though the files are fantastic, they don't look like film - not better or worse, just different.
    I suspect other people will have answers for the rest of your questions.
     
  4. Tim, I can give an opinion on a couple of your points.
    Many have tried to not use Phocus but, in the end, it is the way to get the best from your digital images. I export as PSD from Phocus while lots of people use TIF. Either way, you get the best out of the image with all corrections. Phocus now has dust removal which can be batch applied.
    If you want long exposures as I do, then the current Hasselblad star is the H4D40 at 256 seconds. I had an H3D which I think is the same chip as the CFV and it is exceptional at 64 seconds as is the H4D at 256.
    I would consider an H4D and the CF adapter but most people who go that way end up dumping their V lenses and going for the H lenses but it is something to consider.
    I would also suggest hasselbladdigitalforum.com There are a lot of folks there with more knowledge on the CFV than me.
     
  5. 3) Digital backs attach to the V Series bodies with a setscrew, so you'll need to carry the special screwdriver with you. Otherwise, no problem.​
    Hmmm... Don't know where that idea came from, but it's not true.
    The CFV-39 attaches to the body exactly the same way as an A12 magazine. You can easily change from film to digital and back whenever you wish, with no issues or aggravation.
    There is an additional interlock button that you must press in order to release the back, to prevent it from coming off if you bump the release button, but it requires no tools.
    ----
    A strong +++++ for the Phocus program. It's very intuitive, easy to use, with a wide assortment of features. It even has optical correction built in for all the 500-series lenses. Download for free from Hasselblad at www.hasselbladusa.com and get free updates.
    I have Lightroom, but would never consider using it with the CFV-39. I also have PS/CS5 but almost never use it.
    ----
    Do you need long exposures due to film speed or because of subject issues? You can always user a higher film speed. If there's an exposure issue I just use more light. :D
    ----
    I always shoot at ASA 100. That's the same speed I use for film in 120, 4x5, and 8x10, so I'm accustomed to all the issues.
    ----
    My decision to add the CFV-39 was based on the fact that I have a full set of V-series lenses, from 40mm through 500mm, and I think they're the best optics available. That's why I started shooting Hasselblad over 40 years ago, and nothing has happened since to change my opinion.
    - Leigh
     
  6. david_henderson

    david_henderson www.photography001.com

    Here's what I wrote here three years ago on this before giving up on the idea and buying a lower pixel count but more flexible dslr system instead
    "I've been seriously considering the CFV as an alternative to my film MF system. The things that I really can't get round are the fact of a 1.5 crop that makes wide angle close to impossible; the tiny lcd screen no bigger than that on my 10D; and the fact that the MF composition process involving big, expansive views through the finder is emasculated by the fact you don't with the CFV get all you see on the finder. Personally i can't function without a wide angle capability and the prospect of a film/scan for wide, digital for everything else route to be unacceptably clumsy.
    Added to which is a very definite impression created by the people at the Hasselblad London showroom, that their heart is in the H series not the V series with all sorts of mutterings about the V series lenses "not being optimised for digital" and support for the V series not being endless. In short, despite being told repeatedly that I wasn't interested in a H series they seemed quite content to prejudice the possibility of selling a decent used V system plus a back costing several thousand pounds on the off chance of an epiphany about the H series. If the V system/CFV is a long way down their pecking order I can't help feeling that it should be a long way down mine too."
    If as you say most of your work is wide or ultra-wide then how are you going to get that when the widest lens you own gives you the field of view of a 60mm?
    I know it seems neat and tidy to have just one system handling your needs, but it can onlt be effective if you can photograph how you want. Seems to me that you have such a system now.
     
  7. David,
    I hope you'll pardon a blunt reply, but your entire post is pure nonsense, apparently based on a combination of spec sheet review and comments from a sales staff that makes most of their commission selling new lenses.
    Wide angle is certainly not impossible. I use the 40mm and 50mm lenses just like on the 500C/M for that style. The digital image is very slightly narrower than the film image, not enough to even notice.
    What does the LCD size have to do with anything? You still focus and compose using the regular ground glass and viewing system on the camera body. The LCD is just for image review and histogram assessment, if you wish.
    I think Hasselblad desrrves a strong round of applause for suporting their legacy systems with this product. Most mamufacturers would just tell owners of their older products to toss them and buy all new stuff. After all, they don't make a whole lot of money selling a single digital back to somebody who already owns all the lenses and accessories, then giving away the excellent support software for free.
    - Leigh
     
  8. I agree on the applause for Hasselblad. Phase makes a lot more noise but I never regret that I am shooting with the best MF body, back and lenses which work as one together with software to complete the workflow. It's a superb combination.
     
  9. I think Hasselblad desrrves a strong round of applause for suporting their legacy systems with this product.​
    Absolutely. Especially considering that various owners could have neglected the V system users altogether (though to a degree they did).
    Also, I believe the newest backs are something like 50mm square, which is pretty darn close to full frame film.
    As for the release mechanism, I believe they can be screwed in, but the couple that I've worked with have had an adapter plate (maybe just a mounting plate) that allowed them to be easily removed.
     
  10. As for the release mechanism, I believe they can be screwed in, but the couple that I've worked with have had an adapter plate (maybe just a mounting plate) that allowed them to be easily removed.​
    I don't understand where these comments are coming from.
    The CFV-39 has a standard V-series attachment, identical to that on all the film magazines for that camera.
    You mount and remove the back exactly the same as a film magazine, no differences except for the safety interlock that I mentioned previously.
    There may be adapters or whatever available for the CF-39, which is a different product.
    - Leigh
     
  11. That's my understanding Leigh. I had a V96C which came with a V series adapter, just the same as a film back. Anything with V in it means direct connection as I understand it, so any CFV will attach to any 5 series body.
     
  12. I appreciate the input and comments from everyone so far. Some very good points that I will add to my list:
    CFV-39 is 16-bit vs. 12 or 14 bit in DSLR systems.
    I should visit hasselbladdigitalforum.com for more info.
    The images from film and digital do indeed look different so that was a good reminder. I need to really think about that aspect.
    I will assume that Phocus is the best route to go for optimum images from the back.
    Thanks so far and hopefully I will get some more replies.
    Tim
     
  13. Tim, you may be a lot different to me but, once I started with a digital back, it was the end of film for me. I was never a great lover of film, and couldn't wax eloquent about grain etc. To me, it was a PITA.
    It's very difficult to assess but you may find that your love of film is shallower than you think.
     
  14. I am not sure I want to jump into this whole thing about P1 vs Hasselblad, I guess I would say, having used a 503CW for a good while with a digital back, I think it can be a great system for landscapes, I had trouble focusing on people with it quickly and ended up switching to an H system after using the V for many years. mostly for the autofocus, but I miss the 500 series for doing landscape stuff, just easier to carry around. You might find that you are not comfortable taking the digital back on and off like switching film backs, there is always the possibility of scratching the uv glass on the sensor,dropping it and or getting dirt inside it. It does not have a dark slide and you have to fumble around with the protective cover. With any 39mxp sensor back, you will have a slight loss on the wide angle side of things but not too much, capture integration has a lens angle calculator on their website that will give you an idea. The P1 backs do have a better long exposure capability at least the 39mxp 45+, but you also have to use a cable to the sync post on the lens which is a total pain as it seems to come loose and is in the way a lot. The sensor in the P45+ and CV39 are basically the same, it has been around for about 4 years now. They both would do a great job, I have almost 50,000 captures on mine, there are no moving parts. Some of the newer model backs are supposed have not only more megapixels but much more dynamic range, and the newer sensors go edge to edge on the width of the frame same as film. But they are really expensive, about 2-3 times what you are looking at.
     
  15. Well I don't have an answer for you, but one might take issue with some of what's been said, and you may want to consider what follows.
    The things that I really can't get round are the fact of a 1.5 crop that makes wide angle close to impossible.
    That's an important difference between the older CFV he was discussing and the current CFV-39 you're considering. The Older CFV's had 36.7 x 36.7 mm sensors, which is a 1.53x crop factor relative to 6x6. However, the CFV-39 has a 36.7 x 49.1 mm sensor, so if you're printing US-standard larger print sizes with a 4:5 aspect ratio (8x10, 16x20, and 24x30; and 11x14 and 20x24 are close), then the CFV-39's effective crop factor is only 1.22x. So if you normally use a 60mm, with the CFV-39 you'd need to use a 50mm, and if you normally use a 50mm, with the CFV-39 you'd need to use a 40mm. (Now if you normally use a 40mm or 38mm, I suppose it isn't the solution for you.)
    16-bit vs. 12 or 14 bit in DSLR systems
    I would not get too hung up on that--people confuse bit depth with dynamic range. Dynamic range is how far you can go (from dark to light), and bit depth is how precise / small you can make each step of your travel (tonal gradations). The dynamic range captured is independent of the bit depth. One important thing that the CFV-39 should give you over a typical DSLR is the ability to capture scenes with more dynamic range (greater range of brightnesses) without the highlights blowing out white and/or the shadows fading to black. But what 16 bits realistically gives you over 12 or 14 bits is only somewhat more ability to make radical changes in the contrast and/or color without seeing artifacts like banding. (Theoretically it gives you slightly more subtle / delicate / detailed tonality, but I bet you can't see that in any print.) Frankly if you're not making sizeable changes in color or contrast, 8 bits is usually okay, and 10 or 12 bits can cover all but the most radical / demanding changes. Is 16 bits a real advantage over 14 or even 12 bits? Once in a great while, probably so, but not often for the way most people work.
     
  16. Let's debunk this 1.5 crop factor nonsense once and for all.
    That number comes from 35mm cameras, not the CFV-39.
    The factor for the CFV-39 is 1.12, based on its 49mm sensor width.
    - Leigh
     
  17. I have two questions please.
    1. The difference in price between a CFV-39 and CFV-50 is around Eu 3000. Is the differece worth it? Apart from higher resolution what are the benefits of the CFV-50?
    2. Any solutions for using the CFV backs in portrait mode? (I believe Phase One and Leaf can be rotated and used in both landscape and portrait).
    Your valuable comment will be much appreciated.
     
  18. Let's also debunk this 16 bit nonsense once and for all.
    Dave is basically right, but he hasn't told the full story. To do that you have to take the sensor specifications into account. The KAF-39000 CCD in the CFV-39 has a full well depth of 60,000 electrons, so 16 bits output means an ADC gain of 0.9 electrons/ADU. The KAF-39000 has a total per pixel internal noise of 21 electrons rms, which, in 16 bits, is +/- 19 ADU rms. This means that even before we consider any other noise sources (signal and dark shot), every real intensity is scrambled in a gaussian distribution with a characteristic width of 38 ADU. This is a crazy degree of redundant intensity oversampling.
    If the CFV-39 worked in 14 bits, the gain would be 3.7 electrons/ADU, which is still sampling the noise floor by an overly (but not stupidly) generous factor of 5.7. You could torture the data as much as you want and still not see banding.
    There is not a single MF back since the Leaf Volare/Cantare which has had the full well depth to readnoise ratio to warrant 16 bits.
     
  19. Apart from higher resolution what are the benefits of the CFV-50?​
    Carlo, the CFV-50 is a later design generation of chip, based on a 6 micron process rather than a 6.8 micron process. Readnoise per pixel in the KAF-50100 is marginally lower than the KAF-39000 (13 electrons rms as opposed to 16), although the readnoise per unit area is the same. The real benefit however comes in the dark current, which is around 4x better (depending on temperature) in the KAF-50100. This matters if you take long exposures.
    Apart from that, there is little difference in the sensors. The KAF-39000 has slightly (0.2 stops) higher dynamic range, the KAF-50100 has slightly higher quantum efficiency.
    There may be features in the back interface which are better in the CFV-50, but I am not addressing those aspects.
     
  20. "Apart from that, there is little difference in the sensors."

    There's also that thingy about one resolving the image presented to its 36.7 x 49.1 mm large area into 39 MP, the other into 50 MP. ;-)
     
  21. Q.G. -
    There's also that thingy about one resolving the image presented to its 36.7 x 49.1 mm large area into 39 MP, the other into 50 MP. ;-)​
    Did you actually read the question which I was answering? I quoted it, after all. Looks like I'll have to quote it again! Carlo asked:
    Apart from higher resolution what are the benefits of the CFV-50?​
    ...so of course I did not address the 39 MP vs 50 MP difference. He told us that he already knew about that.
     
  22. Thank you for the answers so far. For example, I did not know it was a new generation chip. Because it is physically the same size I assumed it was the same chip with maybe different software implementation. One question is, is it worth the extra cost?
     
  23. "Worth the extra cost?" is a much harder question to answer - there is no physical formula for "worth"!
    I guess it is worth it if you do a lot of long exposures. Otherwise, I'm not so sure. I expect that 50 MP is not that different to 39 MP really unless you print very, very large, and are consistently good at nailing focus. Hopefully an actual CVF-50 user can chip in here.
     
  24. If you shoot landscapes and fine art mainly B&W, what benefits are you expecting by adding a 39mp back to your film workflow?
    Less resolution, dynamic range, spectrum capture options, huge cropped format (i.e. film captures 75% more than the digital back of the lens projected area), etc.
    I suggest you compare a 39mp capture side by side to a film scan of the same landscape first, if you havent yet, especially on b&w.
    Regarding long-term support, Hasselblad is top notch. That said, you can use this life cycle advantage and get a new scanner from them for the same money. Films will continue to be upgraded for you for free.
     
  25. I agree wholeheartedly about Hasselblad support. It's simply excellent. That's where my agreement ends,
    I have also done the comparison that you suggest and am now shooting with an H4D 40 and am very happy. I have no interest in a film vs digital debate, digital simply works better for me.
    You mention dynamic range so I assume that you mean colour neg. No argument there but what a limited palette you get and grain to sand walls with.
    As for size, I used to get about a 280MB scan with film on a 503CW, now I get a 240MB file from the H4D. There is no competition, the digital image can be poked, prodded and resized and still come up smiling. Try that on film.
    I also wouldn't count on film being upgraded forever, either.
    My portfolio is full of both film and digital images her on PN. I don't think that my work has suffered or been limited by going digital, but that's just me.
     
  26. Mauro asked "If you shoot landscapes and fine art mainly B&W, what benefits are you expecting by adding a 39mp back to your film workflow?"
    Mauro, I am not expecting any benefits per say, other than having digital as a second medium to my main staple which is and will always be film. One of the things that I wanted to explore was the feasibility of swapping the cfv back and the A12 back between exposures for landscapes. It sounds like as long as I am protective of the cfv back that should be fine. In the back of my mind I feel like I need to at least explore this as an option for the long run.
    I currently swap my A12 backs all the time because I group my exposures by development (N, N+1, etc). Since I already have a large investment in my V series this is the main reason I was exploring the idea of the cvf over any of the other options. And, I absolutely love my V series equipment. I can always rely on it to help me produce fine prints. I suspect I will own and still be using my current V series equipment for the rest of my life.
    The bulk of my fine art work is done in large format but there are times when my old bones just can't get that gear where I need it and my V series has been a great alternative in these cases. There are also times when I have no access to developing my film for long periods of time and it would be nice to be able to preview my scenes although that is more of a secondary thought that anything. I still print all of my fine prints in the darkroom and I have no plans of changing that at this point. The digital back in my mind is just another medium which is different and is not a replacement in my mind.
    My plan is to rent a back if I can and if I can't do that then I know H4 systems are readily available which should give me a good idea of what to expect.
    This has been a good thread with a lot of great information. Definitely a good group here and I appreciate the help and feedback.
    Tim
     
  27. I agree with Tim, great thread and very valuable information. I have a large investment in the V system and the option of adding a CFV would expand my horizons, especially when I spend extended time on another continent where film developing and scanning are not an easy option. Having a digital back there will be a big advantage.
     
  28. The dynamic range captured is independent of the bit depth.​
    Well, digital sensors work in a linear way and because of that, there is a relation between minimum required bit depth to assure you can record the dynamic range of the sensor.
    Minimum bit depth = DR
    So if the sensor is capable of 14 f/stops DR, then you need at least 14 bits bit depth.
    The reverse is not true, increasing bit depth will not increase DR
     
  29. Tim, if you could rent a back not only to test the results but also to bring it to the field and see how you relate to the experience would be great.
    For your overall concern about the back's long term support, I wouldn't worry about it if you find that it brings value to your experience.
    i am a B&W shooter myself and I stopped bringing my DSLR to the field a long time ago. I found I would never pick a digital file for print; and using it as a Polaroid was not needed since I either meter by head or occasionally with a spot meter.
     
  30. Well, digital sensors work in a linear way and because of that, there is a relation between minimum required bit depth to assure you can record the dynamic range of the sensor.
    No. As an extreme example, you can use 2 bits to describe a 15-stop dynamic range. It's just that there will be huge steps from one level of lightness / density to the next, causing coarse tonality / massive banding. More bits means more gradations, which means more subtle / fine / detailed tonality.
    There are 2^[bit depth] gradations. So if you use 2 bits, you get 4 gradations; you can design your system so that the four gradations are black, dark gray, light grey, and white; or very dark gray, dark gray, medium gray, and light gray; or black, medium-light gray, light gray, and white; or whatever. By tweaking the design, you can give relatively more bits to the highlights, midtones, or shadows, and in doing so provide more detailed / subtle tonality in one lightness range at the expese of less detailed tonality in another lightness range.
    If you use 8 bits, there are 256 gradations; if you use 14 bits, there are 16384 gradations; and if you use 16 bits, there are 65536 gradations. (And as pointed out above, at some point, the math can represent more gradations than the sensor can differentiate among.)
     
  31. You can do all that you say in postprocessing, but not at the moment of capture.
    Linear encoding work that way, it is basic digital signal processing theory. If you don't want to believe it, it is fine.
    In linear encoding:
    The brightest (first) f/stop will have half the values
    The second f/stop will have a quarter of the values
    The third f/stop will have an eight of the values
    And so on, until you get to 1 value or your equivalent bit unity, where you cannot go further
    There is a good tutorial: Understanding Dynamic Range in Digital Photography
    Also check this by Norman Koren: Tonal quality and dynamic range in digital cameras
     
  32. Francisco, what you are talking about is not an inherent quality of the digital picture-taking process; all of it assumes (at least) a particular behavior of the A/D converter. If you want to say that maximum analog output from the sensor gets interpreted as full white, and the A/D converter quantifies it as 65535; and one stop darker gets represented as 32767; I get that part. But that linearity is at most a design choice for the A/D converter. (And I suspect the relationship between the sensor's analog electrical output and the A/D converter's output written to the raw file is a little different for the R, G, and B channels of a Bayer sensor.)
    As a practical reality-check on the importance of the bit depth, I'd offer this observation: a regular JPEG is an 8-bit file. Common DSLR's produce 12- or 14-bit raw files. Although the raw files tend to have somewhat more dynamic range than the JPEG's, it's generally less than the difference between 8 and 14 bits. How much of this is due to the least significant bits in the raw file being lower quality, versus other factors, I'm not sure.
     
  33. The brightest (first) f/stop will have half the values​
    But that's equally true with film. That's how photography works.
    What is significant in digital is the quantization of the bit values. The tenth bit represents a change of one one-thousandth (approx) of the total range, regardless of the value of the higher-order bits.
    The most significant aspect of bit depth is noise. In any A/D conversion system the lowest two (or so) bits are considered to be noise, and of little intelligence value.
    So you can expect a 12-bit system to yield 10 usable bits, giving 1024 unique levels. A 14-bit system will have four times that number, or 4096 levels.
    The number of unique levels in a film environment is almost infinite.
    - Leigh
     
  34. But that linearity is at most a design choice for the A/D converter​
    Yes, I was referring to that, which is the common case in consumer digital cameras
    As a practical reality-check on the importance of the bit depth, I'd offer this observation: a regular JPEG is an 8-bit file. Common DSLR's produce 12- or 14-bit raw files. Although the raw files tend to have somewhat more dynamic range than the JPEG's, it's generally less than the difference between 8 and 14 bits. How much of this is due to the least significant bits in the raw file being lower quality, versus other factors, I'm not sure.​
    Here gamma encoding plays an important role, remeber that jpeg are not linear but usually encoded with a gamma of 2.2, which increases the number of levels in the shadows, decreasing the number of levels in the highlights.
    In theory, and this is just from a mathematical excercise, the relation between the possible DR to be represented by a specific bit depth using gamma encoding is DR=Gamma x Bit Depth, so 8 bit x 2.2 = 17,6 f/stops of DR.
    This of course does not happen in practice for several reasons, mainly:
    You don't start with such a high DR
    Color space encoding plays also an important role here, decreasing the DR
    Just as a reference, in one of the previous links I posted, there is a table that compares the number of levels availble for every f/stop or exposure zone. At exposure zone 11, you will have no levels left for 10bit linear RAW, 2 levels left for 12 bit linear RAW and 8 levels left for 14 bits linear RAW. Instead you still have 3 levels for 8 bit at 2.2 gamma and 758 levels for 16 bits at 2.2 gamma
     
  35. Does any of this answer the OP's questions or help his understanding? There seems to be an immutable law of nature that the longer a thread runs the further it digresses from the original question.
     
  36. Jeff, on your basic point--fair enough. The OP had commented on MFDB's having 16-bit depth as an advantage over DSLR's with 12- or 14-bit depth; Francisco and I got into more detail than might have helped the OP about the significance of that bit depth. I guess I will leave it at my view that, although the CFV-39 should capture significantly wider usable dynamic range than a current 35mm-style DSLR, the role of a couple of bits one way or the other (in the A/D converter and raw files) is secondary to various sensor-size and sensor-design issues.
     
  37. Thanks Dave. Without wishing to do what I recently complained of, I often think that most folks who compare MFDB to DSLR have never actually used the MFDB for any length of time. My last DSLR was a Sony A900 with Zeiss glass. I sold it because I was disappointed with the IQ each time that I used it. If you cast a quick eye over my seascapes, in particular, you see a lot of images with lots of tonal wash effects. My H4D delivers this superbly but the Sony just didn't cut it.
    Whether that is bit depth CCD vs CMOS, I neither know nor care. I'm just happy that I get it.
     
  38. My gut reaction is that you would be crazy to spend that kind of money for a digital back. You dont do this for a living, so you cant really look at this as something you will get a return on. That back will depreciate quickly, as digital continues to improve in leaps and bounds. Stick with your current setup, and use that money towards trips you will enjoy and remember forever, with beautiful scenery for getting your landscape shots. If you really need it, either pick up a $500-600 top end compact digital with high quality zoom lens (see the panasonic range that uses Leica lenses, or the Fuji, canon, Nikon equivelants). You will get instant feedback, great results, and much pleasure from such an adjunct to your hasselblad setup, which will still serve the purpose of ultimate quality. And, you wont break the bank and get sucked into the disease of GAS additional lens aquisitions which happen when you get a DSLR. Or, you could get a DSLR and a few lenses for a lot less than $14,000. Maybe more like 7 grand if you go that route.
    Either way, I would drop the idea of dumping so much into a digital back. No offense, just trying to give you an outsiders Reality Check.
     
  39. That back will depreciate quickly, as digital continues to improve in leaps and bounds.​
    The thing is, MF digital has not been improving in leaps and bounds. MF digital back performance has scarcely moved on at all since around 2004. Sure, more megapixels have been squeezed in to some backs, and one manufacturer significantly improved its user interface last year. But there is a frustratingly long list of metrics where there has been very little improvement, in comparison to smaller format digital:
    - readout noise
    - dark noise
    - maximum exposure time (still 1-4 minutes in nearly all cases)
    - frame rate
    - dynamic range (actually shrinking, as megapixel count rises)
    - ISO range (at full resolution)
    - sensor area (still not quite 645, and far short of 6x6, let alone 6x7 or larger)
    - live view/video rate
    - in-camera processing
    So because of this slow progress, and the lack of moving parts that could wear out, MF digital backs actually depreciate a lot slower than other digital cameras, and hold their values quite well.
    To take one illustration: last year I bought a Kodak MF digital back, for 20% of its original price 8 years previously. Compare that to this year when I bought the last-and-best Kodak/Nikon-F5 DSLR, using almost exactly the same (scaled down) sensor technology and user interface, for 4% of its original price 10 years previously!
     
  40. My last DSLR was a Sony A900 with Zeiss glass. I sold it because I was disappointed with the IQ each time that I used it. If you cast a quick eye over my seascapes, in particular, you see a lot of images with lots of tonal wash effects. My H4D delivers this superbly but the Sony just didn't cut it.
    Whether that is bit depth CCD vs CMOS, I neither know nor care.​
    Jeff, I know you don't care, but that won't stop me surmising!
    I believe it is neither due to bit depth, nor to CCD vs. CMOS, but rather, due to filtration (both the spatial and spectral types). We know that the DSLR uses an AA filter, which increases spatial correlation of the signal between pixels. We also know (from DxOMark) that Sony doesn't always play fair with its RAW files, applying some sort of smoothing filter which correlates the noise between pixels, at least at higher ISO. And finally we know (again from DxOMark) that some DSLRs (Canon are most implicated here) use excessively broad-spectrum filters in their Bayer arrays, boosting their ISO sensitivity at the expense of colour discrimination. I give the 2 DXOmark links below.
    These are not properties inherent to bit depth or to the use of CMOS - they are external design decisions.
    http://www.dxomark.com/index.php/Publications/DxOMark-Insights/Half-cooked-RAW/Noise-reduction
    http://www.dxomark.com/index.php/Publications/DxOMark-Insights/Canon-500D-T1i-vs.-Nikon-D5000/Color-blindness-sensor-quality
     
  41. Taking some advice that was given in this thread I reached out to a friend of mine that is a digital photographer. He has a Nikon D3X. I headed out this morning at 5 a.m. for a sunrise landscape with the digital equipment and my 503CW. I used the 50mm CFi on my 503CW and the Nikon 24mm prime lens on the D3X. On the 503CW I shot 6 exposures of Tmax 100 and 6 Velvia 100. On the D3X I took the same 6 exposures. I rated the TMAX at EI 64, Velvia at 100 and the Nikon D3X at ISO 100. The best lighting was from about 30 minutes till sunrise to 12 minutes. I plan on going out again tomorrow morning at the same time but a different location to complete the rest of the TMAX and Velvia exposures. Exposure times today ranged from 4 minutes to 15 seconds. Once I get through these tests I will post again with my observations and comments.
    Tim
     
  42. I wonder what are you trying to test/prove? The cheapest way to high quality images is to use film on your Hasselblad and a reasonably good scanner. At least if you only shoot a relatively small number of images in a year (scanning takes time per each image). If you put the CFV39 on that Hasselblad you will get a better image than with either film, or the D3X. The D3X or Sony A900 or Canon 5D2 would almost certainly be good enough for your needs. But I thought the whole point of your original question was to justify the CFV back that can still make use of all your existing equipment.
     
  43. Ilkka, I am not trying to prove anything. You must be misunderstanding me or maybe I did a poor job at explaining myself. I was simply weighing out the options and possibility of bringing my current V system into the digital realm with the cfv-39 back in the context of how I work. I wasn't even sure it was logical and had concerns about the long term support of the back with the recent change of hands with Hasselblad again.
    I have been a film photographer since 1983 with a full darkroom shooting primarily large format 4x5 and 8x10 for my landscapes and fine art subjects. The bulk of all my prints are fine prints made to archival standards in the darkroom with the ocassional scan on my system or a high res tango drum scan for a client and printed digitally. There are times when the LF systems are not an option or desirable so I bring my V series. It is light, extremely reliable, no batteries required and top-notch glass that always produces high quality images. I was considering the idea of shooting both film and digital for my fine art projects and exploring some other possibilities in my current business model. The cost of the back at $14k is big factor for me personally and with so many other options that surfaced as a result of my discussions I need to think about the return on investment for the cfv back vs some of the other options that need to be considered. One of the most simple but profound pieces of feedback I got was "digital looks different than film". I haven't shot much digital so that simple thought did not occur to me at first. I have spent 30 years building a large portfolio of images based on my style and vision. My "look" sources from film as a medium and that is something that I really need to explore before going too much further. I did the tests with the D3X because it was available to me and from what I understand a goo quality DSLR. I thought it was at least help me close the gap on the different looks.
    I needed to get through the basics first before taking the next steps and that is exactly what the feedback from this thread has given me. I am still on my journey with all of this but I am much more organized and prepared as a result of everyone's help and input. I appreciate your comment and if you have any specific input or feedback I would be interested in listening to what you have to say.
    Tim
     
  44. Tim, Thank you for the explanation. It makes it much clearer. Digital does look different from film, but not that much in a final print, more so on 100% magnification on computer screen. I have a fairly large size book where I have used medium format film, small format film (35mm), DSLR and even 5Mp point and shoot digital in one or two images and dobody has been able to tell them accurately apart in the final printed copy. It is good for you to make that comparison, even more so if you make a real world comparison of final products that you would actually produce with the system.
     
  45. Hey Tim,

    About a year ago, I'd borrowed a CFV39 for my 503cw. I too have a considerable V system. The image quality w/the 39 is fantastic. I used it with the crank and w/ the winder as well. For this, you use the supplied cable from the back to the winder. I've also used the H system and I don't like it. It may be innovative but it doesn't feel like a Hasselblad.

    I think the 'Blad backs have come a long way. My first experience was with the Ixpress 96 which you had to tether to the image bank. Things are much better now. My biggest issue was with shooting verticals. I have the pop-up and the PM5 45˚. It was pretty disorienting. I hope you have a 90˚ finder. Also I didn't have any finder masks and would crop into horizontal subjects if I wasn't careful. Other than this, it worked pretty well. My widest lens is a 50mm. I believe you can rent a 30mm and finder (I'm in NYC). I think if I owned one, I'd have the view camera adapter for it. I believe only Leaf makes a back that you can rotate and I'm sure it can't be touched for 14k.

    Incidentally, I got a 5d2 four months ago. I like it but, it's very different than working w/ MF but I'm sure you know this too.

    I also think if you're serious (I believe you are) and in a major city, a Hasselblad rep would loan you a CFV w/ the bits and pieces.

    Good luck.
     

Share This Page