Ultimate digital chip for LF

[MJSfoto1956]

The flaw with this methodology is that scanning film introduces attenuation that exaggerates grain (even more so than using a condenser enlarger vs. a diffusion enlarger). The other flaw is that importing a (noisy) digital image via Adobe RAW "cleans up" the noise resulting in the appearance of a noise-free image (in fact it can be demonstrated that ALL digital CCD and CMOS sensors exhibit significant internal native noise). Needless to say, cleaning up the data also alters it -- there is no way to avoid that. I believe this post-polishing is the "look" that film purists correctly complain about when it comes to digital. It certainly does have a "look"! So the statement that my XYZ sensor is noise free while my scanned film is grainy is simply wrong-headed and the result of flawed methodology (and hyped marketing).

This is the same flawed comparison that Luminous Landscape and other sites have used for years to "prove" the superiority of digital over film. Digital certainly is superior from an economics/convenience point of view (which is why commercial photographers have overwhelmingly endorsed it). However, infering that today's one-shot digital backs produce superior quality to large-format film is wishful thinking. A fairer comparison will be made in the second issue of MAGNAchrom where we scientifically compare a pure analog workflow AAA photo to a completely digital workflow DDD photo. And only the final print will be scanned for the publication, not the film. I think you'll be surprised at the findings.

Hint: yes, a digital print is indeed quite different than a film print!

J Michael Sullivan
Editor/Publisher, MAGNAchrom
www.magnachrom.com

[Gordon Moat]

I definitely agree with the points made by J. Michael Sullivan and Ted Harris. While that Leaf back is probably having the images going through the Leaf software, it is indeed being polished. That someone would not do the same polishing in PhotoShop or high end scanning software like Creo oXYgen seems inconcievable to me.

As far back as 1996, on far worse scanning gear, I was getting those smooth grainless images from scanned film. This involved a few steps in PhotoShop 3.0, which is now much more easily accomplished with newer PhotoShop or third party add-on software. This post-processing, whether to a RAW capture or scanned film, can sometimes take the same amount of time. As a professional, the amount of time makes a difference in most post processing, though enthusiasts and amateurs might be less inclined to worry about post-processing time.

Rather than compare a Leaf digital back to a low end scanner, I think a more valid comparison would be to a Creo scanner, especially when the software is very similar in operation. However, the idea of a digital back is partially to not have film expenses, even if your prior work practice was to charge each client for your film (zero cost item), or incorporate it into each invoice; rather the advantage should be cutting out drop-off and pick-up time at a lab. That time needs to be balanced against any extra time processing RAW images, or in archiving numerous digital files. Suppose a professional film photographer had 100 actual photography days in a year, with the rest of the time in meetings, writing proposals, delivering work, or contacting clients. If switching to a digital back eliminated lab visits, then maybe 150 days of actual photography work could be done in a year. Thus the saving might not be film costs that you should have been billing to clients, it is instead a way to enable a photographer to do more billable assignments each year.

So if you previously ran a zero cost film billing set-up, a comparible scanner would be near the price of a Leaf back. If you did not bill out for film and developing, for whatever reason, then the costs of film for the expected two years professional usage lifetime of the Leaf back would need to be deducted from the amount one might spend on a scanner; maybe a Creo iQSmart 2 instead of an EverSmart or iQSmart 3 . . . or some other choices. Another expense is that storing film is low cost, and not every shot on film needs to be scanned, so only the scanned images need archiving and computer storage space, while all the direct captures from the Leaf are probably images you want to store; this implies a need for more computer storage and archiving space for a digital back workflow than for a scanning film workflow, though that depends upon what you choose to keep of your images.

The last issue is focusing on the ground glass. While a digital back is double 35mm or less than 6x4.5 in size, the issue of stiching four shots makes it closer to 4x5, making the ground glass composing the shot step more familiar. Stitching, even with really good software, adds time to post processing, and slows down the original capture. Using a Leaf back as a single shot device seems like it would make more sense when paired with a 645 medium format camera, rather than the back of a view camera.

Ciao!

Gordon Moat
A G Studio
http://www.allgstudio.com

[David Luttmann]

I don't know Ted....

I've just finished going over the sample scans on DVD from the Luminous Landscape test. The Aptus 75 is not that different from the P45 (34 vs 39MP). The printed crops I made from a 40" size had the P45 looking sharper than that of the scanned film....not much, but definitely there.

Based on all the sample prints I've made, it's far closer than most people believe....at least up to 40" anyway. Of course, we won't mention the cost

[MJSfoto1956]

The Luminous Landscape DVD is simply flawed, in spite of the endorsement of some of today's top lumineries. What you have is a cleaned-up/sharpened digital file (via RAW processing) compared unfairly to a sheet of scanned film (whose grain is attenuated by the scanning process itself) which does not have the same benefit of noise reduction + sharpening. So of course the DVD can show whatever they want it to.

This DVD is not science -- it is psuedo-science posing as fact.

J Michael Sullivan
Editor/Publisher, MAGNAchrom
www.magnachrom.com

[Sheldon N]

Not to debate the merits of the LL tests, but my understanding is that the DVD contains the RAW files from all the respective cameras. That would mean that if you are checking out the comparison files from the DVD, you are free to control all sharpening and post processing to make it as fair as you want.

I don't think the DVD or Michael purports it to be any fact at all, but rather he put it out so that he wouldn't have to defend his own conclusions or his processing of the files. I'm sure there's a bit of a financial motivation to it as well.

In a realistic sense though, both LF and state of the art MF digital are so good that very large high resolution prints are easily attainable with either. For us mere mortals, one option is financially feasible, and the other is a pipe dream. For commercial photographers pursuing MF Digital, I'm sure the competitive environment, demand for fast turnaround, and workflow considerations are just as important (if not more so) than which one is "better".

As for me, I'm always curious what the current state of the art digital is. However, I have no plans to take out a second mortgage to buy one, and if I really want more resolution there's always 8x10 or 11x14.

The Luminous Landscape DVD is simply flawed, in spite of the endorsement of some of today's top lumineries. What you have is a cleaned-up/sharpened digital file (via RAW processing) compared unfairly to a sheet of scanned film (whose grain is attenuated by the scanning process itself) which does not have the same benefit of noise reduction + sharpening. So of course the DVD can show whatever they want it to.

This DVD is not science -- it is psuedo-science posing as fact.

J Michael Sullivan
Editor/Publisher, MAGNAchrom
www.magnachrom.com

[Robert Payne]

Yeah, sorry Mr. Sullivan, but the files on the Luminous Landscape DVDs are all RAW files, meaning, they are there for YOU to process them/sharpen them/clean them up, etc...and actually, if you have any experience with digital at all, and you shoot it correctly, most of the files, for me at least, don't need any "cleaning up."

Unfortunately, film enthusiasts seem to think that all digital files are edited to bloody hell. That's not the case if you're simply a skilled photographer who captures images on a digital sensor, instead of film. Now if you're Bob the digital wedding photographer wannabe, it might be a different story...

I have to say though, the Luminous test is done with a subject that is at close distances. Digital is very good in these situations (portraits, product, etc). However, I believe if the test shot was not in the studio, but was a landscape with vast detail in the distance (basically, what most people on this forum shoots), the test might have been a lot more difficult for the digital backs and DSLRs to keep up with the LF and MF film.

[paul r w freeman]

I think its a valid comparison because it reflects what I was actually doing in practice. It isn't a valid comparison of the media and I didn't claim that it was.

Because of where I live, its simply not practical to get professional drum scans and processing done in a timely manner. Hence my comparison is not a technical lab comparison, but a practical one. Driving to the nearest lab open after 5pm absorbs 3 hours of my time. Then there is the wait while all the processing is done.

If I were to have my own colour darkroom, develop my own film using one shot developer, dry it in a professional dryer, then scan the film on my own drum scanner then I might be able to beat the quality of the digital back. This would probably cost as much to set up as it cost me to buy the back, and the eventual process would take longer.

My point is not what the theoretical quality of one medium is over another, only how it works for me in practice. Thats a very different thing. Maybe if I had more time, lived in a more convenient location, had bought better film equipment, worked harder at 5x4 film I would have had better results.

In any case, my point is not merely a film comparison, more to point out that high quality digital results are a reality, for me, not a figment of a future chip that may never exist.

The problem is that its hard for people to get their hands on this kind of gear to make the practical comparisons because of both the cost and the complex fittings needed to hook backs onto view cameras. As a result, people have to rely on lab tests which may not reflect a real life workflow. For this reason all sorts of misconceptions arise like "film lenses haven't got the resolution for digital".

[David Luttmann]

The Luminous Landscape DVD is simply flawed, in spite of the endorsement of some of today's top lumineries. What you have is a cleaned-up/sharpened digital file (via RAW processing) compared unfairly to a sheet of scanned film (whose grain is attenuated by the scanning process itself) which does not have the same benefit of noise reduction + sharpening. So of course the DVD can show whatever they want it to.

This DVD is not science -- it is psuedo-science posing as fact.

J Michael Sullivan
Editor/Publisher, MAGNAchrom
www.magnachrom.com

Michael,

The file I just worked on was processed with Capture One with the noise reduction settings at 0. The file was not "cleaned up". I sharpend both the digital and film files to taste and printed them. There is nothing "psuedo-science" here. I think you have a little bit of reading to do on this topic as there appears to be a bit of apparent bias in your statement.

What I found one print was this:

The digital file had no noise. It had better accutance than the film no matter how much the film was sharpened. Detail between the two were similar. This was a cropped print from a 40" enlargement.

Please tell us where the LL DVD is "pseudo-science." How is the scanned film vs a RAW digital shot flawed?

[bglick]

Great comments guys....

A few things I would like to add...

First, LL uses digital lenses on the digital back, but uses conventional LF lenses with film.....these digital lenses are ultra sharp, even when used on film. Although the digital lenses are not quite big enough to cover 4x5, some can come close, I would like to see that thrown into the mix to help level the playing field. As the superior digital results when be reduced also if this latest generation of lenses were NOT used.

Next, the digital back is acheiving its greatest advantage by allowing to extra "free stops" of aperture diffraction to attain the same DOF. When DOF is NOT an issue, the MF digital can shoot at optimized f8 - 11, whereas most LF lenses are optimized at around f22, this was never considered a degrading consequence of LF, till now.... i.e. highly dense recording media (digital sensor) provides the benefits of stepping up in format size in resolution, without the obvious draw back of being forced to use longer fl lenses to acheive the same composure.


Its been rumored that Canons replacement for the 1dsII will be a 24MP body, at the same price price both the 1ds and 1dsII were introduced at... $8k. If this becomes a reality and Canon comes through with their version of digitar lenses, we may see a new "film vs. digital" war. Now, a simple two stitch image from a 35mm camera will yield results better than 4x5 and 5x7, and most 8x10. (based on f stop used, or DOF required) The smaller 35mm format enables even lower f stops, allowing even higher lens resolution to the sensor than MF digital. Scary, huh...

In the end, there is lot more to the equation then scanned dpi vs. pixel count... and unfortunately all advantages move in the direction of digital....its more phyiscs than pixie dust...

One poster mentions they can continue to climb the LF ladder if they want more resolution... I would offer some qualifiers to that statement. IF DOF is required in the shot, jumping past 4x5 offers little benefits (if any) as diffraction will eat up any potential gains, this assumes the 4x5 shot was taken at f22, which therefore requires f45 for 8x10, f64 for 11x14, etc. However, if you are shooting flat subjects or at infinity, then the gains can be appreciated in larger formats as you do NOT continue to pay the diffraction penalty when climbing the LF film ladder.


The one area color neg film remains strong over digital is exposure lattitude...it seems digital chip makers have not been able to tackle this problem as of yet.


Also, as pointed out, to make an apples to apples comparison, the users should have the ability to apply all available tools to produce the best final image as possible to each medium. I think this is another area that digital excels in... I have experimented with both, and it's clear that digital images are more pixel friendly than film. In other words, noise cleans up nicer than film grain. It could be a result of more sophisticated software chasing the digital market, as its the bigger source of future revenues. When I take pix with my 8 MP tiny digicam, yeah its noiser than larger digital sensors, but its amazing how well it cleans up using some of the noise reduction programs.

LF and a low cost scanner will clearly remain the choice for the hobbiest or those not willing to throw a second mortgage at new technology, or who just prefer the many benefits of film use in the field vs. electronics. If there was ever a comparison made of the cost of LF set up with low end scanner vs. good (maybe not the best) digital, the value would surely be in the LF system. This assumes low volume usage per year of course. If one is a professional, it's an entirely different game..... as economics is a part of every business.


We just have to hope the next round of digital development is not so strong and cost effective (as Canon rumors above) .... such that it puts an even bigger dent into film sales. It's unfortunate the two big LF color film makers got so damn huge from years of film profits...... but, now what made them (film), will soon be seen as loosing division. Not trying to start the "will they stop making film thread"..... just noting the huge jumps in digital that many of us could not fathom just 3 - 4 years ago when much of this "is film dead" started.


Oh, yeah, B&W shooters, you're lucky....... its mostly us color guys who beat this issue up. The fact small companies make B&W film will probably assure availablity of film supply for our life time....

[Eric Leppanen]

First, LL uses digital lenses on the digital back, but uses conventional LF lenses with film.....these digital lenses are ultra sharp, even when used on film. Although the digital lenses are not quite big enough to cover 4x5, some can come close, I would like to see that thrown into the mix to help level the playing field. As the superior digital results when be reduced also if this latest generation of lenses were NOT used.

While at a theoretical level I agree with you, as a practical matter LF film lens design has essentially ceased, and most likely there will never be any next-generation products brought to market. So I think using currently available LF film lenses amounts to a valid comparison. No sense in using higher resolution LF digital lenses if none cover the 4x5 format, with no prospect of any such lenses becoming available in the future. Perhaps the LL folks might have achieved a smidge more resolution using a Sironar-S versus a Sironar-N, but I doubt this would have made much practical difference. Given the limited objectives of their test, I think their setup was reasonable.

I'm an LF color film guy too, and I've also been watching the digital capture scene to see when it may become necessary to jump to the dark side. Canon has much work to do with wide-angle lenses, greater bit depth, etc. for a 24MP DSLR to reach its full promise. I shoot mostly 8x10 these days and earlier this year we were able to persuade Fuji USA to inventory 8x10 Velvia 100 in the U.S., and one would doubt Fuji would do this if they foresaw an abrupt end to color film manufacture. So I suspect we have several more years before color film availability becomes a serious problem (at least from Fuji). By then, who knows what digital products (some presumably for view camera platforms) may become available?