Posted by Marc Adamus on April 14, 2006 at 11:33 AM
I had this really interesting conversation with a professional landscape photographer, Vincent Taylor on P.Net the other day. The guy still uses 35mm Velvia film for 100% of his work and has no intentions about moving digital anytime soon. Furthermore, he believes that by using his Eversmart Supreme II scanner or by sending files out to be scanned on a Tango drum scanner, he can acheive giclee enlargements that rival those possible to attain from a 13mp digital RAW file, such as the produced by the 5D.
I have a LOT of scanning experience myself (but am not any sort of tech-guru or print expert in the slightest way), and have seen the best reprints possible from 35mm in the galleries of Rowell, Wolfe, Mangelsen, ect. In my opinion, I haven't seen anything that clearly rivals the enlargement capability I have using my 5D (another probable reason pros like Wolfe have made the switch to digital). That said, I have never personally scanned with the Eversmart or Tango. I have experience only with a Nikon 8000ED, 9000ED and Imacon Flextight system. Using those systems, the 5D is producing much sharper files after processing.
I realize that the resolution of scanned film is WAY higher than 13mp, but I can't find that this is entirely a useable resolution or tack-sharp at 100% like my digital files. My scanned 35mm is never 'tack' sharp at a comparable resolution. If you think drum scanning 35mm would produce significantly better results and have personal experience doing so, I'd like to hear from you.
The differences in capture technology are pretty complex so there's really no basis for objective comparison. One medium excels in some things and the other in other things. Detail (i.e. resolving power) is just one part of what constitutes "quality" which, in turn, is inherently subjective. There's also dynamic range, grain, spectral response ("curves"), etc.
Rather than try to find a definitive answer where there is none, the only recommendation I can make is that anyone faced with a choice examine what each alternative provides, along with personal factors of cost, convenience etc. and make the decision themselves and without prejudice.
At the end of the day an image lives or dies by its emotional appeal first, not by the capture medium.
Guy
Posted by Marc Adamus on April 14, 2006 at 1:00 PM
You're right, Guy. To make comparisons between the two is very difficult. I tried to isolate particular elements for comparison, such as detail visible at 100% at a certain resolution (similar to that 39mp digital vs. LF film sample on Luminous Landscape? I believe), but when it comes to actual print making, it may be impossible to formally compare the two mediums. I also wanted to get opinions on whether I might be significantly better off sending 35mm to be drum scanned as opposed to doing the scans myself on a top of the line desktop model.
I agree completely that the choice to go with either medium should be based on individual factors and I'm not trying to convince anyone otherwise.
Marc, why not just send a favorite film image off to be drum scanned? Compare the scan and then a print to a version that you already have from a ccd scan. 35mm drum scans are pretty reasonably priced.
You could go even further by taking a film shot side be side with your 5D (I assume you have a film body still), have it drum scanned, and then do a print of each image. Obviously the print is the ultimate matter of concern here, so that should be the real basis of comparison. You have shot enough 35mm film to get a pretty good idea of where it stands in terms of resolution, so I'd be suprised if your gut instinct about how it compares wasn't correct. Still, I think its always good to see the "real world" reality of the difference, just for the sake of personal knowledge if nothing else.
To tell you the fact -
I dunno about 35mm but MF still has lots of details. But as Guy mentioned -
"Detail (i.e. resolving power) is just one part of what constitutes "quality" which, in turn, is inherently subjective. There's also dynamic range, grain, spectral response ("curves"), etc.
Rather than try to find a definitive answer where there is none, the only recommendation I can make is that anyone faced with a choice examine what each alternative provides, along with personal factors of cost, convenience etc. and make the decision themselves and without prejudice.
At the end of the day an image lives or dies by its emotional appeal first, not by the capture medium. "
Posted by Jerry Greer on April 14, 2006 at 4:23 PM
We've beaten this around before, 35mm film guys try to justify this by resolution. They have the "no rezzing" ideal branded onto their forehead! I went through this same thing with a photographer out of Colorado, he owns his own drum scanner. Honestly, I think that’s why he’s so hung-up on it the notion that a 35mm drum scan is better than 13mp’s. If I had $40,000+ scanner sitting there I’d try everything to justify it, even lying to myself. I’ve seen this side by side and the 13mp image kills the film image. It’s MUCH cleaner, MUCH sharper, MUCH more dynamic range….I can keep going. There is nowhere that I can see where even a drum scanned 35mm tranny even gets close to a properly processed 13mp+ image, period! I think that the real word here is a PROPERLY PROCESSED DIGITAL FILE.
I don't look at this issue as the technical merits of film v. digital. I believe with digital more chances are taken and more problems are worked out in the field. With equal skills, I believe the digital photographer will have more exhibition or 'A 'grade prints at the end of the year as compared to an equally skilled 35mm film shooter.
An analogy might be in comparing lens sharpness. The old Canon 300/f4 is apparently sharper than the newer 300 f/4 IS (with image stabilization). That is, it can be proved that the old lens is sharper by locking the lenses down on a tripod and comparing resolution charts. But in practice, the image stabilized lens will produce more sharp prints. So which lens is "sharper".
The real question is which is the best image producing SYSTEM.
Is there any film being shot at National Geographic?
35mm film shooters have a high opportunity cost with shots that were never taken
Posted by Les Barstow on April 14, 2006 at 5:00 PM
I know I personally will never go back to 35mm film. I've got about 25 drum-scanned 35mm slides that were the best of my early pro slide shooting experience; I can state outright that the 12+MP images I get from my D2X easily outweigh the quality of those shots in enlargement potential, detail resolution, and dynamic range.
The things I liked most about film - saturation and color - I can achieve through PhotoShop on my digital shots.
Posted by Marc Adamus on April 14, 2006 at 5:50 PM
Thanks Les. Jerry, my thoughts exactly. I don't know if I'll ever waste the time to get 35mm drum scans made, but I know I'm not going back to 35mm. I've shot several thousand rolls of film and made several thousand digital images just in the months since I've switched. I can't find a compelling case anywhere for sticking with 35mm personally.
Nowhere in the above replies does anyone state what film was being drum scanned when the comparison with digital was made. Remember, garbage in equals garbage out. If the scans were of chromes using a higher speed film or inferior lenses, then there is no basis for comparison.
I have too many chromes, Provia 100F and Velvia, which have much more detail than we ever gotten with our D1X shot raw with at mpxls. Still, 90% of our images are now digital.
Posted by Marc Adamus on April 14, 2006 at 11:52 PM
Sorry David. The above know my work was almost exclusively with Velvia 50 and less often Provia 100.
Posted by Robert Cable on April 15, 2006 at 10:45 AM
For large prints of 35mm film images (16x24 and up) I have found that drum scans are better.
I have used the LS2000, LS4000, and now the LS9000. The drum scans give you greater shadow detail and overall sharpness. I have never scanned anything myself and printed at 16x24 and larger because I don't have a big printer, so I can't speak for the comparison between desktop scans and drum scans at large print sizes. I have always been wowed by the large lightjet prints from Nancyscans or Photocraft when I get them back. Let them do a scan and make you a print for your reference.
I know how you feel about what I am about to say with regard to timing and workflow, but try stitching as few as 3 - 4 digital images together and then the conversation really ends. I probably lean towards the overkill side. But a 16x20 of (4) dSLR images is just incredible.
I used to shoot Velvia 50 and have a Canoscan 4000 (4000 dpi)... recently I went back to edit some of thos files and just thought, ughh... what a mess to clean some of them up.
Posted by Jerry Greer on April 15, 2006 at 6:18 PM
T. David,
Most of us guys shot professionally with film so it sure will not be high-speed film. I most always used Velvia Classic with 35mm. Once the new films came out, Velvia 100F, Astia 100F, Velvia 100, I had already switched to large-format.
I have drum scans from two year old $60,000 ICG drum scanner in 645and have compared them to my 5D. Both images that I compared were taken with approx. 45mm lenses (35mm). The drum scanned film image has a touch more detail but I think most is due to the film grain. But the super clean image from the 5D holds its own even @ 20". The digital image is just so clean! If I throw some grain back into the digital film using PK Sharpener, grain equal to the film scan, the digtal file really takes on a sense of more detail. When you factor in the increased dynamic range of the 5D then it's a no-brainer! 35mm is not even close!
Posted by Les Barstow on April 17, 2006 at 1:39 PM
The scans I have lying around are 4000dpi drum scans professionally scanned by PhotoCraft (full treatment - oil mount, etc.) of Provia and Velvia 100F film. I don't know that I could have done much better, and Roger above mentions how much he appreciates scans and prints from PhotoCraft.
The images I have from my D2X just flat out-perform both these early scans and my later scans from my Minolta DiMAGE 5400. I think Jerry sums it up; the extra perceived "detail" from a 35mm scan is really more about noise than about actual detail. I can take my 35mm scans to 16x24", but I don't think I would dare to sell those images above that print size - edge sharpness degrades too much at that point. OTOH, I am comfortable offering full-frame D2X enlargements to 20x30" and larger.
Discussion
14 repliesI have a LOT of scanning experience myself (but am not any sort of tech-guru or print expert in the slightest way), and have seen the best reprints possible from 35mm in the galleries of Rowell, Wolfe, Mangelsen, ect. In my opinion, I haven't seen anything that clearly rivals the enlargement capability I have using my 5D (another probable reason pros like Wolfe have made the switch to digital). That said, I have never personally scanned with the Eversmart or Tango. I have experience only with a Nikon 8000ED, 9000ED and Imacon Flextight system. Using those systems, the 5D is producing much sharper files after processing.
I realize that the resolution of scanned film is WAY higher than 13mp, but I can't find that this is entirely a useable resolution or tack-sharp at 100% like my digital files. My scanned 35mm is never 'tack' sharp at a comparable resolution. If you think drum scanning 35mm would produce significantly better results and have personal experience doing so, I'd like to hear from you.
The differences in capture technology are pretty complex so there's really no basis for objective comparison. One medium excels in some things and the other in other things. Detail (i.e. resolving power) is just one part of what constitutes "quality" which, in turn, is inherently subjective. There's also dynamic range, grain, spectral response ("curves"), etc.
Rather than try to find a definitive answer where there is none, the only recommendation I can make is that anyone faced with a choice examine what each alternative provides, along with personal factors of cost, convenience etc. and make the decision themselves and without prejudice.
At the end of the day an image lives or dies by its emotional appeal first, not by the capture medium.
Guy
I agree completely that the choice to go with either medium should be based on individual factors and I'm not trying to convince anyone otherwise.
You could go even further by taking a film shot side be side with your 5D (I assume you have a film body still), have it drum scanned, and then do a print of each image. Obviously the print is the ultimate matter of concern here, so that should be the real basis of comparison. You have shot enough 35mm film to get a pretty good idea of where it stands in terms of resolution, so I'd be suprised if your gut instinct about how it compares wasn't correct. Still, I think its always good to see the "real world" reality of the difference, just for the sake of personal knowledge if nothing else.
I dunno about 35mm but MF still has lots of details. But as Guy mentioned -
"Detail (i.e. resolving power) is just one part of what constitutes "quality" which, in turn, is inherently subjective. There's also dynamic range, grain, spectral response ("curves"), etc.
Rather than try to find a definitive answer where there is none, the only recommendation I can make is that anyone faced with a choice examine what each alternative provides, along with personal factors of cost, convenience etc. and make the decision themselves and without prejudice.
At the end of the day an image lives or dies by its emotional appeal first, not by the capture medium. "
An analogy might be in comparing lens sharpness. The old Canon 300/f4 is apparently sharper than the newer 300 f/4 IS (with image stabilization). That is, it can be proved that the old lens is sharper by locking the lenses down on a tripod and comparing resolution charts. But in practice, the image stabilized lens will produce more sharp prints. So which lens is "sharper".
The real question is which is the best image producing SYSTEM.
Is there any film being shot at National Geographic?
I would not have even tried this (weird) shot with film:
http://www.photoportfolios.net/portfolio/pf.cgi?a=vp&pr=21693&b=up&st=1&la=38&ph=1&sid=19753&pi=DMILLER&u=19753
35mm film shooters have a high opportunity cost with shots that were never taken
The things I liked most about film - saturation and color - I can achieve through PhotoShop on my digital shots.
I have too many chromes, Provia 100F and Velvia, which have much more detail than we ever gotten with our D1X shot raw with at mpxls. Still, 90% of our images are now digital.
I have used the LS2000, LS4000, and now the LS9000. The drum scans give you greater shadow detail and overall sharpness. I have never scanned anything myself and printed at 16x24 and larger because I don't have a big printer, so I can't speak for the comparison between desktop scans and drum scans at large print sizes. I have always been wowed by the large lightjet prints from Nancyscans or Photocraft when I get them back. Let them do a scan and make you a print for your reference.
I know how you feel about what I am about to say with regard to timing and workflow, but try stitching as few as 3 - 4 digital images together and then the conversation really ends. I probably lean towards the overkill side. But a 16x20 of (4) dSLR images is just incredible.
I used to shoot Velvia 50 and have a Canoscan 4000 (4000 dpi)... recently I went back to edit some of thos files and just thought, ughh... what a mess to clean some of them up.
Most of us guys shot professionally with film so it sure will not be high-speed film. I most always used Velvia Classic with 35mm. Once the new films came out, Velvia 100F, Astia 100F, Velvia 100, I had already switched to large-format.
I have drum scans from two year old $60,000 ICG drum scanner in 645and have compared them to my 5D. Both images that I compared were taken with approx. 45mm lenses (35mm). The drum scanned film image has a touch more detail but I think most is due to the film grain. But the super clean image from the 5D holds its own even @ 20". The digital image is just so clean! If I throw some grain back into the digital film using PK Sharpener, grain equal to the film scan, the digtal file really takes on a sense of more detail. When you factor in the increased dynamic range of the 5D then it's a no-brainer! 35mm is not even close!
The images I have from my D2X just flat out-perform both these early scans and my later scans from my Minolta DiMAGE 5400. I think Jerry sums it up; the extra perceived "detail" from a 35mm scan is really more about noise than about actual detail. I can take my 35mm scans to 16x24", but I don't think I would dare to sell those images above that print size - edge sharpness degrades too much at that point. OTOH, I am comfortable offering full-frame D2X enlargements to 20x30" and larger.