The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Scanning negatives with Phase One and Rodenstock digital

IBICO

New member
Anyone tried out this? To digitize negatives using a MF back and Rodenstock digital lens?

Would be great to compare this vs a scan from "cheap" drumscanner, Epsonv750 and other options.

If it is good enough for large prints, then I could justify spending lot more on a new back instead of a new drumscanner.
 

torger

Active member
I have done some camera scanning with DSLRs, using stitching to achieve high resolution.

Shooting 35mm film is relatively easy, but already at 645 film you get issues with film curling so you need a good film holder that holds the film flat. Wet mounting may be worthwhile.

The achievable quality is certainly better than Epson V750, you can use multiple exposure with different shutter speeds (HDR) to get amazing dynamic range from hard-to-scan films like fuju velvia, and you can adjust focus to perfection assuming you have a good film holder and copy stand so you get corner-to-corner parallelism.

A expertly made drum scan will be better though, the weakness of lighttable type of copying is haloing around high contrast edges, described here: High Contrast Edges | Cheap Drum Scanning

I would say though that most would consider camera scanning "good enough", but if you want "the best", drum scanning is still it.
 

dougpeterson

Workshop Member
We do this at a high level with our Cultural Heritage Film Scanning Kit. For anything 4x5 or smaller, assuming use of an 80mp back and a 120 ASPH lens at f/8 our results are comparable to drum scans as measured by the FADGI standards for film digitization.

For large institutions these systems are far, far, far more price effective on a cost-per-item-scanned (accounting for the cost of labor and facility overhead) because the negatives can be scanned as quickly as they can be placed into the holder (rather than a 4-20 minute scan time on a film scanner) and are imaged in raw (meaning that a session of 1000 negatives can be inverted, pushed/pulled, curved, rotated, cropped en masse etc without the open/save/close of a TIFF workflow).

Note that there are no shortages of challenges involved. You'll need methodology/equipment to hold the camera and film in perfect alignment as even a small fraction of a degree of non-parallelism will cause a soft corner. You'll need to keep the film quite flat for the same reason. And you'll need a very diffuse lighting source to avoid/minimize the effect of edge rendition that torger referred to. And you'll need a very good lens used near-wide-open to avoid diffraction. And you'll need to manage the LCC workflow to ensure properly-even exposure and color across the frame.

Solving these problems was quite time consuming for us, but then, our quality standards were quite high (and were being measured numerically for compliance with international standards).

 

IBICO

New member
Cultural Heritage Film Scanning Kit look interesting, but I guess it is not for free? And shipping outside us?

It almost look like a repro stand, with a lightboard?

If I could find some dias frames for 4x5, then much would be solved. I have a drumscanner now, but it is very large. Also need old OS9 on a G4 mac.

Do anyone have some files from scanning with this film scanning kit and digital back?
 

dougpeterson

Workshop Member
Cultural Heritage Film Scanning Kit look interesting, but I guess it is not for free? And shipping outside us?

It almost look like a repro stand, with a lightboard?

If I could find some dias frames for 4x5, then much would be solved. I have a drumscanner now, but it is very large. Also need old OS9 on a G4 mac.

Do anyone have some files from scanning with this film scanning kit and digital back?
Our Cultural Heritage products are shipped internationally.

But this product, like all of our Cultural Heritage products are designed, built, warrantied and priced with institutional and large-volume service bureau in mind. Such entities are digitizing hundreds of thousands of pieces of film so the longevity and error-proofness of the equipment must be exceptionally high.

If you're just digitizing a few hundred pieces of film every once in a while I'd advise you to build a cheap DIY rig. Your time-per-capture will be much higher as you fuss with things like alignment, but the setup cost (time and financial) to get to a rock-solid rapid-capture setup like ours is not going to make sense spread out over only a few thousand captures.

We have samples from our system from a variety of eras, emulsions, and sizes including some cool glass-plate negatives. But I don't think they will be of much relevance to you. Better for you to build your own cheaper setup with minimal investment and see how good you can do.
 
Last edited:

timparkin

Member
We do this at a high level with our Cultural Heritage Film Scanning Kit. For anything 4x5 or smaller, assuming use of an 80mp back and a 120 ASPH lens at f/8 our results are comparable to drum scans as measured by the FADGI standards for film digitization.



Just had a good look at the standards and they're probably a good basis to avoid awful archiving but to be honest they're fairly low standards to hit, even the four star standards.

For instance you can get away with half of pixel of colour fringing across 20% of the image and still get 4 star rating. You can also pass with four stars with delta E's of 3 (or worse) which are quite visible differences (and you in areas you can have delta E (lab) of 6 for large areas of your scans and get away with it.

The biggest issues is definitely the mentioned halos around dark/light areas which I've seen on every system that takes more than single point readings and I've come to the conclusion is inevitable apart from with drum scanners.

These halation issues might not occur obviously in many images but they're there never the less.

Everything is a trade off and it's good to know what those trade offs are.

That said if all I had to match was the FADGI standards and I had tens of thousands of images to scan I'd be buying a system like the above (or possibly knocking up one based on a D800 - perhaps - depending on budget!).

Tim
 

dougpeterson

Workshop Member
Just had a good look at the standards and they're probably a good basis to avoid awful archiving but to be honest they're fairly low standards to hit, even the four star standards.
If there was a FADGI 5-star we'd hit it :). For instance FADGI requries (off the top of my head) 92% sampling efficiency and we regularly achieve 99-100% (as rounded to the nearest digit) across the frame.

Unfortunately for us the government doesn't give out A+s.

Trust me when I say we'd be delighted if the grading curve was even tougher :).
 

Egor

Member
This is from a thread I did last year but is a pretty good comparison.
Using a Leaf Aptus2-80, PhaseOne 120mm Macro AF and 1:1
Tango is "loafing" at a mere 6000PPI but hey, its for comparison :)
Doug's set up would be better for sure, but this wasnt bad. Tango beats it, for sure, but was told to use a better lens and possibly a repro stand setup and maybe get closer. This is a color tranny, of course, but I havent seen any other comparisons like it.

e
 

Egor

Member
I found some B/W examples!
As a followup to those tests last year, we won a contract with the Getty to use the Leaf/Phase but only for contact sheets (over a thousand of them)
The job was a cultural history type thing of Cree Indians in Canada back in the 70's
Selects were then scanned using the Tango.
I will try and post some side by side comparisons of the drum vs leaf B/W but it wont be anywhere near a fair fight because the contact sheets were for reference selection only and the Tango is...well...a Tango drum scanner, and each file is oil mounted, analyzed and custom scannned PMT. There is no comparison. But I still think the Leaf, with a proper setup like Doug's and a better lens, may have been "close enough"
 

dougpeterson

Workshop Member
This is from a thread I did last year but is a pretty good comparison.
Using a Leaf Aptus2-80, PhaseOne 120mm Macro AF and 1:1
Tango is "loafing" at a mere 6000PPI but hey, its for comparison :)
Doug's set up would be better for sure, but this wasnt bad. Tango beats it, for sure, but was told to use a better lens and possibly a repro stand setup and maybe get closer. This is a color tranny, of course, but I havent seen any other comparisons like it.

e
What f-stop was the Leaf used at? Looks a bit diffracted. Might get better results at a larger aperture - with the caveats that alignment and flatness become even more important and lens-sharpness-at-the-corners is harder to pull off without switching to the lenses like the Schneider 120mm ASPH.
 

dougpeterson

Workshop Member
I found some B/W examples!
As a followup to those tests last year, we won a contract with the Getty to use the Leaf/Phase but only for contact sheets (over a thousand of them)
The job was a cultural history type thing of Cree Indians in Canada back in the 70's
Selects were then scanned using the Tango.
I will try and post some side by side comparisons of the drum vs leaf B/W but it wont be anywhere near a fair fight because the contact sheets were for reference selection only and the Tango is...well...a Tango drum scanner, and each file is oil mounted, analyzed and custom scannned PMT. There is no comparison. But I still think the Leaf, with a proper setup like Doug's and a better lens, may have been "close enough"
I was just out at Getty last month. They have several of our systems on both the research side and library side of the institute. Great people as well.

And man is their building and property beautiful! Meier was the architect, who, my friend was quick to point out was a Cornell grad (guess which college my friend graduated from).
 

Egor

Member
Hi Doug,
Yes, diffraction was inevitable because of our need to pull DoF even at that minute a scale, MF has none. So I believe that was at F16-F22. When we opened up to reduce diffraction it got fuzzier and we gave up trying anymore. The clients and the ethnographers wanted drum scans and got tired of us fiddling with the MFD setup.
We will revisit the setup soon as a new project is on its way to us of thousands of film images from Pre-Wall Berlin; as well as a series from Afghanistan. We will wade back into the MFD scan/capture waters again soon once funding is a certainty. These are for the Corcoran and Smithsonian.
The Getty is a nice museum, and has facilities on their campus, but in many cases, the artists or owners are not willing or able to send their images to them for capture.
I am a "recovering architect" (CooperUnion then VPI class of '86) and am not a fan of Meier's work (imo, it doesnt hold up well over time, requires too much maintenence, and distracts from the artwork); but hey, everybody's a critic, right? ;)
I am a HUGE fan of Cornell, btw! :)
 
Last edited:

yongfei

New member
For DIY, what would be a good light source? Will flash be better than a LED based light? I researched the Kaiser lightbox. It seems that they offer some quality lightboxes. Will that be any good? For exmaple:
Product code: K2423
Manufacturer: Kaiser
5000 Kelvin color temperature. Uses special fluorescent tubes for high color fidelity. Color rendition index CRI = 90-100. High-quality acrylic-glass plate of exceptionally uniform transmission over the entire visible spectral range
 

yongfei

New member
This is from a thread I did last year but is a pretty good comparison.
Using a Leaf Aptus2-80, PhaseOne 120mm Macro AF and 1:1e
The result looks very good. Do you have a set up picture? Also, what kind of light source you are using?
 

dougpeterson

Workshop Member
Hi Doug,
Yes, diffraction was inevitable because of our need to pull DoF even at that minute a scale, MF has none. So I believe that was at F16-F22. When we opened up to reduce diffraction it got fuzzier and we gave up trying anymore. The clients and the ethnographers wanted drum scans and got tired of us fiddling with the MFD setup.
We will revisit the setup soon as a new project is on its way to us of thousands of film images from Pre-Wall Berlin; as well as a series from Afghanistan. We will wade back into the MFD scan/capture waters again soon once funding is a certainty. These are for the Corcoran and Smithsonian.
We have several systems at Smithsonian as well :).

The use of f/16 would definitely account for the softness. Our results vis a vis drum scanning are considerably closer (with many medium sizes like 35mm slides going in favor of the MFD system) With the magnification that would be an effective f/22 which is strongly diffracted. But I totally understand the choice. That's the origin of the "over-built" solution we have; shooting at f/8-f/11 (to avoid diffraction) on a homegrown solution, even a very good quality one, requires such fiddling that the time-per-capture rate suffers greatly. Only with a very, very, solid setup can you truly hit the capture rate mass digitization projects these call for. Given you had a preference to drum scan the selects your solution and aperture selection make perfect sense. Our industrial-grade solutions aren't inexpensive and can only be justified when there is enough volume.
 
Last edited:
For DIY, what would be a good light source? Will flash be better than a LED based light? I researched the Kaiser lightbox. It seems that they offer some quality lightboxes. Will that be any good? For exmaple:
Product code: K2423
Manufacturer: Kaiser
5000 Kelvin color temperature. Uses special fluorescent tubes for high color fidelity. Color rendition index CRI = 90-100. High-quality acrylic-glass plate of exceptionally uniform transmission over the entire visible spectral range
You're spot on with the needs for high CRI lamps, to me that plus even illumination are more important than colour temp (e.g. unfiltered tungsten lamps are full spectrum) and is true for any copy work. For a cheap setup you can actually rig up tungsten lamps through a diffuser, a lot of good tips for a dark room enlarger applies to this problem.

If you can, check the light source with a spot meter to ensure the light is even.
 

Egor

Member
Yes, that is the same thread I think. I started out using a Macbeth or JustNormlicht light box, but ended up going with strobe through a diffusion panel. I think there is a pic of the setup there. very crude, very redneck, very effective. ;)
Since then we purchased some old Beseler metal neg holders for better flatness and a larger glass frame for contact sheets. Camera still held in place using a giant Cambo studio stand and extended arms, but the diffusion panel is a custom flat box held over the diffusion strobe head.
I love the sheer output and speed of the strobes and the full spectrum light output. Also don't worry so much about camera shake at 1/10000 sec burst durations.
Edge2edge evenness doesn't seem to be an issue when it comes down to such a small area.

Doug, do you have an example of a 35mm tranny or neg done with your system I can look at? ....And dare I ask, borrow to drumscan and compare? :) we are interested and may have enough volume to justify your system (depending on what it's real costs are for us..we already own a 80mp back so maybe...)
 

IBICO

New member
Might a setup with Rodenstock Digital Macro 180mm lens with digital back be better maybe? they should be high quality glass.
 
Top