Film vs. Scanning resolution

Shadow 2

A
Shadow 2

  • 0
  • 0
  • 22
Shadow 1

A
Shadow 1

  • 2
  • 0
  • 21
Darkroom c1972

A
Darkroom c1972

  • 1
  • 2
  • 35
Tōrō

H
Tōrō

  • 4
  • 0
  • 40

Recent Classifieds

Forum statistics

Threads
198,826
Messages
2,781,494
Members
99,718
Latest member
nesunoio
Recent bookmarks
0

Lachlan Young

Member
Joined
Dec 2, 2005
Messages
4,943
Location
Glasgow
Format
Multi Format
Lachlan, Lachlan...

Look, a DIY emulsion has many technical limitations, one of the limitations is that, while you may increase the speed of the emulsion, latitude won't increase what we may desire, if you blend two or more DIY emulsions of different speeds you in fact increase the resulting latitude, this is the reason why we may mix several DIY emulsions.

Another reason to blend emulsions if making a variable contrast emulsion for photopaper, you mix ortho emulsions with a raw blue sensitive (color blind) emulsions to coat paper or to print on glass. IMO, the carce people making dry plates don't usually blend much the emulsions for a greater sharpness, anyway mostly an straight emulsion is used because most of the joy comes from the involved adventure.

It's pretty clear that you have very little inkling about either dye ballasting & very monodisperse emulsions being needed to get variable contrast to work well or equilibration & why it affects emulsion blending. If you did, you would know that making a good variable contrast emulsion, or even a good, fast high contrast paper emulsion is very technologically challenging within the limits of single run emulsion making.

And because you can adjust latitude/ define your curve shape in many ways at the making stage in polydisperse emulsions, you should only even contemplate blending once you have totally exhausted all other possibilities of adjustment (change your addition rates & iodide % for a start - it's all documented on here - you have either done almost nil research, or actively dismissed the writings of actual experts as knowing less than you think your random googling tells you). Manufacturing only went over to blended emulsions once monodispersity became prevalent. For example, while the old Efke/ ADOX CHS films were polydisperse single emulsion materials, the current CHS-II is a blended emulsion (likely because it uses much more monodisperse emulsions) coated in a single layer.

Here we were speaking agout the well known multi-layer effect in the latitude boosting. The 1 layer kodachrome was a pitfall, still slides have a way narrower latitude requiring less the latitude boost, while print film (color/bw) has evolved to deliver insane highlights latitude, provided by the multi layer: 9 layers, 3 per color in good color print film !! Why they take that effort ? Are you aware of the latiude boost it provides ?


A small amount of searching on here will tell you accurately about the benefits and shortcomings of polydisperse emulsions (and yes, Kodachrome was polydisperse historically - one emulsion per colour - what you are trying to claim about latitude is inherent to transparency film design). Latitude via multi-emulsion blends is a matter for worrying about only once you have a properly monodisperse emulsion - polydispersity by its nature gives an inherent wider latitude. Multi-layer is for when you run into problems with emulsion being so dissimilar in iodide content that you get equilibration problems, or that the internal layer turbidity is worse than interlayer reflectivity when you have attempted to control either through the use of absorber dyes.

Let me repeat it a final time: The core technology of the TMax films (TMX, TMY, TMY-II, TMZ) is laying extraordinarily well aligned ultra flat crystals over cubic crystals, all TMax films sport this core concept to provide insane latitude, insane linearity and fine grain.

With TMX you easily get 14 stops, or 18 with a careful processing, For that you require multi layer. If you blend the cubic in the T you destroy the alignment provided by the SHEAR THINNING (https://en.wikipedia.org/wiki/Shear_thinning) in the coating cascade, but you need the cubic emulsion for the linear extreme highlight latitude, so you won't find a way to avoid multi-layer for the TMX if wanting those 14 to 18 stops.

It is totally ridiculous your theory of TMX being a totally different concept than TMY, beyond speed, they are different flavors of the same, sharing linearity and insane latitude. You won't find a serious evidence stating that TMX is single layer, this is a urban legend with no base.

Your deduction that TMX is single layer because of glass plates is simply ridiculous, this is not being aware of the core technology that shapes the TMax product range.

I won't say more about this off-topic.

Basic methods of how to adjust the curves of a blended multi-emulsion T-grain (or any other monodisperse grain character film) have actually been disclosed/ strongly hinted at on this site (silver and iodide adjustments between the emulsions - finer grain emulsions gain contrast faster than coarser grained ones). But you either don't know what you are looking for or refuse to learn. You are attempting to blather your way out of another mess of your own making by throwing around randomly googled rheological terminology without considering that you are making fundamentally self-contradictory statements: if your claims are correct, how can 5 separate colour emulsions (let's say green sensitive) be blended into three layers for coating when they might all contain high aspect ratio/ t-grains? Because that is what is done in reality. Quite apart from the fact that the T-Max 400 electron microscope image clearly shows several quite different T-grain sizes within what you have spent a lot of words claiming to be one discreet layer...

And if you had done your primary source research properly, you'd have found that when T-Max films were first introduced, it was T-Max 400 that was regarded as the most challenging of the three in its highlight behaviour. That has changed, following multiple revisions. Consequently, applying blanket statements to all three T-max films as they currently exist is inaccurate. Absent further information to the contrary, the most likely constructions are: T-Max 100, single layer, multiple T-grain emulsions; T-Max 400, multilayer, one using high aspect ratio T-grain emulsion(s), the other more 3D emulsion(s); T-Max 3200, multilayer, multiple t-grain emulsions. This would be in line with industry norms for these speeds of film (which essentially means 'we copy what Kodak does - with good reason').
 
Last edited:

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
Ah, yes, packet couplers - it seems that it never proceeded much after multilayer curtain coating showed itself to be workable in the very early 1950's - I was getting confused by the way that there seemed to be an implication that it was a late 70's research project - from Ron & Bob's comments, packet coupler work for papers seems to have carried on into the 1960's, then multilayer fully won out.

There are some really interesting papers from the early 70's looking at image structure/ content - the Mike Kriss one Ron often referred to as being 'seminal' in its impact takes some tracking down, but it pretty clearly shows the analytical framework from which the information in those datasheets is derived. If you enjoy 60 pages of calculus that is.
You know, it's quite possible that I misunderstood, or some kind of misremembering was at play from either of the sources.
It is pretty thin ground to build a theory on. :-S
Listening through the video now, I have a hard time finding the place where he mentions single layer Kodachrome. I could have sworn it was there I heard about it though.
I am certain that I heard a reputable source mention it, and my Kodak correspondent agreed.

Is packet couplers the same as the coupler dye that doesn't affect other crystals, that Hanson mentions in the video?
 
  • Helge
  • Helge
  • Deleted
  • Reason: Ad hominem and responses

Lachlan Young

Member
Joined
Dec 2, 2005
Messages
4,943
Location
Glasgow
Format
Multi Format
You know, it's quite possible that I misunderstood, or some kind of misremembering was at play from either of the sources.
It is pretty thin ground to build a theory on. :-S
Listening through the video now, I have a hard time finding the place where he mentions single layer Kodachrome. I could have sworn it was there I heard about it though.
I am certain that I heard a reputable source mention it, and my Kodak correspondent agreed.

Is packet couplers the same as the coupler dye that doesn't affect other crystals, that Hanson mentions in the video?

Yes - it's a single layer, it's in the chapter on Kodachrome in the 2nd edition of Bob Shanebrook's book - Ron was the source for research on packet couplers in paper emulsion making continuing into the 1960's. This is the Godowsky patent in question I think. Bob's source for some of the story seems to be Paul Crough who was seconded from Kodak to assist Godowsky.

[Response to ad hominem remark deleted.—moderator]
 
Last edited:
  • MattKing
  • MattKing
  • Deleted
  • Reason: Ad hominem and responses

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
Yes - it's a single layer, it's in the chapter on Kodachrome in the 2nd edition of Bob Shanebrook's book - Ron was the source for research on packet couplers in paper emulsion making continuing into the 1960's. This is the Godowsky patent in question I think. Bob's source for some of the story seems to be Paul Crough who was seconded from Kodak to assist Godowsky.

[Response to ad hominem remark deleted.—moderator]
Thank you!
I think the moderator needs a recap on literary history.
It was Lachlan I tongue in cheek compared to Don Quixote. And that was only meant as a compliment. It's hard to find a more idealistic, resourceful and sympathetic character in classic litterature.
 

runswithsizzers

Subscriber
Joined
Jan 19, 2019
Messages
1,758
Location
SW Missouri, USA
Format
Multi Format
We can confirm that from our own countless resolution tests over the years.
The articles of Clarke and Tim Vitale unfortunately have many errors and give overall a wrong assessment of the performance of film. We cannot recommend them.

And they deserve it. The work of Tim Parkin, Henning Serger and Carl Zeiss belong by far to the best of what has been published so far concerning that topic.
We can confirm their results from our own work for years in that area.
ADOX is offering the highest resolving photographic medium for many years now (CMS 20 / CMS 20 II). And with HR-50 / Scala 50 we have another film in our product portfolio which offers very high resolution.
And of course a huge amount of tests is needed to evaluate such materials and their performance. Including all the imaging chains our (potential) customers would probably use.
For the best possible evaluation of film resolution we are using a microscope. With that you really see what incredible detail is recorded by excellent films.
Then further tests are made with classic optical enlargements: We've found that APO enlarging lenses have an outstanding performance, very close to microscope results. Users of negative film get the best performance using this workflow. It's surpassing even drumscanners by a very significant margin.
The same is true for slide projection: Performance with the best projection lenses is about on the same level as the APO enlarging lenses. As we are offering two BW reversal films we've done lots of tests in this field as well.
With these two workflows film users get the best resolution values for the final pictures. We highly recommend them. Also because you get the best performance by very low costs (often lower compared to scanners).

The second best performance concerning resolution are offering drumscanners. Really very good performance, sufficient for huge prints, exhibition and gallery work. Drumscanners are unmatched in the field of shadow and highlight detail capture and recovery (e.g. if you have done a mistake in exposure). That is due to their PMT technology.
We have some drumscan examples on our website, e.g some made by the leading German drumscan company www.high-end-scans.de. The expert behind this company, Mr. Ventzke, has an excellent reputation.

After that, on the next lower resolution level you'll find scanners like the last Nikon Coolscan series (5000; 9000), the Reflecta RPS 10M (and his Prime Scan sibling), and minilab scanners like the Noritsu HS-1800. There are some smaller resolution differences between them, but not huge ones.
We will probably also do some tests in the coming months with the new Plustek Opticfilm 120 Pro.

Next lower resolution level are scanners like the Reflecta MF 5000 (multi-format scanner) or the Plustek 8200 (35mm).

And at the bottom, with the by far worst detail resolution performance, there are the flatbed scanners.

We've done a huge amount of different tests with all these options in the imaging chains over the years (and continue to do so). And the ranking above is the result of these numerous tests.

ADOX - Innovation in Analog Photography.

Thanks so much for sharing your results and ranking the various scanning technologies!

Does ADOX have any plans to test resolution by the copy-with-digital-camera method? It would be interesting to see how that method compares with the others, and it would fit in with your goal of "Including all the imaging chains our (potential) customers would probably use."
 

SCHWARZZEIT

Member
Joined
Sep 6, 2008
Messages
12
Format
Multi Format
Thanks so much for sharing your results and ranking the various scanning technologies!

Does ADOX have any plans to test resolution by the copy-with-digital-camera method? It would be interesting to see how that method compares with the others, and it would fit in with your goal of "Including all the imaging chains our (potential) customers would probably use."
I would take a ranking based on these categories with a grain of salt as there are too many variables. For example, the high-end flatbed scanners such as the Creo/Scitex/Kodak Eversmart or Screen Cezanne lines are very capable film scanners and definitely not the worst method of digitization. And the top Imacon/Hasselbad Flextight scanners have even slightly better resolution than my ICG 370HS drum scanner from 35mm scans. But having seen many Flextight scans so far I have not seen one at the highest resolution that did not have any issues with having all of the image area consistently in focus. Obviously, having the film perfectly flat and parallel to the sensor is a challenge for 8000 ppi resolution, and it showed in the Flextight samples I've seen. For wet mounted drum scans the focus of the scan is mostly very consistent across the full image area. Drum scans can show other issues at high resolution like jittered (slightly misaligned) pixel lines.

In response to Adox's comments I wrote a post and included a camera-based digitization sample showing the resolution potential of Adox CMS 20 much better than a drum scan could. But my post took a day or so for the moderators to approve and therefore only appeared on the previous page while the thread had moved on. Here is a link in case you haven't seen it:
https://www.photrio.com/forum/threads/film-vs-scanning-resolution.177544/page-7#post-2315135

As I mentioned in that post, camera-based scanning is highly modular and thus the quality really depends on the quality of the components used, the precision of the alignment and the skills of the operator. Compared to dedicated film scanners, those film copy rigs are open systems with many options for fine tuning but also many places where things can go wrong. As the quality of camera scans can vary so much from system to system, from low-end setups based on a smartphone camera to very high-end solutions that exceed the quality of drum scans, it doesn't make sense to fit the method of camera-based scanning into a ranking versus other methods. I would just say that the potential of camera-based film scanning is higher than any other method of digitization.

-Dominique
 

runswithsizzers

Subscriber
Joined
Jan 19, 2019
Messages
1,758
Location
SW Missouri, USA
Format
Multi Format
I would take a ranking based on these categories with a grain of salt as there are too many variables. For example, the high-end flatbed scanners such as the Creo/Scitex/Kodak Eversmart or Screen Cezanne lines are very capable film scanners and definitely not the worst method of digitization. And the top Imacon/Hasselbad Flextight scanners have even slightly better resolution than my ICG 370HS drum scanner from 35mm scans. But having seen many Flextight scans so far I have not seen one at the highest resolution that did not have any issues with having all of the image area consistently in focus. Obviously, having the film perfectly flat and parallel to the sensor is a challenge for 8000 ppi resolution, and it showed in the Flextight samples I've seen. For wet mounted drum scans the focus of the scan is mostly very consistent across the full image area. Drum scans can show other issues at high resolution like jittered (slightly misaligned) pixel lines.

In response to Adox's comments I wrote a post and included a camera-based digitization sample showing the resolution potential of Adox CMS 20 much better than a drum scan could. But my post took a day or so for the moderators to approve and therefore only appeared on the previous page while the thread had moved on. Here is a link in case you haven't seen it:
https://www.photrio.com/forum/threads/film-vs-scanning-resolution.177544/page-7#post-2315135

As I mentioned in that post, camera-based scanning is highly modular and thus the quality really depends on the quality of the components used, the precision of the alignment and the skills of the operator. Compared to dedicated film scanners, those film copy rigs are open systems with many options for fine tuning but also many places where things can go wrong. As the quality of camera scans can vary so much from system to system, from low-end setups based on a smartphone camera to very high-end solutions that exceed the quality of drum scans, it doesn't make sense to fit the method of camera-based scanning into a ranking versus other methods. I would just say that the potential of camera-based film scanning is higher than any other method of digitization.

-Dominique

I did see your previous post, but, at the time, because the sample image was described as being from a "HXY Scanner" I did not realize that your sample image was "camera-based." After re-reading your post, I now see you plainly say so. Without knowing more details about your HXY Scanner I cannot guess how it may fit into the wide spectrum of other camera-based film copy solutions. Is your HXY Scanner something you bought, or made, or ?

I certainly agree that generalizations are not very useful in this situation, and I agree that the quality of results from camera-scans are quite variable. For such comparisons to be useful, testing equipment and variables would need to be plainly stated - but isn't that equally true when testing film scanners? In their brief summary of testing results, Adox did mention some specific models of film scanners (useful), but did not name what brands and models were chosen for flatbed scanning, making their generalization much less useful. Adox made a reference to only one drum scanner, without mentioning the brand by name - but that is of less concern to me because I am not likely to ever own a drum scanner.

If the goal of Adox was to survey and compare "... all the imaging chains our (potential) customers would probably use," then I don't see how camera-scanning can be ignored, despite the difficulty of doing meaningful testing. I would speculate that a higher percentage of Adox customers are using cameras to digitize their negatives than are using drum scanners, but that is a wild guess on my part.

Perhaps Adox could define two or three camera scanning test rigs, such as Budget, Mid-range, and High-end? Just as with their testing of drum scanners, film scanners, and flatbeds, there would need to be a certain amount of trust on our part that Adox has the resources and experience to follow best practices for each method.
 
Last edited:

138S

Member
Joined
Dec 4, 2019
Messages
1,776
Location
Pyrenees
Format
Large Format
Sounds like an argument for TMAX over more traditional films like Tri-X. Is there a downside to TMAX?

Phil, IMO TMax films are technically more advanced, but also IMO TMax lacks the TX cultural/aesthetic heritage that TX has accumulated since WWII. After releasing TMax kodak tried to discontinue TX but a legion of photogarphers would have set fire to the kodak buildings... so they abandoned the idea... :smile:

TMax is a bit less suitable than TXP (in special) for those following the Zone System recipes and printing in the darkroom. In the Zones System we make a visualizarion of the final print before shutter release, in that visualization some Zones are "linear" but extreme zones are compressed, this is the classic way to fit an scene with oversized dynamic range in the paper's limited range...

Instead TMax is a bit more suitable for "the JohnSexton way", consisting in taking a flexible linear capture and later manipulating the print what necessary, instead making a negative that's easy to print like we want.

Then we have grain, some hate grain and some love the aesthetics of the grain, of course. TX is very dramatic because it sports more grain the shadows, but here all depenmdes on the format. Prsonally I feel that TX shines in MF, in general I feel TX grain a bit excessive for 35mm but by MF it delivers a relative grain size I like. For LF grain is more dificult to see...

The Darkroom Cookboock compains a bit about TMax, but I guess that debate is over. This is about mastering our tools and having fun, it is the indian and not the arrow.
 
  • 138S
  • 138S
  • Deleted

PhilBurton

Subscriber
Joined
Oct 20, 2018
Messages
467
Location
Western USA
Format
35mm
Phil, IMO TMax films are technically more advanced, but also IMO TMax lacks the TX cultural/aesthetic heritage that TX has accumulated since WWII. After releasing TMax kodak tried to discontinue TX but a legion of photogarphers would have set fire to the kodak buildings... so they abandoned the idea... :smile:

TMax is a bit less suitable than TXP (in special) for those following the Zone System recipes and printing in the darkroom. In the Zones System we make a visualizarion of the final print before shutter release, in that visualization some Zones are "linear" but extreme zones are compressed, this is the classic way to fit an scene with oversized dynamic range in the paper's limited range...

I go back far enough that I once heard Ansel Adams talk. I don't remember any of the details, but he gave his talk in a large auditorium that was packed.

About 40 years ago, I could have bought a nice-sized print of "Moonrise over Hernandez" for under $1000. I kick myself sometimes, but at that time I was living a low-budget life. Back then I read all of Ansel's books, and wanted to try the Zone System, and even read some of Fred Picker's books. But since I have been out of the darkroom, even for film development, for many years now, I never got a chance to apply the Zone System to roll film. I did, however, buy a Pentax Spotmeter from eBay, just in case.

That all said, if I can get back into B&W film photography after developing the 60+ rolls of film my freezer, I don't know if I will go for the Zone System or the John Sexton approach, since I will have to scan the negatives either way. So I don't know right now what I will do when I can shoot B&W again. I still have my Nikon F2 body, and even had it reconditioned a few years ago.
Instead TMax is a bit more suitable for "the JohnSexton way", consisting in taking a flexible linear capture and later manipulating the print what necessary, instead making a negative that's easy to print like we want.

Then we have grain, some hate grain and some love the aesthetics of the grain, of course. TX is very dramatic because it sports more grain the shadows, but here all depenmdes on the format. Prsonally I feel that TX shines in MF, in general I feel TX grain a bit excessive for 35mm but by MF it delivers a relative grain size I like. For LF grain is more dificult to see...

The Darkroom Cookboock compains a bit about TMax, but I guess that debate is over. This is about mastering our tools and having fun, it is the indian and not the arrow.
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
Comparing Tmax 100 and Tmax 400, how large an print would there have to be to see practical differences in resolution, tones, or anything else?
That's always a tough one. 400 has without doubt more speed/dynamic range/more tonal range than 100.
100 has the edge with regards to resolution.
I bet there are people who actually find 400 sharper though, that is, it has better acutance.

The psycho optic response to the above parameters doesn't change linearly with linear change in size of reproduction though.

For a really large print I think 100 would win.
I never had the chance to directly compare though.
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
This is not I was told when I was an student...

View attachment 253326
https://en.wikipedia.org/wiki/Nyquist–Shannon_sampling_theorem

Of course reconstruction depends on the nature of the function, but an image is a "worst case" because the function is totally arbitrary, still we may consider that our function has a Fourier Transform, to apply that theory body. That "worst case" scenario implies that you need that safety margin in the samplling rate to not loss too much.

If you downsize to 2600dpi a larger image that is 2600 effective you will lose the 2600 performance. Try to sample a 20 line pairs (40lines) with 40 pixels, you'll get pure aliasing !!!

View attachment 253329





Again, from a super pixel-peeping, see how many pixels an edge takes for the transition, this not worth the half, clearly under 2600.

View attachment 253328
You manage to prove yourself wrong in reverse order within this post. Take a look again.
You never ever want an abrupt square waveish transient in a scanned image, or any kind of real world sampling. That would mean aliasing and beating as you touch on in the first half. And would mean that you need to filter, or increase sampling rate.
There is clearly more information in the image. The scanner just wasn't up to the job of extracting it.
You read the the sampling theorem and information theory like the devil reads the bible.

You might want to look through this one: https://www.wescottdesign.com/articles/Sampling/sampling.pdf

In general I'm reminded of this famous scene:



In conclusion, films resolution is in the common "web mind", for lack of a better term, vastly underrated.
I don't for the life of me, see why you persist with your crypto-misantropic flim-flam shenanigans, in the face of overwhelming evidence to the contrary‽
 
Last edited:

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
That would be me. Thanks for the kind words.

As I have mentioned in my previous post, I have developed a high-end camera-based film digitization system (HXY Scanner) that outresolves any drum scanner and is also superior in dynamic range (high density rendition and shadow recovery) and color fidelity.

Drum scanners had been the best quality solution for film digitization for a long time. But their development stopped about 20 years ago while CMOS sensor technology matured to a level where it can match and exceed the quality of PMT-based digitization. Camera scanning is a highly modular field with a vast range of options from very affordable entrance DIY setups over highly sophisticated custom rigs and up to very expensive turn-key solutions for cultural heritage institutions as offered by Digital Transitions and Phase One.


Since my ICG 370HS drum scanner is optically limited to roughly half of its specified 12000 ppi resolution, the 8000 ppi representation is as good as the ICG can do. Some degree of oversampling yields slightly better results than scanning at 6000 ppi only.

Here is a sample of a 35mm Adox CMS 20 frame, shot handheld with the Canon EF 15mm Fisheye on a Canon EOS 5 in 2009, developed in the original Adotech and digitized recently with my HXY Scanner at 24110 ppi (almost 800 MP) showing much more of the film’s resolving power than a drum scan:

DV09_ADOX_CMS20_30_res144.jpg


Detail crop of the original 24110 ppi scan:

DV09_ADOX_CMS20_30_24110ppi.jpg


-Dominique
WOW!
There has to be at the very least the equivalent of 200 real megapixels there?
The higher scan resolution isn't nugatory of course, since it supports the structure of the film and avoids artefacts.

Do you have any details you can/want to share on your HXY setup?
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
We can confirm that from our own countless resolution tests over the years.
The articles of Clarke and Tim Vitale unfortunately have many errors and give overall a wrong assessment of the performance of film. We cannot recommend them.



And they deserve it. The work of Tim Parkin, Henning Serger and Carl Zeiss belong by far to the best of what has been published so far concerning that topic.
We can confirm their results from our own work for years in that area.
ADOX is offering the highest resolving photographic medium for many years now (CMS 20 / CMS 20 II). And with HR-50 / Scala 50 we have another film in our product portfolio which offers very high resolution.
And of course a huge amount of tests is needed to evaluate such materials and their performance. Including all the imaging chains our (potential) customers would probably use.
For the best possible evaluation of film resolution we are using a microscope. With that you really see what incredible detail is recorded by excellent films.
Then further tests are made with classic optical enlargements: We've found that APO enlarging lenses have an outstanding performance, very close to microscope results. Users of negative film get the best performance using this workflow. It's surpassing even drumscanners by a very significant margin.
The same is true for slide projection: Performance with the best projection lenses is about on the same level as the APO enlarging lenses. As we are offering two BW reversal films we've done lots of tests in this field as well.
With these two workflows film users get the best resolution values for the final pictures. We highly recommend them. Also because you get the best performance by very low costs (often lower compared to scanners).

The second best performance concerning resolution are offering drumscanners. Really very good performance, sufficient for huge prints, exhibition and gallery work. Drumscanners are unmatched in the field of shadow and highlight detail capture and recovery (e.g. if you have done a mistake in exposure). That is due to their PMT technology.
We have some drumscan examples on our website, e.g some made by the leading German drumscan company www.high-end-scans.de. The expert behind this company, Mr. Ventzke, has an excellent reputation.

After that, on the next lower resolution level you'll find scanners like the last Nikon Coolscan series (5000; 9000), the Reflecta RPS 10M (and his Prime Scan sibling), and minilab scanners like the Noritsu HS-1800. There are some smaller resolution differences between them, but not huge ones.
We will probably also do some tests in the coming months with the new Plustek Opticfilm 120 Pro.

Next lower resolution level are scanners like the Reflecta MF 5000 (multi-format scanner) or the Plustek 8200 (35mm).

And at the bottom, with the by far worst detail resolution performance, there are the flatbed scanners.

We've done a huge amount of different tests with all these options in the imaging chains over the years (and continue to do so). And the ranking above is the result of these numerous tests.

ADOX - Innovation in Analog Photography.
Thank you for that answer Team ADOX! You really are invaluable to film users.
I really should shoot more CMS 20 II. Really there is very little excuse once you are in daylight. Else just use a tripod and/or external flash.

Is there any online source for the Zeiss tests you write about?
I've heard of them for years, but have never been able to find them, in print or online.
 

Team ADOX

Partner
Joined
Mar 11, 2019
Messages
318
Location
Germany
Format
Multi Format
Thanks so much for sharing your results and ranking the various scanning technologies!

You are welcome.

Does ADOX have any plans to test resolution by the copy-with-digital-camera method? It would be interesting to see how that method compares with the others, and it would fit in with your goal of "Including all the imaging chains our (potential) customers would probably use."

There is quite a momentum recently in this technical field. We are in contact with several specialists concerning that topic, discussing methods and exchanging results. The current best solutions are unfortunately very expensive. But we will certainly explore more in that direction.

ADOX - Innovation in Analog Photography.
 

Team ADOX

Partner
Joined
Mar 11, 2019
Messages
318
Location
Germany
Format
Multi Format
I would take a ranking based on these categories with a grain of salt as there are too many variables.

Not needed at all. Our post was a short summary of the relevant (main market relevance) options. Not more, not less. Please see below:

For example, the high-end flatbed scanners such as the Creo/Scitex/Kodak Eversmart or Screen Cezanne lines are very capable film scanners and definitely not the worst method of digitization.

Completely agreed!
Our category "flatbed scanners" refers to the Epson and Canon flatbed scanners, which represent 99,999% of all flatbed scanners on the market.
The high-end Heidelberg, Creo, Scitex etc flatbed models represent only an extremely tiny niche. And are mostly used in professional environments.
Of course we could have more differentiated. But there is always the trade off of having a shorter post which is read by the members, or having an extremely detailed post being very long and therefore ignored by most readers..........

And the top Imacon/Hasselbad Flextight scanners have even slightly better resolution than my ICG 370HS drum scanner from 35mm scans. But having seen many Flextight scans so far I have not seen one at the highest resolution that did not have any issues with having all of the image area consistently in focus. Obviously, having the film perfectly flat and parallel to the sensor is a challenge for 8000 ppi resolution, and it showed in the Flextight samples I've seen. For wet mounted drum scans the focus of the scan is mostly very consistent across the full image area. Drum scans can show other issues at high resolution like jittered (slightly misaligned) pixel lines.

That is also absolutely in-line with our results over the years. Completely agreed again.
But the resolution advantage of the latest Flextight is quite small, and real drumscanners like the Heidelberg Tango or the ICG 370HS have also advantages in capturing shadow and highlight detail and grain rendition.

ADOX - Innovation in Analog Photography.
 

Team ADOX

Partner
Joined
Mar 11, 2019
Messages
318
Location
Germany
Format
Multi Format
Thank you for that answer Team ADOX! You really are invaluable to film users.
I really should shoot more CMS 20 II. Really there is very little excuse once you are in daylight. Else just use a tripod and/or external flash.

Is there any online source for the Zeiss tests you write about?
I've heard of them for years, but have never been able to find them, in print or online.

You are welcome.
The Zeiss results were published in their former editions of "Camera Lens News". They were also online for many years, but not anymore.

ADOX - Innovation in Analog Photography.
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
There is quite a momentum recently in this technical field. We are in contact with several specialists concerning that topic, discussing methods and exchanging results. The current best solutions are unfortunately very expensive. But we will certainly explore more in that direction.
One of the best things about camera scanning is that you can get ninety percent of the way there with very modest equipment.
An old DSLR body is more than good enough. A slow but good macro or enlarger lens. A copy stand or enlarger that can be converted to copy stand and a good light source and you are good to go.
The most important thing is rigidity, so you don't have to adjust stuff too much between shots.
Microscopically small deviations mean the difference between average flatbed resolution and better than drumscan.

If someone where to make a commercial scanner, I'd think a contact print approach would be the best and simplest solution.
Pressing a very thinly protectively coated sensor on the film, with highly collimated backlight, with both in a fluid scanning bath would seem to cut out a lot of the fat and potentially image degrading factors.
Flat film, no lens and modest phase change veiling sounds like the ticket.
A 12 MP monochrome CMOS phone sensor if stepped over a 135 Barnack frame would yield in the neighborhood of 250MP.
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
You are welcome.
The Zeiss results were published in their former editions of "Camera Lens News". They were also online for many years, but not anymore.

ADOX - Innovation in Analog Photography.
Do you have the issue number?
And/or alternatively do you have the address of the webpage the results where on? Archive.org might have a backup.
 

SCHWARZZEIT

Member
Joined
Sep 6, 2008
Messages
12
Format
Multi Format
I did see your previous post, but, at the time, because the sample image was described as being from a "HXY Scanner" I did not realize that your sample image was "camera-based." After re-reading your post, I now see you plainly say so. Without knowing more details about your HXY Scanner I cannot guess how it may fit into the wide spectrum of other camera-based film copy solutions. Is your HXY Scanner something you bought, or made, or ?
I designed and built the HXY Scanner around my Phase One IQ4 digital back and high-resolution industrial optics. The HXY naming comes from the H-shaped design of the copy stand where the camera is mounted on a horizontal rail between two columns and the XY capture stage that allows the film carrier to be shifted in X and Y direction for stitching film formats up to 12x20” at any resolution. It is a high-end solution optimized for ultimate quality film digitization that I offer as a commercial service.

While I originally planned to replace my drum scanner with just the HXY scanner, it turned out the difference in rendering from the softer tonal transitions of the drum scanner are reason enough to keep the drum scanner as an option for clients preferring that particular output. Quality is not always about resolution. And sometimes a softer rendering with appealing character is preferable over an accurate sharply rendered presentation of the film grain where the added resolution leads to harsher tonal transitions.

When I finally get to updating my website (www.high-end-scans.de) and add the HXY scanning options I will have a set of comparison files available for download.

WOW!
There has to be at the very least the equivalent of 200 real megapixels there?
The higher scan resolution isn't nugatory of course, since it supports the structure of the film and avoids artefacts.
Well, the finest details captured on the negative require a much higher pixel density than 12200ppi (which is about 200 MP at 36x24mm). Though speaking of a film’s megapixel equivalence is always a bit misleading as it somehow suggests a comparison with a digital sensor. A native 200 MP digital image would look much cleaner and be more detailed for low contrast image content. But the peak resolution of Adox CMS 20 for high contrast details is extremely high and beyond the resolving power of a potential 200 MP full frame sensor.

-Dominique
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
Well, the finest details captured on the negative require a much higher pixel density than 12200ppi (which is about 200 MP at 36x24mm). Though speaking of a film’s megapixel equivalence is always a bit misleading as it somehow suggests a comparison with a digital sensor. A native 200 MP digital image would look much cleaner and be more detailed for low contrast image content. But the peak resolution of Adox CMS 20 for high contrast details is extremely high and beyond the resolving power of a potential 200 MP full frame sensor.

-Dominique

But a lot of the "cleanness" of the digital capture is guessed at by various algorithms, and perhaps lately recursive networks (I'm aware of the higher QE of electronic sensors, but that has been a known for decades before the proliferation of digital cameras, and is mostly overrated as an arbiter of IQ).
Resulting in "nice" clean mush at some levels in the frequency domain, and artificial harsh transients in other parts. I'm not just talking de-mosaicing here, but about the plethora of techniques for de-noising and antialiasing that is applied to the signal, very often also mandatory to the even the RAW file, and under any circumstance part of any "finished" digital photo.

It's something, like the sweetness of Pepsi vs Coke in the legendary Pepsi challenge, that the human visual system is inclined to like in smaller doses, because it makes the task of processing the scene easier.
But in larger doses it becomes cloying and irritating. If you still have something to compare to, that is.
Otherwise it just becomes the new normal and everything else becomes subtly weird and quaint.
That's where photography is now.

It would be interesting (and in short order very useful) to teach a neural network to subtract the grain and extract low contrast details from film. Essentially what our own visual system is kind of doing when we look through grain of a film photo.
 
Last edited:

138S

Member
Joined
Dec 4, 2019
Messages
1,776
Location
Pyrenees
Format
Large Format
You read the the sampling theorem and information theory like the devil reads the bible.

You might want to look through this one: https://www.wescottdesign.com/articles/Sampling/sampling.pdf

In general I'm reminded of this famous scene:



:smile: Nice video, it looks a personal attack but s funny one, you have a more refined style than Lachlan :smile:.

In a dayly basis I C++ hand code those kinds of things in firmware. Even I've been teaching this stuff in a college many years ago, right know I'm implementing an adaptative FIR filter, with varies the finite impulse response in real time, optimizing the FIR configuration as it is optimal.


Let me repeat my reasoning:


What I exactly say is that if we take a 5200 dpi image worth 2600 dpi effective and we reduce size (sampled) to 2600 then we have an image thas worth less than 2600dpi effective, if we expand it to 5200 then this time we'll have loss compared to the original 2600 effective.

Do you agree with that or not ?
 

Helge

Member
Joined
Jun 27, 2018
Messages
3,938
Location
Denmark
Format
Medium Format
:smile: Nice video, it looks a personal attack but s funny one, you have a more refined style than Lachlan :smile:.

In a dayly basis I C++ hand code those kinds of things in firmware. Even I've been teaching this stuff in a college many years ago, right know I'm implementing an adaptative FIR filter, with varies the finite impulse response in real time, optimizing the FIR configuration as it is optimal.


Let me repeat my reasoning:


What I exactly say is that if we take a 5200 dpi image worth 2600 dpi effective and we reduce size (sampled) to 2600 then we have an image thas worth less than 2600dpi effective, if we expand it to 5200 then this time we'll have loss compared to the original 2600 effective.

Do you agree with that or not ?
Yes, but I’m not sure how that relates in a meaningful way to the question(s) at hand?

And it’s neigh on impossible not to get personal when replying to someone’s writing.
The detached, rational reply has been proven long ago academically to be a sham. So why pretend?
There is though a huge difference between being personal, and being nasty and only going for a reaction.
In danger of being insultingly obvious, posters can of course always take solace in that for most cases we don’t, and never will know the real “you” outside the infinitesimal part exposed here.
For better and worse. ;-)
 
Last edited:

138S

Member
Joined
Dec 4, 2019
Messages
1,776
Location
Pyrenees
Format
Large Format
Yes, but I’m not sure how that relates in a meaningful way to the question(s) at hand.

Let me explain it.


Thread Tittle: Film vs. Scanning resolution

1) Depending on film resolution you require more or less scanning resolution.

2) We were evaluating the resolution of an image posted by you, that clock...

3) Here someone says: "A simple test to check the effective level of detail of an image file is to downsize it (e.g. in Photoshop) to see at which resolution the image loses image content details. You can download that crop, load it into Photoshop, set the resolution to 5400 ppi (without resizing) and then resize to lower resolutions. Resizing the scaled down image back to the original resolution and adding appropriate sharpness will help seeing which details can be retained and which are lost in the process."

I pointed that downsizing to the checked resolution is not rights,a safety margin has to be accounted, IMO.

Then insisting: "The other crop of the Minolta 5400ppi scan contains finer details and some higher contrast lines that spread just over a single line of pixels indicating that the effective resolution and level of real detail in that area is very close to the sampling resolution."
(https://www.photrio.com/forum/threads/film-vs-scanning-resolution.177544/page-5#post-2314547)

Then I aslo provide evidence that your image is not 5400 worth but half: the number of pixels it takes the transition of an edge: https://www.photrio.com/forum/threads/film-vs-scanning-resolution.177544/page-5


If you downsize the image to 2600 and later you inflate it to 5400 you have some loss (try it), bit this is not because the image is worth more than 2600, this is because point 3) is wrong or a bit coarse precision.


So this may be interesting because we debated that the shrinking-inflating is not an accurate test and that we have the (super peeping) edge transition criterion to evaluate if we are scanning at overkill dpi, but first we have to evaluate the scanner's native edge transition to be aware of when the scanner can be the limiting factor.

But if one wants to debate this seriusly in Nyquist terms, I'd be happy...

______

Let me add another (relatively close) analogy. Human hearing is band limited to 20KHz, but as audio is an arbitrary signal and because of that CDs are sampled at 44.1KHz, around 2F0, you have also 96KHz and beyond... suggesting that 2600 dpi are not to record 2600 effective, as it hapens when requiring 20KHz but having to sample 44, we can go into the detailed math of that, if wanted.

______


All of this is interesting to realize what true scanning performance we require to exhaust what shots have. The Serger's Porsche in CMS and is challenging the hassie at its 35mm sweet point !! But this is not the usual situation, by far.
 
Photrio.com contains affiliate links to products. We may receive a commission for purchases made through these links.
To read our full affiliate disclosure statement please click Here.

PHOTRIO PARTNERS EQUALLY FUNDING OUR COMMUNITY:



Ilford ADOX Freestyle Photographic Stearman Press Weldon Color Lab Blue Moon Camera & Machine
Top Bottom