lens that can take advantage of hi resolution films like CMS20

[brighamr]

hi

I see that adox CMS20 can resolve 800 line pairs and is available in 4x5
are there any lenses in any format that can get any where near this
could microfilm cameras ?
what lens did they use ?

if not then
what is the point of such a film

robin

[Nodda Duma]

Yes a diffraction-limited (or close to) visible optic faster than f/2.3 will resolve that spatial frequency.

There are fast telescope optics out there which will do this (the Schmidt Camera comes to mind)

[brighamr]

Thanks Nodda

That's interesting
I'm now looking at microfilm cameras that formally photographed newspapers into 16mm film
Used in reverse it may well image jewellery onto 5x4 or even ULF

Found one for £6 online ... So we will see

Best

Robin

[tjvitale]

I have a friend who used to manufacture digital scanbacks. He tried to commission a lens that could match 10,000 pixels per 2.67" (3750 ppi). He couldn't find a vendor to do the job. Expensive telescope optics aside, microfilm photography is done but the strangest folks. I had some working for me when I was a department head at a government agency. There is little high quality going on there.

I doubt you can find a off-the-shelf lens to do the job.

I'd be happy to be wrong. Please show me!!

Tim Vitale
Conservator
Oakland, CA

[Nodda Duma]

Tim, I agree: Good luck finding an off-the-shelf lens. The only thing that comes to mind is a Schmidt camera setup.

[tjvitale]

Right, my friend couldn't find large format lenses beyond 80 lp/mm, and 100 lp/mm seemed to be vaporware. I saw some of the tests and they just couldn't hold up.

Is there some application that I don't know about using a telescope as a substitute for a large format lens on a view camera? Really, I don't know.

Tim Vitale
Conservator
Oakland, CA

[Nodda Duma]

I was about to say I have no idea then I remembered that I have a Kodak Aero-Ektar sitting at home lol. This is rather academic, but the need for aerial/satellite reconnaissance in the film era would have seen very fast (faster than f/2.7) telescope-type objectives corrected over a large film plane. Could they realistically resolve 800 lp/mm? I'd have to measure the optical MTF of an example to know for sure. Mine won't..it's an f/6. But something along those lines...designed to look far away at high resolution...would be heading down the right path.

Other than that, large observatory telescopes have really big image circles.

[Emmanuel BIGLER]

hi
I see that adox CMS20 can resolve 800 line pairs and is available in 4x5
are there any lenses in any format that can get any where near this
could microfilm cameras ?
what lens did they use ?

Hello from France.

I have personnally used in the past two kinds of highly specialized lenses capable of delivering well above 300 cy/mm on a pĥoto-chemical detector, silver halide plates or photoresist/chromium photomask plates.

The first example was a lens made in France by Cerco, a 100 mm f/4 lens dedicated to fabricating photomasks from an enlarged drawing on paper. Years ~ 1982 - 1984, our lab had bough a used repro bench from an institution who stopped making photomasks by this technique, already obsolete at the time.
The lens was a f/4 and was diffraction-limited, to be used in monochromatic green light (there was simply a green filter permanently mounted in front of the lens), and in a very limited range of magnifications like 1:20 or 1:10. Plates in use were 2-1/2" square, silver halide high-resolution glass plates with a resolution above 500 cy/mm.

Among glass plates in use at the time there was the Kodak high-resolution plate type 1A, blue sensitive only hence probably unsuitable for use in green monochromatic light, and the legendary Kodak spectrocopic plate 649-F also used for holograms.
I do not remember the manufacturer's specs for the lens resolution, but the absolute cut-off spatial frequency for a diffraction-limited f/4 lens is about 450 cy/mm for a wavelength of 550 nm (green).

This 100 mm f/4 Cerco lens did not cover the 4x5" format, only 2-1/2" by 2-1/2", but I cannot see any reason why it could not have been designed in a focal length of 250 mm by scaling up the 100 mm design.
Most probably, covering the 4x5" format, the resolution would have dropped down to about 200 or 300 cy/mm but by no way such a lens could resolve 800 cy/mm on a whole image field of 4x5".

The second example, years 1989-2000, was a 10X photorepeater lens by Carl Zeiss. Fixed magnification 1:10, object size = 10x10 cm (5" square photomask plates and 4" image field). The output object was a photo-repeated mask, photoresist+ chromium on glass plates.
I have no idea of the focal length and the f-number of this Zeiss lens, for sure the field angle was very narrow like in a telescope lens. It used mercury arc lines for illumination, the G-line at 435 nm and the H-line at 405 nm.
The resolution limit, on a field of 1x1 cm, was 1.4 micron for the ultimate period (0.7 white bar plus 0.7 adjacent dark bar = period of 1.4) and the corresponding spatial frequency is 1000/1.4 = 700 cy/mm. Results obtained only when the machine was properly tuned, and this demanded quite an effort by doing time-consuming crossed-tests for best focus and best exposure time.
The effective f-number N for a diffraction-limited lens capable of passing 700 cy/mm @400 nm is N = 1.4/0.4 = 3.5 ( a f/3.5 lens).

But the requirements were: monochromatic blue or near-UV light ; fixed magnification, and the image field was 1cm square only.
But it can certainly be used reversed with a 1cm object to make an enlarged image on a 4x5" sheet of film ... provided that you are able to properly focus the image ;)
But the resolution limit on the 10x image covering a 4x5" field would only be 70 cy/mm... any good film could be used in this application, no need for a 800 cy/mm microfilm.




if not then
what is the point of such a film
robin

This is a zero-grain film for those who hate the horrible granularity of Tri-X and HP5 ;)

[tjvitale]

I've heard of these (generic) fabled lenses. Thanks for all of that.

Tim Vitale
Conservator
Oakland, CA

[Jac@stafford.net]

[... snip great article ...]
This 100 mm f/4 Cerco lens did not cover the 4x5" format, only 2-1/2" by 2-1/2", but I cannot see any reason why it could not have been designed in a focal length of 250 mm by scaling up the 100 mm design.


Can a lens be scaled up, retaining the proportions to the same resolution? I really do not know, and having wrestled with the issues of down-scaling aircraft as a young man I usually question intuition. I disassembled a Pacific Optical 3" Biogon and found it had fewer elements than the 38mm. (I've no idea whatsoever of the PO's resolution. Never even tried to look it up.)

Thanks for the great article!

[Dan Fromm]

Can a lens be scaled up, retaining the proportions to the same resolution? I really do not know, and having wrestled with the issues of down-scaling aircraft as a young man I usually question intuition. I disassembled a Pacific Optical 3" Biogon and found it had fewer elements than the 38mm. (I've no idea whatsoever of the PO's resolution. Never even tried to look it up.)

Thanks for the great article!

No. Everything scales with size, including aberrations. When a prescription is scaled up, resolution attainable given aperture falls.

About the Pacific Optical "Biogon." USAF bought at least two different 3"/4.5 lenses from Pacific Optical, identified in my USAF data sheets as Paxar (A) and Paxar (B).

If I'm reading the cross-sections correctly, (A) has 8 elements in 5 groups. AWAR on Plus X Aerecon with a 1000:1 target and a W-12 filter is 31 lp/mm @ f/4.5

(B) has 9 elements in 5 groups. AWAR on AWAR on Plus X Aerecon with a 1000:1 target and no filter mentioned is 52 lp/mm @ f/4.5. On S.O. 243, same conditions, 121 lp/mm

The 38/4.5 Biogon as fitted to Hasselblad SWx cameras and to some aerial cameras (mine came from an AGI F135) have 8 elements in 5 groups.

[Drew Wiley]

You're only as good as your weakest link. Your filmholder itself will not hold the film flat enough to worry about this kind of thing. So unless you are talking about
glass plates in a lab environment or some kind of precision vacuum film back, you're barking up the wrong tree. True microfilm has been cut in 4x5 sheets from time to time, but it's rather wretched stuff for normal pictorial work. There are certain graphics arts uses. What is your application for wanting this kind of film?
Also remember that lenses designed for things like mapping are not necessarily going to deliver that kind of performance when normal view camera tilts, swings,
and offsets are involved.

[Andy Eads]

Sharpness is a relative thing. Strictly speaking, it is the sum of all the links in the chain as has been mentioned. It is also partly psycho-physical in that we will perceive as sharper an image with greater contrast when viewing two photos of equal optical resolution. With all the laws of physics conspiring against a sharp photo, it's a miracle we get anything that looks good at all. Searching for great sharpness is a fine goal if you are trying to limit the size of the beastly camera you carry. For me, there are simple limits to how far I will carry the search. If it is interfering with the story I'm trying to tell, the quest for ultra sharpness is a distraction.

[Drew Wiley]

Astrophotographers, the military, and survelliance agencies are decades ahead of us when it comes to this game. There are also forensic and art history usages for special documentation films. Going out and taking pictures and wanting sharp results - heck, that's pretty hard NOT to achieve using any number of modern lenses and large format films. Just how big a print does someone want, anyway? The problems associated with depth of field and diffraction will qualify this horse race every time. But sheer area of sheet film generally wins. I wouldn't even want CMS20 in a 35mm camera - too many micro-zits, tonality issues. But what the heck.
Tech Pan was once cut even in 8x10 sheets. I still use this for pan lith masking in color printing. But for actual pictures? - you gotta be kidding.

[Jac@stafford.net]

No. Everything scales with size, including aberrations. When a prescription is scaled up, resolution attainable given aperture falls. [... snip helpful stuff ...]

Thank you very much for the information, Dan. I was hoping you would answer.

The model B has larger elements than the Zeiss 75mm Biogon. In fact the rear element is larger than the recess on the 4" side of a Sinar 4x5 back. As you may know, many (wow, too many) years ago I bought a lot of these as military surplus, cheap. Of the whole lot, only two were pristine. The rest sit on a back shelf in Tupperware boxes, one of them in many pieces. Maybe I can burn ants with the parts.

The Paxar (Pacific Optical 3") is not worth using on LF despite being inexpensive. They were designed for 5" roll film to be used at infinity. I spent, some would say wasted a lot of time making a camera for a B model. It is sharp wide open, as is expected of aerial recon/mapping lenses, but after borrowing a 75mm Biogon for the Technika, I could find only one advantage - slightly greater coverage due to the larger size of the rear lens cluster.

Current project is to noodle out a couple different electric leaf-type shutters for Metrogon ~152mm lenses ($25 each then.)

[koraks]

Astrophotographers, the military, and survelliance agencies are decades ahead of us when it comes to this game.
And, most notably, the photolithography industry. If you're looking for extreme resolution, you might want to start taking apart some decommissioned lithography tools. Which, of course, is a purely hypothetical thing. Anyone with a basic knowledge of this technology knows it would be horribly impractical for pictorial purposes even in a very well-equipped lab setting.

[Nathan Potter]

Thanks for the nice summary Emmanuel. During the 1960s and into the 70s I made use of a couple of types of Ultra Micro Nikkor reproduction lenses for mask making on Kodak High Resolution Glass Plates. Those glass plate were rated at about 2000 lp/mm capability. The Nikkors could achieve (as listed) some 800 lp/mm resolving power using the mercury 365 nm line. But the setup had to be rigged using a high precision optical bench on a vibration isolated table. The image size was limited to about 1 X 1 cm. or less. Those early Ultra Micro Nikkors have evolved into current stepper lenses now requiring deep UV light sources (even 190 nm laser sources). These lenses are still all small field (1 X 1 cm. or so) and designed for 4X to 10X demagnification ratios and are used by stepping a chrome on glass plate image across a silicon wafer and into deep UV sensitized photoresist, not silver emulsions. Current costs are in the $500K to $5 million range for the imaging lenses. All require narrow monochromatic light sources to achieve the nominal 1000 lp/mm resolution.

As far as I know there are no lenses that can deliver 800 lp/mm using a white light spectrum. Such performance can only be achieved using monochromatic light of narrow bandwidth.

Nate Potter, Austin TX.

[Dan Fromm]

The Paxar (Pacific Optical 3") is not worth using on LF despite being inexpensive. They were designed for 5" roll film to be used at infinity. I spent, some would say wasted a lot of time making a camera for a B model. It is sharp wide open, as is expected of aerial recon/mapping lenses, but after borrowing a 75mm Biogon for the Technika, I could find only one advantage - slightly greater coverage due to the larger size of the rear lens cluster.

JAC, those monstrosities make sense for people who need what they can do. Even with the cost of putting it in shutter one can cost less than the equivalent Zeiss-made 75/4.5 Biogon, an equally specialized lens.

Steve Grimes rated the PO lenses high enough to buy what he thought were the last three pallets of them. They were in beautiful condition and he was ideally situated to put them in shutters so visions of fortunes danced through his head. And then C&H popped up with more of the same.

[jp]

Maybe wet plate is where the action is at for pictorial high res stuff. It's UV (shorter wavelength) and does not have grain as it's liquid silver, not crystalised silver.

[Jac@stafford.net]

JAC, those monstrosities make sense for people who need what they can do. Even with the cost of putting it in shutter one can cost less than the equivalent Zeiss-made 75/4.5 Biogon, an equally specialized lens.

True. I forgot how expensive the Zeiss 75/4 is. Here is a PO in an Alphax #5 shutter (http://www.digoliardi.net/super-biogon-3-inch.jpg). With 20/20 hindsight it appears it could have been machined to take a smaller shutter. Both my #5 shutters have broken.

[Drew Wiley]

I've seen true-color film shots that make any lens you've ever seen look like a pathetic toy by comparison. Two things I wasn't allowed to see is the lens itself,
or what the film was. Let's just say whatever the film was made the resolution on something like Ektar or Velvia look like shotgun splatter by comparison. In fact,
I wasn't even allowed to know what kind of vessel or plane or whatever had the camera, but I suspect it was a surfaced sub.

[Jac@stafford.net]

I've seen true-color film shots that make any lens you've ever seen look like a pathetic toy by comparison. Two things I wasn't allowed to see is the lens itself,
or what the film was. Let's just say whatever the film was made the resolution on something like Ektar or Velvia look like shotgun splatter by comparison. In fact,
I wasn't even allowed to know what kind of vessel or plane or whatever had the camera, but I suspect it was a surfaced sub.

Seems like a legend to me. No image, no source, no facts, no nothing.
.

[Carsten Wolff]

True. I forgot how expensive the Zeiss 75/4 is. Here is a PO in an Alphax #5 shutter (http://www.digoliardi.net/super-biogon-3-inch.jpg). With 20/20 hindsight it appears it could have been machined to take a smaller shutter. Both my #5 shutters have broken.

Thats not an Alphax shutter either. :)

[Jac@stafford.net]

Thats not an Alphax shutter either. :)

What is it, then? Did I misspell?

[Dan Fromm]

Originally Posted by Carsten Wolff
Thats not an Alphax shutter either.
What is it, then? Did I misspell?

Ilex, not Alphax. Quite a bad spelling error

[Jac@stafford.net]

Ilex, not Alphax. Quite a bad spelling error

Ach! getting old sucks.

[Emmanuel BIGLER]

Hi!

Thanks all for the nice exchange regarding this ultimate quest for the highest possible number of cy/mm in a photo-chemical detector combined to an ultimate lens.

I fully agree with Drew Wiley that for our photographic uses, we prefer good tonality and a nice gray scale to no-grain. And those microfilms require such a very special low-contrast chemistry in order to deliver some acceptable gray scale. A burden, to say it bluntly: an we can no longer use our own beloved soups!

In other terms, at least for Kodak, there is a market for Tri-X film, but no longer for hi-res plates type 1a ;)

Well, not kidding, thanks to Nathan for reminding us the good ol' days of Kodak hi-res plates type 1A in a square 2"-1/2 size. At the time, I was told that type 1A was similar to 649-F, but non-chromatised, blue-sensitive only. I suspect that Kodak had the technology to make those plates before the laser era, and that people experimenting with the first holograms in the sixties naturally used them as commercially available hi-res plates.

It also should be mentioned that hi-res silver halide plates are old stuff, suffice to remember Lippmann plates, fabricated before 1900.

And also, in a totally different resolution range, that photographic paper being a very slow medium exhibits fine silver grains, I would say a few microns in size. But nobody cares for a visual examination of a print!

To continue with the good old days, I can say that actually I have used Hi-Res Kodak type 1A plates during my Ph.D. years, we exposed them to X rays with synchrotron radiation.
Colleagues told us: you are crazy! The plate will be fogged at its maximum density in a microsecond! But since this was not a X-ray film, with an emulsion thickness of a few microns, X-ray sensitivity was exceedingly low. Definitely unusable with a conventional X-ray source.
And exposed to X rays, the type 1A plated retained its no-grain capabilities, and, surprisingly, featured a certain amount of nice gray levels. The X-ray imaging process was contact imaging.
We processed the plates in red safelight with a GEPE COMBI-PLAN tray, with Kodak D-19, a standard high-contrast microfilm developer.
A no-grain silver-halide image seen through an optical microscope is fascinating, this is the reason why, in those years, for my personal photographic activities, inoculated with the no-grain-Hi-res virus, I only swore by AGFA APX 25 and never experienced Tri-X ;-)

I have recently browsed through the Harman-Ilford web site (actually looking for fresh info about their direct-reversal paper, which is back again, this is another story)
http://www.harmanexpress.eu/holo.html
and found the price list for their holographic plates ; the 2"-1/2 square glass plate standard is still in use, 2 mm thickness, and a box of 6 plates costs only £20.25 (excluding postage), about the same price as 4x5" colour slide cut film ;-)
Resolution specs are: 7000 cy/mm!!! But holography is also another game.
To the best of my knowledge, Slavich in Russia also sells holographic plates, hence the know-how behind those strange beasts is not lost.

And just for being very precise in the explanations, when Dan says (hello, Dan!)
Everything scales with size, including aberrations. When a prescription is scaled up, resolution attainable given aperture falls.
we could add that in an hypothetic diffraction-free word, only governed by geometrical optics, aberration spots would scale exactly like lens element radii and spacings. In this imaginary world, for a given lens prescription, scaling up the design would scale the focal length and the image size in direct proportion of the scaling factor, and would scale the achievable resolution exactly in the inverse proportion.

Diffraction works quite differently. In another imaginary world, free of all geometrical aberrations, scaling a design would not change the diffraction spot size at all, if we keep the same f-number!
Hence in this other dreamed world, bigger lenses would achieve the miracle of keeping the 450 cy/mm of the green-light-only 100 mm CERCO lens, up to focal lengths of 1000 mm for ULF use, covering an image field of 25"x25"!!

Our real world combines both physical phenomena ... plus some other discouraging, down-to-Earth practical things ... fortunately we do have discussion forums to let us dream of those imaginary optical worlds ;)

[Bill Burk]

I'd be happy to have a sheet of this film, on which a resolution test chart was step-and-repeated each 3/16ths of an inch, either as a full sheet full of these marks or a few columns, to use in contact printing test strips on my everyday film tests... So that I can include resolution information in my test results for the film I use every day.

[Dan Fromm]

I see that adox CMS20 can resolve 800 line pairs
are there any lenses in any format that can get any where near this ?

Hmm. Lessee now. The aperture at which the diffraction limit is around 800 lp/mm on axis is f/2. So, before we start to worry about practical matters like portability and shutters and off-axis performance, are there any f/2 and faster lenses that cover 4x5 or larger?

I looked in my USAF data sheets, which are incomplete. They list several lenses f/2 or so lenses as covering 4x5 but their focal lengths and claimed angular coverage come out smaller than 4x5. Then there's the 12"/1.5 Perkin-Elmar Recon 660 that covers 162 mm. Its AWAR (area weighted average resolution, gives more weight to far off-axis than near the axis) on Aerographic 5424 is 34 lp/mm. Not quite 800.

The French Air Force flew Omera aerial cameras, some with a 200/2 S.F.O.M. lens. I've had one, it wasn't easily usable. I saw one on offer that was engraved "Kinoptik" so its probably a decent lens but ...

Dallmeyer made a line of f/1.9 Super Six lenses, 6/4 double Gauss types. The longest two were 6" (late ones engraved 152 mm)/1.9 and 8"/2.0. Perhaps decent lenses but its hard to believe they cut 800 lines on axis wide open.

If there's a wonder lens for 4x5 or larger I'm not aware of it. Can anyone direct us to one?

At the tiny tiny end, I recently used a Leica stereo microscope. Sorry, not sure which model, but a very recent one. I liked it so much that I fantasized buying one, even looked in Leica catalogs. Their best will resolve slightly more than 1,000 lp/mm. "This resolution means you can see details that are smaller than a hundredth of the diameter of a human hair." Naturally I can't afford one.

Hi, Emmanuel.