'repro' lens question

[PaulFurman]

I'm looking at a specialty 'repro' type lens (description somewhat obscured), it's about 100mm, fairly fast & optimized for 1:1 with extremely high resolution & exceptional correction, etc. It's a symmetrical design. Other models in this line are rated for .25x to 4x but the only info I can find is this is just for 1x only. Is that an omission or something would go wonky on it at a different mag? The resolution ought to support 3x or more for a 12MP FX sensor. Or will it be compromised at a different working distance? In that case it might be better on micro 4/3. It throws a somewhat larger image circle than 35mm, almost double.

Also I guess it's going to be awkward & heavy with a long extension to achieve those magnifications, it comes with a longish tube & (without that?) it's spec'ed for almost 200mm flange mount distance and almost 400mm working distance (object to sensor I guess?)

[PaulFurman]

If my theory is right about limited magnification range, it should work well with a teleconverter on the camera body although that probably spoils all the spiffy color corrections & stuff somewhat :-)

[ChrisR]

I have an apo-rodagon 75mm f4 optimized for 1:1.
I haven't done /can't do the subtlest of testing, but it seems to be much the same as other "good" lenses at 1:1 or 2, like a Rodagon 50 apo, the ubiquitous El Nikkor 50mm f2.8 or the 63mm.
On a short -stacked pic I'd be struggling to see any significant differences, though my sub-optimal technique may be masking somewhat.
I imagine it might be particularly good at flatness of field and lack of distortion, but they aren't important to me.
This is using a 12MP FX sensor, which looks like a limiting factor. Maybe more pixels would show more difference.
Some pics here

[Harold Gough]

The latest lens I bought is the 80mm Olympus OM bellows lens which was made specially for 1:1 or thereabouts. This is said not to focus at infinity but it does if you reverse it. The protruding back of the lens prevents attachment to the camera body but it is a reversed design anyway, like the other bellows lenses in the series.

Harold

[PaulFurman]

My suspicion is it's just optimized for that distance so maybe the super-performance falls apart at higher mags due to the different angle of light??? I found spec from the manufacturer & the others give a wider range of mag in the same chart, same page. In that case it would be fantastic on medium format at 1x but that's not what I've got.

[enricosavazzi]

...
Is that an omission or something would go wonky on it at a different mag? The resolution ought to support 3x or more for a 12MP FX sensor. Or will it be compromised at a different working distance? In that case it might be better on micro 4/3. It throws a somewhat larger image circle than 35mm, almost double.

It depends on the specific lens model, but you should be safe at least between approximately 0.7x and 1.5x. I tested a Zeiss S-Planar 74mm f/4, which is also optimized for 1x, and found it excellent between 0.5x and 2x. Outside this range it starts to show slight aberrations, but can be used satisfactorily up to 5x.

[dmillard]

Hello Paul -

I also have two lenses optimized for 1X, a 74mm S-Planar and a 75mm Apo-Rodagon. You can see here the ratios recommended for the Rodagon, but like Chris and Enrico, I have found both lenses perform very well on either side of this magnification range.

On a separate note, I previously owned a 75mm Apo-Rodagon 2X, designed for duplicating 6X7 format film. It may have just been sample variation, but its performance was marginally inferior to both of the other lenses even at its designated 0.5X image-taking magnification.

Regards,
David

[PaulFurman]

Thanks for the comments folks. The lens in question is a Printing Nikkor 105mm f/2.8 and I'm not going to get it, I was just looking at it & drooling a little but it's too big & awkward and no way to take advantage of the resolution.

Here's my research on it:

----------------------------------
Current version for machine vision:
http://www.daitron.com/tochigi_nikon/pd ... X-data.pdf
Developed specifically for high resolution CCD (line sensor) cameras, all Rayfact Machine Vision Lenses are distortion free (0.0%) at standard magnification. High resolution and the uniformity of efficiency are guaranteed from the center to the periphery. No vignetting (0%). Multi-layers of anti-glare film are applied to prevent the lens surface reflection...

These lenses are designed after the Printing-Nikkor Series and are ideal for inspection using high resolution line scan cameras such as detailed pattern inspection of the FPD substrates and the PCBs. For chromatic aberration, the apochromatic color collection is used in the wide range from 400nm to 800nm.

Code: Select all

Focal Length                    127.0mm          95.0mm            104.5mm
Standard Magnification    - 3.5X            -2.0X -0.5X                -1.0X
Image Size                      64mm ø      32mmø 64mmø         60mmø
Overall Working Distance  651.3mm          397.6mm           388.1mm
Working Distance             70.8mm       87.0mm 212.4mm    140.1mm

----------------------------------
Mounted on an SLR:
http://akiroom.com/redbook/kenkyukai05b/200508-44.jpg

----------------------------------
Chosen by a researcher for photographing crystals:
http://akiroom.com/redbook-e/collection2/tnk01.html

----------------------------------
Printing / Copying macro lens. Focal length 103.9mm Max. aperture f/2.8 Min. aperture f/11 Lens construction 14 elements, 6 groups Usable magnification range 1/1X Corrected chromatic aberration range 400 - 800 nm Vignettings 0 % Distortion 0.0000000000% (1/1X) Aerical resolving power 240 lines/mm (center, 1/1X) Aerical resolving power 180 lines/mm (60mm circle, 1/1X) Image size 60 mm dia. Original size 60 mm dia. Image distance at standard magnification 388mm Weight 375g (courtesy Michio Akiyama)

----------------------------------
http://www.macrolenses.de/ml_detail.php?ObjektiveNr=349
Japanese spec sheet for a slightly newer version A:
http://www.daitron.co.jp/item/data/pdf/ ... Nikkor.pdf
resolution: 240 lp/mm (where DSLRs can only manage about 80 lp/mm

----------------------------------
Discussion:
http://forum.getdpi.com/forum/showthread.php?p=156670
Old 23rd August 2009, 03:22 dcouzin New Member Join Date: Aug 2009 Posts: 12
Re: A superfast normal on the G1
Revision of the moderator-deleted post.

Quote:
Originally Posted by Vivek View Post
The question you need to ask yourself is, is photography an exact science? To me, the answer is no. Mind you, I have and use(for special purposes) some the best lenses (perfect lenses to be precise) ever made. I can use a color corrected Printing Nikkor 95/2.8 for distortionless (0.0000% distortion) copy purposes, for example. That lens was incidentally made for copying movie films.
Funny that you mention a Printing-Nikkor lens. In 1978 got a small NEA grant for motion picture optical printing research and immediately blew $2600 of it on a Printing-Nikkor 105/2.8. That's the 1:1 lens in the series, and being fully symmetrical it is distortionless at 1:1. (Every fully symmetrical lens, even a single element, is distortionless at 1:1.) The Printing-Nikkor 95/2.8 you mention was designed for 2:1 and is not distortionless at 2:1. According to the Nikon literature it has 0.2% distortion at 2×. Every photographic lens is of course "color-corrected". Nikon described the Printing-Nikkor's as "apochromatic" in its first brochure.
I believe photography is an art which rests on much exact science. I've done serious work to help students and artists find simple lenses for their purposes. For example in Good cheap lens for 16mm optical printer (1986) I suggest a $100 lens to substitute for the (by then) $4600 Printing-Nikkor. This was based on sophisticated measures while being down to earth.
That old 1986 problem resembles this strand's problem. An inexpensive lens is desired for a certain imaging task where none is made. Seemingly inappropriate lenses are tried out with some pretty good results. Mounting the lenses is a little pain. What's missing from this strand is the application of any lens sharpness evaluation methods beyond looking at JPEGs with less than a million pixels.
A big difference between the 1986 problem and this strand's problem is that the former involved optical printing and the later involves picture taking. In optical printing the big analog woe was loss of sharpness from generation to generation. The sharpness demands on the optical printer lens were therefore extreme.

----------------------------------
http://photo.net/large-format-photography-forum/0036S1
Dec 06, 2001 B&H for a mere $6,478.50.

----------------------------------
http://www.patentstorm.us/patents/63043 ... ption.html
"Another problem with the optical printers used in the film industry today is that they use lenses, such as the Printing Nikkor lens, which contain multiple refractive elements. These lenses, though of relatively good quality, are limited in terms of the veiling glare, due to a large number of glass components."

----------------------------------
They have recently sold on ebay used for $700 to about $1500, there's one listed for $1700 with a scratched up body, supposedly perfect condition otherwise but removed with a wrench from a copy machine by a salvage guy. Anyways I'm not getting it, just a little too curious :-)

[[Who?]]

The one you mention that is listed has actually been reduced in price over the past few months it's been listed. It used to be on for $2400 odd. I actually brought one last week from a seller in Italy so I'll have to say how I get on. The 45mm thread is a bit of a pain in the arse! Interestingly my serial # is 323553 the one on eBay is 500351 !!

Just checked the original printing nikkor brochure the 105mm pictured is serial #323023 or 323033 ( I can't quite tell exactly)