Lens Resolution (was 'Nikon Multiphot brings $3,550')

Hokan · Post by **Hokan** » Fri Oct 23, 2009 4:28 pm

Barry wrote:Interesting data.

Recently I tested a 12cm Macro-Nikkor, 63mm Luminar, 50mm and 63mm El-nikkors (N-series) and a Tominon 105mm.
Camera was Canon 450D (now replaced). Bellows draw available up to 600mm. Illumination was fibre optics through transparent paper.
Did not print the images - my opinion is based on viewing at 100% at computer screen.

I found that, at all magnifiations, the Macro-Nikkor and Luminar were superior to the EL-nikkors and Tominon. No doubt. As to be expected, from litterature.
Sharpness is at least slightly better, colour and contrast certainly better.
The apertures I found best were at both macro lenses ap=1.5 (1 is a little bit soft), the EL-nikkors 5.6. For up to 2x I now use the Macro-Nikkor, for 2x-4x the Luminar.

BUT, another view may be:
The EL-nikkors and Tominon were bought for €25,-- (US $35), which is a big contrast compared with the macro lenses.
And in which cases are the difference in quality (post-processing included) really noticeable?

Cheers,
Barry

Hello Barry,

Not disagreeing with your findings or conclusions. I would just be curious as to what specimen(s) you used across all lenses and what type of illumination you used. I've found the El-Nikkors to be less then their best when used with specimens requiring more than 0.1mm DOF. Again using 35mm sized sensor as standard. However within their 0.1mm or less DOF they come into their own sweet spot. Stacking would quickly revise my last statement.

regards,
Hokan

Barry · Post by **Barry** » Sat Oct 24, 2009 9:18 am

Hi Hokan,

My tests were with stacked images, of either an insect (moth) or a macrofossil. Images were shot at 0.05mm interval, so within the 0.1mm you mention.
But my camera was a Canon 450D, thus no 35mm sized sensor, which is a big difference. I can imagine each type of sensor reacts different.
For illumination, I use a fiber optic illuminator with double goose neck, 100watt, diffused through a cylinder of semi-translucent drawing paper.
Often I try to find a compromise between diffused and direct light, to create some extra detail in the image.

If I may insert some images in this thread as a comparison:

EL-nikkor N 63mm at f5.6 and Luminar 63mm at ap=2 (corresponds to f6.3).
Below are 600 pixel crops of a moth, taken at 2x, no post processing. Fiber optic illumination was kept constant between shots. (0.05mm focus interval).
Note that probably 2x is more ideal for the Luminar than the EL-nikkor. But the difference is there at any magnification.

(note: pictures were taken with Auto White Balance, colour may not be good. Now I use a grey card and am more satisfied).

EL-nikkor:

Luminar:

Best wishes,
Barry

Eric F · Post by **Eric F** » Sat Oct 24, 2009 10:01 am

Hi Hokan,

Thanks for your remarks and observations on DOF. Your math and logic seem perfect (though I admit to not following the math part closely, as I've always tried to avoid that subject...). My needs for macrophotography are now mainly driven by the working geometry of my photostand setup, and the working distance between my subject (usually a dead fly specimen on a pin) and the lens being used, while simultaneously considering how to best deliver suitable light. After lots of experimentation, I keep coming back to using a Twin Lite flash (designed for use with the Canon MP-E macro lens -- but very good for many other lenses also). So, my current photostand geometry (for >5x magnification) fits best with a working distance of about 10 to 40mm, and the use of a small-diameter lens (less than 50mm). Lens diameter is important to the proper placement of the Twin Lite along the extension tube alignment -- as the metal ring holding the Twin Lite is just over 50mm in diameter. For example, I have great difficulty using my Zuiko 20mm/f2 (= 60mm wide) with the Twin Lite. (I'm now experimenting with placing the two flash heads of the Twin Lite on independent flex arms -- just to use with this lens.)

Anyway, good lenses of around 20mm focal length fit my working geometry and magnification needs.

Cheers,

Eric

Charles Krebs · Post by **Charles Krebs** » Sat Oct 24, 2009 11:26 am

Barry,

Don't have a Luminar, but I do have a 63mm El Nikkor. (And several 50/2.8's as well).

So I can't speak to the Luminar, but something is definitely not "right" with the El-Nikkor shot you posted above. You don't indicate how large a crop this is, but even if were seeing it at 100% the results should be far (far!) better.

Hokan · Post by **Hokan** » Sat Oct 24, 2009 8:13 pm

Eric F wrote:Hi Hokan,

Thanks for your remarks and observations on DOF. Your math and logic seem perfect (though I admit to not following the math part closely, as I've always tried to avoid that subject...). My needs for macrophotography are now mainly driven by the working geometry of my photostand setup, and the working distance between my subject (usually a dead fly specimen on a pin) and the lens being used, while simultaneously considering how to best deliver suitable light. After lots of experimentation, I keep coming back to using a Twin Lite flash (designed for use with the Canon MP-E macro lens -- but very good for many other lenses also). So, my current photostand geometry (for >5x magnification) fits best with a working distance of about 10 to 40mm, and the use of a small-diameter lens (less than 50mm). Lens diameter is important to the proper placement of the Twin Lite along the extension tube alignment -- as the metal ring holding the Twin Lite is just over 50mm in diameter. For example, I have great difficulty using my Zuiko 20mm/f2 (= 60mm wide) with the Twin Lite. (I'm now experimenting with placing the two flash heads of the Twin Lite on independent flex arms -- just to use with this lens.)

Anyway, good lenses of around 20mm focal length fit my working geometry and magnification needs.

Cheers,

Eric

Hello Eric,

Sounds like a very workable set up.

Just for your information the Nikkor Macro 19mm has a working magnification range of 15x-40x, sweet spot around 20x. Working distance from subject to front of lens of 14.3mm(15x) to 13.4mm(40x) using a 35mm sized sensor. About 155mm to 700mm bellows extension.

The Zeiss Luminar 16mm has a working magnification range of about 4x to 12x, sweet spot around 10x. Working distance from subject to front of lens 12mm(4x) to 9mm(12x) using a 35mm sized sensor. About 0mm to 90mm bellows extension.

A dramatic difference in design and lens capability in my estimation.
One of the reasons I hesitate to embrace tests showing El Nikkor, Nikkor Macro, Zeiss Luminar, etc. being better or worse than other compared lenses.

Intentionally did not address aperture, type of specimen, DOF, illumination technique. To include all of those would mean I have finished my winter project! Not

Just a note of interest. Oblique illumination appears to produce the "best" contrast and definition for opaque specimens as the magnification increases, 5x and up.

regards,
Hokan

ChrisR · Post by **ChrisR** » Sun Oct 25, 2009 1:21 am

63mm El Nikkor, something not "right" - indeed.
That reminds me of results I got before the reversing adapter arrived :-& !

Eric F · Post by **Eric F** » Sun Oct 25, 2009 9:13 am

Hokan,

Thanks for the info on the Macro Nikkor 19 and Zeiss Luminar 16. The Nikkor sounds like a fantastic lens, and the specs for the Luminar 16 are very good for my needs (though I could use some extra working distance...). Have you used the Luminar 25 much? Do you have specs for that one? Seems like there are quite a few Luminar 25's available on eBay ('minty' & otherwise!).

Cheers,

Eric

Charles Krebs · Post by **Charles Krebs** » Sun Oct 25, 2009 3:12 pm

Don't forget to keep in mind that most of us are now using cameras with sensors in the size range from about 15x23mm to 24x36mm in size. To be of any real use, the images recorded with these cameras need to be magnified... either on screen or in a print.

Many of these lenses were designed to be at their best at certain magnifications (but are certainly very good over a range of magnifications), and were originally used with equipment that used either Polaroid pack/sheet film or 4x5" (or even larger) sheet film.

You'll find that at their "optimized" magnifications they were superb for use on these larger formats, which needed very little (if any) additional magnification to be viewed. However, when used with our present, comparatively tiny formats, the final result might not look that great. This is because of the need for significant additional magnification of the recorded image. The culprit is resolution loss due to diffraction. The problem is quite literally "magnified" due to the need to increase the size of the recorded image.

Consider the image recording size:

DSLR ... 15x23mm to 24x36mm
4x5" (image size, Fidelity holder)... 95x120mm
(sheet Polaroid is about the same as 4x5", actually a tad bigger image)

Remember that the "effective" aperture seen by the camera is:
effective f-number = f-numbermarked on lens* (m+1)
(for a lens with a pupillary magnification factor of 1)

So an f2.8 lens used at 20X has an "effective" aperture of about "f58".

Then play with the "calculator" toward the bottom of this page:
http://www.cambridgeincolour.com/tutori ... graphy.htm

When an image will be presented at a small size there is considerable leeway.... but once you get up to about 10" width you really need to be very "aperture sensitive" with small formats.

However, keep in mind... with a given subject you need to photograph at a lower magnification to "fill the frame" with a smaller sensor/film. If you have a subject sized 6mm (left to right) and you wanted to "fill the frame" (left to right) with that subject you would need to photograph at a magnification of 20X on 4x5" sheet film, but only 4X-6X for a DSLR. This will have a large effect on the "effective aperture" obtained.

Charles Krebs · Post by **Charles Krebs** » Sun Oct 25, 2009 4:06 pm

I had meant to include another thought in my previous post...

Sometimes it seems to make more sense to work "backwards". Instead of starting with a specific lens, start with the sensor/film size to be used and the desired subject sizes. Have some idea of the size of the final usage/reproduction. Are you "stacking" or will it be a single frame image?

With these things in mind you can then start to search out optics that will meet your needs.

For example, say you are using a "reduced frame" DSLR and know that you need a 10X magnification on sensor. You will be "stacking" images. You would like to have an image that will stand up to being printed to a 10" width.

Based on these needs, you know that the "marked" aperture set on the lens should be multiplied by 11 ("m+1") to obtain the effective aperture. Ideally it would be nice to keep your effective aperture somewhere between f13 and f16. At 10X this is very tough to do... so you loosen your criteria to a limit of about f22. This means that you should be looking at lenses that can be used (ie: perform well) at a "marked" aperture of about f2. This really limits the choices! If you need to use a lens with a smaller aperture the results will have lower resolution. (However... if the final image was not stacked, but a "one shot", then this could well be a compromise you need to make in order to obtain any semblance of depth-of-field!)

This is one reason many of us here are enamored with certain 10/0.30 and 10/0.25 microscope objectives on bellows. At 10X these provide "effective" apertures of about f16 to f20. DOF is so shallow that stacking is nearly always required, and it can be really rough to work with some subjects... but when you "nail" one of these images the results are stunning.

Hokan · Post by **Hokan** » Sun Oct 25, 2009 7:19 pm

Eric F wrote:Hokan,

Thanks for the info on the Macro Nikkor 19 and Zeiss Luminar 16. The Nikkor sounds like a fantastic lens, and the specs for the Luminar 16 are very good for my needs (though I could use some extra working distance...). Have you used the Luminar 25 much? Do you have specs for that one? Seems like there are quite a few Luminar 25's available on eBay ('minty' & otherwise!).

Cheers,

Eric

Hello Eric,

Yes I've used my Zeiss Luminar 25. A very sharp lens when used to its strengths. I like the bokeh it produces.

For the Zeiss Luminar 25, (Zeiss defines its focal length as 26mm), Magnification 3.8X to 7.8X, Sweet spot around 6X. Working distance from specimen to front of lens is around 19.5mm(3.8X) to 16.2mm(7.8X). Bellows extension is about 30mm(3.8X) to 102mm(7.8X). Using a 35mm sized sensor.
Charles brings up a very important point to keep in mind. The rule of thumb I use. My D300 mounted on my Axiomat I use a 1.6X projection lens, for my D700 I use a 2.5X projection lens. The Axiomat has its own camera module for 35mm and 4x5 large format. The Multiphot was origninally configured for 35mm to 4x5 LF and everything in between. ie. MF, etc.. On those rare occasions when I use my Axiomat's 4x5 LF camera I know that my Axiomat will magnify the image from my objective by 10X. Thus giving me parfocal with my 10X eyepieces. Now if I only had the money for a 4x5 Digital scanning back.

Realistically about 4 years away.

regards,
Hokan

Eric F · Post by **Eric F** » Mon Oct 26, 2009 10:45 am

Charles, thanks very much for your most informative thoughts on the use of these microlenses and objectives. Unfortunately, I haven't kept certain of these concepts in mind (and wasn't even aware of some of them...) -- as I have tried to come up to speed in photomacrography. (Another problem is my inexperience in photography in general!) I do use a small sensor (15x22mm for the Canon EOS40D), so these concepts of format size, effective aperture and diffraction are, no doubt, especially important.

One thing is clear: my immediate goals are to produce stacked images suitable for use in illustrating the detail of small insect parts -- for eventual use in journals or on the internet -- to help in identification and systematic study of same. A format showing a printed image of around 6" should be enough for this (although now online publications are including links to even larger images, backing-up the 'normal' page shots!). Excellent examples of the kind of work I hope to eventually produce are the Tabanidae pages that NU has shown us.

I would say it is pretty clear (from your photos, many other photos in these forums, even from my own limited experience) that the very best images in the 5x to 15x area are produced by 4x and 10x objectives. However, using an objective lens requires about 5 to 10 times as many images in the stack, compared to what is needed by a very good microlens. I would like to find microlenses that could approximate the results of an objective lens -- at 1/5 to 1/10 the time & effort. (And, I admit to being very envious of Andrew & others for their ingenious mechanized stages!)

Eric

Eric F · Post by **Eric F** » Mon Oct 26, 2009 10:53 am

Hokan,

Thanks again for your info -- this time on the Zeiss Luminar 25mm. Sounds like the lens is best at just beyond the limits of the MP-E 65mm, so maybe I need something with a little more magnification. The working distance is perfect, however.

Cheers,

Eric

NikonUser · Post by **NikonUser** » Mon Oct 26, 2009 11:11 am

Eric:
In its Instruction Manual, Nikon gives specifications for the El-Nikkor 50mm f/2.8 N enlarging lens:
standard magnification 8X, usable mag. range 2X-20X.

I know it produces excellent image at about 4X when reversed on bellows but can't recall having tried it at higher mags. It would need a lot of extension at higher mags.
f-stop goes from 2.8 - 16

Barry · Post by **Barry** » Mon Oct 26, 2009 11:57 am

So I can't speak to the Luminar, but something is definitely not "right" with the El-Nikkor shot you posted above. You don't indicate how large a crop this is, but even if were seeing it at 100% the results should be far (far!) better.

Charles,

You are right this particular stack was not in favour of the El-nikkor 63mm, it is indeed at 100%. The EL-nikkor 50mm gave better results but I thought the 63mm was good as it had the same focal length as the Luminar.
Traded the lens later, so can't do the test again, sorry.
I can post a crop of the 50mm EL-nikkor vs the Luminar if you'd like.

Cheers,
Barry

Charles Krebs · Post by **Charles Krebs** » Mon Oct 26, 2009 12:41 pm

Barry,

A brief comparison I made between the 63/2.8 and 50/2.8 El Nikkor showed very similar results (at least with my "samples"). Doubt I could ever distinguish between them. In "operation" you do, as expected, get a little more working distance and need a little more extension with the 63mm.

It would be interesting to see a crop from a sharply focused single image from the stack... of a detail something like the serrated edge of a scale like those seen in the lower left quadrant on your posted comparison. Seems like it should be "crisper" (in the Luminar image as well) and a single image could indicate if somethings going on in the stacking process.