Nikon CF N Plan 4x's comparison -images added

[ChrisR]

Your resolution figure I believe allows only 2 pixels per line, depending on MTF and all that, and a perfect lens.
I have some objectives which
theoretically outresolve my sensor, and others which
theoretically outresolve my sensor by even more

The fact is, I can see a difference.
( eg 10x 0.3 vs 10x 0.46, 8μ pixels)

In the old days the formula
1/(Achieved resolution)= 1/(lens res) + 1/(Film res)
was bandied widely.
Taking the reciprocals, everything adds its own blur. It makes sense to me.
Digital sensors obviously have a sharpish cutoff which limits resolution, but with the filters built in, there is a slope to the MTF.


The reason I tested the 4x 0.2 at about 7x was to compare with an Olympus camera lens, which I couldn't get nearer to the camera, at the time.
People with larger sensors are interested in using objectives outside the envelope they were designed for. One person's 5mm field width may mean 4x, when for another it's 7x.

[Blame]

You have caught me using rules of thumb. Fair enough.

But what was the lowest magnification (or shortest total extension) that by your subjective standards resulted in sharp corners? I know that your answer can't be definitive but when definitive answers are unavailable, a subjective evaluation is far better than nothing.

I supose what I am searching for is a new rule of thumb. A ratio between a reputable manufacturers estimate for field of view of a top quality objective and the "best" in practice. It is going to be a wooly figure, and dependent on the relative quality of objectives and cameras. As I tried to point out, the higher the definition of the camera the less the disadvantages of pushing the field of view to the limits.

Given that the lowest magnification of camera occular used to be 2.5X back in the old film camera days, one could gess that the optimal limit for a diagonal was at best 45/2.5= 18mm. This at a time when field of view was quoted at around 25mm to 28mm. My guess is that the better digital cameras of today shift the optimum outwards. Certainly some enthusiasts are going all the way out to 27mm.

I ask because I am starting my own project and it doesn't include a bellows. If I have misjudged the usable field width it is going to cost me time and money. The quicker I correct a mistake the less lost.

[Cyclops]

How would this lens compare to a Meiji 4X/10 plan objective?

[ChrisR]

Blame - you'll find there's no very simple rule of thumb, it varies enormously between lenses.
From memory -
a NIkon Mplan 20x is OK for full frame (24 x 36)
the 4x 0.2 (http://www.photomacrography.net/forum/v ... 6012#56012) covered 36mm diameter reasonably at 7x but the corners of the frame were bad - the 0.13 was a little better.
and Nikon 2x ( can't remember which, one's an APO) produce a circular image only, which I don't think quite covers a DX format.
If you're desperate for 24x36 to be covered, stick a teleconverter on the camera. Sorted!

[Blame]

Thanks Chris.

45X4/7=25.6

That works out at about a 26mm diagonal but at the expense of poor corners. - Probably a fair compromise.

But I take your point - It is going to be a very rough rule of thumb indeed.


Does anybody have experience with mitutoyo objectives?

[Charles Krebs]

Microscope objectives were typically designed to "cover" one of two image circles... both are physically limited by the size of the eyepiece tubes. Microscope eyepieces have either a 23mm or 30mm diameter. So a typical "normal" objective was made so that the image circle diameter "seen" by the eyepiece (in a 23.2mm tube) was generally 18-20mm (at best... sometimes smaller. This is also why a 2.5X was the de facto standard projection eyepiece for 35mm format... it recorded an 18mm (17.3mm to be exact) field number which provided a good match for the most commonly used 10X viewing eyepieces.) An objective that was made to produce an image suitable for the 30mm eyepiece tubes (often called "super-widefield") was made expecting the observed image circle to be about 26.5mm in diameter.

A 4/3 sensor has a diagonal of 21.6mm
An "APS" sized sensor has a diagonal of about 27-28mm.
A 24x36mm sensor has a diagonal of 43.3mm

So, on a bellows, if you have an objective that was made to be OK with 30mm eyepieces you should be totally safe with 4/3, most likely good with APS-sized, and questionable with 35mm format. Beyond this who knows! You just need to try it and see.
(We're talking finite objectives used directly on bellows at the designed "tube length", and thus magnification. Remember, the DIN standard called for the intermediate image to be located 10mm down from the top of the eyepiece tube... so the "optical" tube length would be 10mm less than the "mechanical". But not all were DIN ).

Naturally some objectives will be pretty bad even within their "range", and it's not unreasonable to expect others to perform well even beyond these "guideline" image sizes. But we're definitely talking "try it and see".

Extending more will increase magnification and may improve corner performance with "borderline" sensor sizes. At some point increasing extension alters the design "geometry" to the point that other problems creep in (most noticeably spherical aberration and excessive diffraction effects. Although increase in SA is usually not a big issue with low NA objectives.).

Mitutoyo literature "states" intermediate images of 30mm diameter for their objectives. I've used one on an APS-sized sensor with no problem at all in the corners. Never tried it on 24x36.

[Blame]

Charles

You have been very helpful.

I see I got the diagonal for full frame wrong. Thanks for pointing it out.

That makes the "poor corners" field of view for the Nikon "4X 0.2" 43.5X4/7= 25mm. Sour grapes on my part, but that at least gives hope that another objective can be found that gets close with careful choice of magnification. There are not enough of this particularly good Nikon for all of us.

Field of view is clearly a difficult issue to predict which is why is is so helpful that you share your experience. It absolutely stellar that you have already tried a mitutoyo. Which one was it, and what did you use as a tube lens?

[ChrisR]

What coverage are you looking for, at what magnification?

[Blame]

Well I have (I think) standard Mitutoyo 5X and 10X objectives. My uncertainty is that they are labeled Mitutoyo QV-2.5X and QV-10X, but physical dimensions and specifications are the same as best as I can tell.

I am looking to cover a canon cropped sensor - that would be 27mm diagonal, 22.5mm diameter.

I am planing to use a 150mm lens as tube lens (although I do have a tamron 180mm nikon fit macro - a bit more bother, and I was hoping to sell it, but if needs must)

I am happy to correct vignetting, some chromatic aberration, distortion, and curvature of field, but serious loss of definition is uncorrectable and to be tolerated only in the extreme corners.

[Charles Krebs]

Which one was it, and what did you use as a tube lens?

It was a Mitutoyo 5/0.14 M Plan Apo and I used an Apo Nikkor 210/9 as the tube lens.

(Talked a bit about it in another thread: http://www.photomacrography.net/forum/v ... php?t=8995 )