Minolta 50/1,4. MC vs. MDI

[papasito]

About the resolution power of both lenses, in https://pbase.com/kkawakami, you can see:


#1


#2


Commentary of Koji Kawakakami, author of the page and member of this site, is :

"This lens achieved 250 lppm at centre @f5.6 first time historically in japanese photographic
lenses, it was not surprisingly almost comparable to Leica's Summicron-R 50/2 first version.
These graphs are for the first version of MD 50/1.4 (5 group / 7 element).
See > http://www.pbase.com/kkawakami/rokkor_md_50mm_f14_55mm"

As the central resolution of the MD I, based on the those graphics information, is really great, I bought one of this gems-

Do you know how compare the MD III with the MD I central resolution power?

Any confirmation or refutation of the 250 lppm?

[Teemō]

Actually, I own the source of these tests too, as well as the Leica and Pentax editions and some limited Contax. They are from the Asahi Camera journal's 'New Face Clinic Diagnostic Room' (yes, that is their official English translation, although I believe 'Face' フェース may also be intended as 'surface (usage face) tool'.
If there's a better meaning, I'm curious.

The tests were performed at the National Institute of Advanced Industrial Science and Technology, National Research Laboratory of Metrology (計量研究所). My understanding of the text is that the test was done on 'photosensitive material' (this could mean either semiconductor or film) but probably it was measured directly through the lens. It is clear that the test can provide any resolution, so it is measured, not estimated. The target might have been traditional slanted line pairs as this can provide for the tangential and sagittal resolution where (as far as I know) a Siemens Star cannot. Across other tests, the specified test magnification is mentioned as 1/63 or in cases where the lens is long such as a 135mm (test space was limited), the magnification was given. The maximum resolution for this machine and test condition is 280 lines per mm. For telephoto lenses (above 135mm) the maximum was 180.
Lenses are purchased at random from a retail store rather than requested from the manufacturer. The review team normally consists of a photographer, an engineer from the national industrial institute, an optics professor from a university, and also an official from the Japan Photographic Instruments Inspection Association (JCII).

Keep in mind that the figures given are given in lines per mm, not line pairs per mm - unless it is mentioned elsewhere, but these are reprints of historical tests. There are no contrast figures provided so we really have no frame of reference for just how well these lenses perform at a given resolution, but we can at least assume that the contrast level was consistent (it's probably safe to assume 10-20% contrast at these resolutions, as a typically diffraction-limited point and compared against the modest resolution at 30-40% of MTF graphs in that period).
The only thing I really question with these tests it that 240L/mm and 112L/mm are common figures. It's too coincidental and I suspect it is a condition of their testing method, and also deliberately rounded. Anyway, I don't think the coincidence is malicious.
I suppose the test of the MD-III has not been conducted or else it would have been included. Probably the MD-I was already considered an exemplary result for the series.

---

The Leica lens tested in my edition was the Summilux-R II, from 1998 - so far more modern than the Minolta lenses but still without asphericals. Erwin Puts wrote about this lens compared to the first version: "The new 1,4/50 Summilux-R defines the current state of the art of large aperture standard lenses. It outclasses the previous version of the 1.4/50 Summilux-R by a clear distance. It edges ahead of the current Summicon-R and improves upon the current Summilux-M 1.4/50.

In comparison to the Sumicron-R 2/50 the new Summilux is an improved design. Optical aberrations ( especially curvature of field and astigmatism) are tightly controlled, and at f/4 to f5.6 the Summilux-New edges ahead of the Summicron-R."

The optical test results in Asahi generally agree that the astigmatism is well-handled but indicate that field curvature is definitely not. You can read another independent test here to put Put's words into context: https://photo.imx.nl/leica/lenses/lenses/page57.html
"If we set the previous Summilux-M at the beginning of a line and the new Summilux-M ASPH at the end of the line, the current Summilux-R would by on a point, located at about two thirds of the line segment from the start point."

---

Resolution

The Leica lens is capable of high-resolution in the mid-field and corners when the focus is shifted, only 10-25% less than the centre resolution when focused for the highest resolution in each area of the frame. For brevity here, I only wrote the nominal focus result (regular flange focal distance of the R-mount) to match the Minolta reports and only the number that is highest of Rr and Rt. Radial distances (0, 3.9, 7.8, 11.7, 15.6, 19.5) of the image frame are the same.

At F1.4, it achieves: 160, 130, 55, 38, 24, 39 | Average: 56
At F5.6 it achieves: 250, 238, 139, 67, 40, 129 | Average: 113

Field Curvature

As written on your images, for the MD50/1.4, the best across-the-frame focus is 0.03mm (30um) closer to the lens than the best centre focus.
At F5.6, the best focus coincides across the frame.

For the MC50/1.4, the best across-the-frame focus is 0.06mm closer to the lens.
At F5.6, it grows to 0.08mm, while the best centre focus shifts 0.03mm closer to the lens.

The Summilux is worse, with the best across-the-frame focus being 0.09mm towards the lens' rear.
At F5.6, it grows to 0.16mm.

------

I was slightly amused, reading more on the issue, to find that Roger Cicala at Lens Rentals didn't understand Field Curvature issues against MTF himself. Had he been an avid Asahi camera reader (or consulted a university), he would have followed the same method and saved himself 100 hours of confusion.

The deviation of even 0.03mm is quite significant, considering that Kodak gives the emulsion thickness for several of their aerographic films as between 7.6 and 21.3 microns thick. It is easy to see why macro lenses must remain in a category of their own.

It is mentioned of the Summilux that spherical aberration is extremely well corrected and that softness does not come from flares or glare. Actually, they do provide a MTF graph for the Summilux and it would show the performance has increased over time against Leica's current MTF graphs (manufacturer graphs are usually theoretical/computer-generated or of a 'master production sample'). They provide a 2nd set of MTF at 50cm focus distance: wide open the MTF is V-shaped for the 10cy, and flat at 40cy. At F5.6, the overall contrast is excellent but the details at 40cy are A-shaped across the image circle. The MTF of the MD 50/1.4 are about 15-20% worse, as to be expected - contrast is simply lower.
With regards to the high inward field curvature, the reviewers suggest that it is better to have a curved image plane to take a three-dimensional subject, and that this is clearly a lens designed for general photography such as portraits and landscapes. I know my Minolta 50mm and 58mm lenses suffer terrible field curvature at macro distances, compared to my 35mm lens.

The original test dates of the lenses:
MD 50/1.4: 1978 (MD-I)
MC 50/1.4: 1973
Summilux 50/1.4 R-II: 1991

You can see what maximum resolutions modern lenses can theoretically be pushed to in a lab setting here: https://www.lensrentals.com/blog/2017/07/experiments-for-ultra-high-resolution-camera-sensors/ and https://www.lensrentals.com/blog/2019/10/more-ultra-high-resolution-mtf-experiments/

[Alun Thomas]

Just a quick tangential question: How many version of the Minolta manual focus 50/1,4 are there? I had assumed any 5/7 MD-I 50 1/4 must share the same optical construction as the MC-PG version. is this the case or is the MD-I 5/7 version different again?

[papasito]

[Ernst Dinkla]

[Teemō]

[Ernst Dinkla]

I think the MC was better on overall IQ than the first MD versions, not in all optical aspects though. The last MD versions improved a lot but I have no tests for that. How they all behave on a digital sensor, FF or APS, counts more now.
I have not enough compared my early Olympus OM 50mm 1.4 with the MD I that I have, but the corner quality of the OM at 4.0 and 5.6 is very good, I doubt the MD I equals it there. That said the T-stop of the OM is 1.65 where the MC/MD lenses score 1.52/1.53. Something you see in more OM lenses, smaller lens elements and by that more vignetting sacrifices actual speed but gaining some corner quality that way.

[cbass]

I want to confirm I understand this correctly.

MD 50mm f/1.4
This is a 7 element 6 group lens?

MC 50mm f/1.4
This is a 7 element 5 group lens?


@f/5.6 the MD 50mm f/1.4 did 250 lppm in the center?

@f/5.6 the MC 50mm f/1.4 did 224 lppm in the center?

What is the second row for the MC 50mm f/1.4 that doesn't exist for the MD? There is one with 224 and another with 100

[Ernst Dinkla]

OP's first message is about an MD I and the MC, both 5/7.

[Teemō]

[cbass]

If both were 7/5 lenses, then that means there have been some improvements although they use the same number of elements or groups. Is that correct or is it just sample variation?

[Teemō]

[Ernst Dinkla]

Firefox displays the images well.

[papasito]

My copy of the MDIII 50/1,4 has better coating, ore contrast and is sharper than my copy of the MDI.

[Antoine]

http://www.rokkorfiles.com/Battle%20of%2050s2.htm

No, although same number of elements/groups, MC and MD 1 are different designs as shown in Rokkor files.

[stevemark]

According to official Minolta documents, the new MD-I [7/5] computation was mainly made to reduce the length of the lens (from >95% of the focal length to <90% of the focal length), while maintaing an "acceptable" performance and - equally important for a SLR lens - the backfocal distance. To reduce costs, the large and thick 2nd lens was made from less expensive glass (nD 1.69, v=49). At the same time the thickness of the 5th lens (made of more expensive glass with nD 1.79, v=46) was reduced.

I have updated the artaphot site with the corresponding MD-I 1.4/50mm lens section:
http://www.artaphot.ch/minolta-sr/objektive/155-minolta-50mm-f14

Stephan

[Alun Thomas]

[papasito]

AFAIK, the 50/1,4 MD II (v. 2) was born with 49 mm filter diameter and the same with the MDIII.

The MDI and MDII (v.1) both have 55 mm filter diameter.

The MDII V.1 is the most rare un my experience

[stevemark]

[papasito]