Review: Minolta MD Rokkor 50 mm f/1.2 (MD-II)

The MD Rokkor 50 mm f/1.2 is the successor of the legendary MC Rokkor-PG 58 mm f/1.2, a.k.a. the “bokeh king”. It uses a different optical design, is smaller and lighter. The change from MC to MD in this case also meant going from 8 to 6 aperture blades. Because of this step back and a generally different image characteristic, MD f/1.2 is often considered to have inferior bokeh compared to the MC f/1.2.

The lens features a solid weight, focus is pleasingly smooth and handling on a NEX-5T is nice. Focusing at f/1.2 is challenging due to the extremely thin depth of field combined with the rather glowy image, which reduces the focus peaking ability of the NEX. Optical qualities wide open and the effective focal length of 76 mm on APS-C cameras make the MD f/1.2 a very flattering, ultra fast portrait lens.

For the MD Rokkor 50 mm f/1.2 I’ve created lens correction profiles, which are available for download. For further details on the lens like weight and dimensions, have a look at its entry in the Minolta SR mount lens database.


Condition of my copy

Optics: Very good. No scratches or cleaning marks and only a handful of dust particles inside. A small dot of fungus behind the second lens element had to be removed and left no marks.

Mechanics: Very good. Uniform and smooth, aperture clean and fast.

Exterior: Excellent. Only slight marks on the mount and a missing red indicator dot for the mount alignment, which was replaced.


Optical performance on NEX-5N / 5T

Update (14.03.2015): Test shots repeated with a little more love put into focusing at f/1.2 and f/2. The lens now performs slightly better wide open. Yet, visible astigmatism in the test shots (horizontal lines are rendered sharper than vertical ones) points to the fact that the lens is not perfectly centered. Corner performance and CAs could possibly improve if it was.

The lens is hazy and rather soft at f/1.2 with mediocre contrast. Contrast and haziness improve strongly when going to f/2, but the softness stays. At f/2.8, the lens makes a large step forward in terms of sharpness and the haze disappears almost completely, improving corner performance from mediocre to good. At f/4, sharpness reaches very good levels with an even performance across the image. Stopping down to f/5.6 or f/8 makes the image that last tiny bit crisper, raising the sharpness score to excellent. Diffraction starts at f/11 and softens up the image at f/16.

A slight red and very dark cyan glow is visible at f/1.2, but mostly blurred away by the haze. At f/2, it starts to stand out a little more. From f/2.8 on, the lens shows light to medium red/cyan CAs which stay constant until f/5.6 and then grow a tiny bit if stopped down further. A good performance, especially considering the extremely fast maximum aperture.

Vignetting is surprisingly low with only 2/3 of a stop at f/1.2 and is nearly gone by f/2. The lens further exhibits a slight barrel distortion of -0.5%. The effective T-stop at f/1.2 is approximately T1.4 (-0.5 EV), which is a little disappointing. This, however, is not primarily the fault of the lens, as DxOMark has proven. It seems to be related to the problem that digital sensors register less light the further the incident angle of the light diverges from 90°. And the higher the f-stop, the higher the fraction of light that reaches the sensor at smaller angles. If you read the DxOMark investigation, keep in mind that the pixel pitch of the NEX-5T and 5N is 4.76 µm, close to the α7R’s 4.86 µm. Looking at their graph, even the modern f/1.2 lens they used for the test didn’t perform better than the Minolta MD 50 mm f/1.2 in terms of light transmission.

Compared to the MC Rokkor-PG 50 mm f/1.4, the performance of the MD f/1.2 is very close. A little less haze wide open as well as higher micro-contrast up to f/2.8 favor the MC. The latter also shows less CAs. But resolution-wise, the lenses are close to being equal at all f-stops and from f/4 on, perceived sharpness is identical. Compared to the legendary MC Rokkor-PG 58 mm f/1.2, the lens shows nearly equal sharpness with a very slight edge for the MD 50 mm up to f/2.8. However, the 58 mm f/1.2 shows significantly less lateral chromatic aberration (CA) and is mechanically superior – not by much, but you can feel it. One of the main downsides of the 50 mm f/1.2 – heavy longitudinal CA, sometimes termed bokeh fringing or color bokeh – is also almost unbeknown to the 58 mm. On the plus side, the 50 mm is much more compact and roughly 160 g lighter, which is “about half a lens” in terms of normal lenses. Overall, it’s probably more a question of taste whether you prefer your Minolta f/1.2 to be a 50 mm or a 58 mm. Both lenses show the typical characteristics of vintage ultra wides and can give you a razor thin depth of field when needed.


Test charts

The following images are pixel-level crops from the test chart. They may appear scaled in your browser window. Click on them to view the crops in full size and cycle through them easily. For more info on the test setup, visit the details page.



Test chart overview

(Cropped areas marked in orange)



MD Rokkor 50 mm f/1.2 @ f/1.2



MD Rokkor 50 mm f/1.2 @ f/2.0



MD Rokkor 50 mm f/1.2 @ f/2.8



MD Rokkor 50 mm f/1.2 @ f/4.0



MD Rokkor 50 mm f/1.2 @ f/5.6



MD Rokkor 50 mm f/1.2 @ f/8.0



MD Rokkor 50 mm f/1.2 @ f/11



MD Rokkor 50 mm f/1.2 @ f/16


Leave a comment

Your email address will not be published. Required fields are marked *

2 thoughts on “Review: Minolta MD Rokkor 50 mm f/1.2 (MD-II)

  • Blz

    What you are seeing in the corners wide open is not a hazy image or haze, but vignetting. This is caused by the image circle of the lens not being uniformly illuminated out to the edges of the image fornat when the lens is wide open.

    • Benjamin Post author

      I think we are talking about two different things here: Vignetting is the darkening of the image in the corners. It’s most pronounced wide open and disapperas when stopping down. You are correct in the sense that this is caused by non-uniform illumination.

      The “haze” I’m referreing to shows as the bleeding of lighter into darker areas of the image, best seen on the resolution digits in the corner crop: The digits are much less black than, for example, the center of the slanted square. I think that this haze is caused by spherical aberrations, which become less pronounced when stopping down. There might also be some longitudinal CA involved, which is responsible for the very slight dark red to purple glow on contrasty egdes, best seen in the center crop (look at 200%). The LoCas further lower the micro contrast.