50mm lens shootout (Summicron DR, Summar and Elmar Red Scale)

[rogerzilla]

Test details: heavy tripod and cable release, Reala film, flat outdoor lighting, 2 metres shooting distance, 3600dpi scan (3300dpi in the real world) with no post-processing at all. Focused carefully (and I have checked the rangefinder) - all lenses focused to exactly the same scale point, which was reassuring. This is just meant as a comparison of resolution and contrast, not as a means of quantifying resolution, but in some cases the lenses do get down to scanner resolution*. EDIT: the scans here are 100% so represent a piece of the negative less than 2mm wide. They are all from the centre and I haven't attempted to measure edge sharpness.

The target is the good old USAF test chart; just the middle section. The smallest "1" and "2" which you can make out on the f/4 scans are about 1-2mm tall on paper.

At f/2

Summicron DR followed by Summar. The DR wins by a country mile. Focusing is absolutely critical though, and it's easy to get an unsharp picture at f/2 if you're half an inch off the plane of true focus.

At f/4

Summicron DR/Summar/Elmar. The DR and Summar are almost identical, with the DR a fraction crisper (you'd need a drum scanner to really show any difference). The Elmar isn't bad considering we're using almost all the glass.

At f/8

Summicron DR/Summar/Elmar. The DR has the edge, but not by much. At smaller apertures they become diffraction-limited so there's no point in a test at f/11 or f/16.

Other observation: the Summar shot is underexposed at f/2, probably because of the lack of coatings.

So...don't be afraid of your Summar, as long as you don't use it wide open (or in any situation where it might flare) and downrate the film a bit to allow for the light loss. Also, don't be afraid of the DR wide open, but your technique must be spot-on. The Elmar will turn in consistently good pictures at any aperture although it never reaches the performance of the f/2 lenses.


*to be specific, the last point at which the difference between horizontal and vertical bars can be seen is to the right of the third smallest figure "4", by which point we are alternating light and dark pixels in Photoshop. Try it and see with the DR f/4 image. MTF is probably very low by this point.

 

[ferider]

The true difference between these lenses will be out of center. You're right, for accurate center resolution measurements focus is critical and difficult, and you'll find that most cameras will be slightly different when used with the same lens. That's why people like Puts focus bracket when measuring.

Anyways, try some flare and out of center tests - you might like the DR even more ...

Cheers,

Roland.

 

[sc_rufctr]

Thanks for going to the trouble of doing this.

I've always liked my Summicron DR... It really is my most used lens and considering when it was designed it's an absolute legend.

 

[raid]

Most 50mm lenses are sharp in the center and when stopped down across the coverage. The older lenses that are less sharp wide open provide an alternative for portraits and scenics with less sharpness when used wide open.

The Summicron 50mm DR or Rigid have high resolution across,and they stand the challenges of time.

 

[rogerzilla]

In a harsh flare test the DR and the Elmar would be about the same, while you wouldn't be able to see the test target with the Summar!

The DR is known to have a doughnut of less-sharpness slightly bigger than the vertical height of the frame, so it's kind of this shape on film: ( )

As this is a characteristic of the double Gauss design, the Summar likely has the same problem.

 

[bobby_novatron]

Thank you for going to the trouble of setting up this test. As the owner of both a Summicron DR and an Elmar, I've often wondered what the inherent differences are between them. Your test is definitely illuminating.

In a completely unscientific but perhaps useful test, I shot my DR and Elmar using color film (Kodak 160NC) ... alternating test shots revealed that the DR and Elmar both rendered the color 'palette' in subtle but different ways, probably due to coatings and lens design. The Elmar was definitely cooler, and the DR was much more 'realistic', and perhaps a touch 'warmer' -- with more pleasant soft contrast.

The Elmar is a great B&W lens, but the DR would be a great tool for either B&W or color, IMHO.

 

[David Murphy]

Nice comparison. My experience however indicates to me that any Cron greatly outperforms any Summar or Elmar, so I'm sure your tests are not complete.

I just calculated the diffraction limit for a camera lens at 5000 Angstroms (blue like film response) used at F11, and it is 12.4 microns (Airy dark ring diameter). Small ISO rating film has smaller grain sized than that, as small as 2 microns in diameter, so you are essentially correct about the diffraction limit. There are a few other issues that could muddle this observation, but it gets involved!

These are all fines lenses though when used within their limitations.