Bokeh of Leica 35mm lenses compared (Summicron IV vs Summicron ASPH vs Summilux ASPH)

[Chriscrawfordphoto]

here's a quick auto levels adjustment. it doesn't really add any new information; i think the test is fine with straight scans.

That's because you didn't add enough. Levels adjustments are not enough, you need a fairly serious s-curve as well. Viewscan with a Nikon scanner produces an extremely flat scan that is, as I said and will say again, worthless until post processed. The OP didn't like hearing that. Fine. But I'm right, and I was trying to help him improve his work. Look at the images after I worked on them. The contrast is more natural, with actual midtone contrast. See the difference in Bokeh between the lenses NOW?

 

[Ron (Netherlands)]

In terms of sharpness the Summicron asph is clearly the better lens, but of course that says nothing about the oof quality....

 

[AhtoT]

See the difference in Bokeh between the lenses NOW?

You are talking like the bokeh difference was not visible on original scans. 🙂 Looking at your images on your web site I see you are prefering darker high contrast images, often with blown blacks. That is fine with me, there was a period I liked these too, but that does not not mean everything else is wrong.

 

[Chriscrawfordphoto]

You are talking like the bokeh difference was not visible on original scans. 🙂 Looking at your images on your web site I see you are prefering darker high contrast images, often with blown blacks. That is fine with me, there was a period I liked these too, but that does not not mean everything else is wrong.

You need to calibrate your monitor. If you see blown blacks in my images, yours is WAY off. No wonder you think those scans you posted look good. On a properly calibrated screen they are FLAT. I'm trying to help you here, I do this for a living. If you printed those negs in the darkroom on grade 2 paper I bet they'd look like my versions of your scans look on a properly setup screen. The thing is Viewscan and a Nikon scanner give VERY VERY flat images that MUST be processed to look normal and have normal contrast range and tonal values. Use the histogram in photoshop and the info readout (tells you the actual numeric values of a tone in the image), you'll see what I'm talking about.

 

[nobbylon]

Minimal differences at best. Asph sum is to my eye the nicest but that's the point, it's how my eye enjoys the pictures taken. Let's face it, most people who look at our photos are not going to be mentally comparing to the last ones you showed them or how the oof appears. I must admit that my 35mm 1.7 Ultron looked just as nice as my asph cron but I still had to get one! My advice is try them all and sell what you don't like. Choosing lenses by looking at comparison tests is IMHO of little practical use apart from a bit of fun. Better to use them yourself and decide if they suit your purpose.

 

[LeicaFoReVer]

Chriscrawfordphoto is right to my opinion that the images look flat. It seemed to me too. His images dont have blown blacks.

Nevermind but it is too harsh to say that is why the test is worthless! It shows that the OOF areas of Cron IV are crispier than others. Look at the halo effect at the boundary btw trees and the sky...

The test can be performed on a larger object that has a depth to test what RITskellar says..

Thanks for sharing...some people needs to learn to appreciate the efforts

 

[sanmich]

That's because you didn't add enough. Levels adjustments are not enough, you need a fairly serious s-curve as well. Viewscan with a Nikon scanner produces an extremely flat scan that is, as I said and will say again, worthless until post processed. The OP didn't like hearing that. Fine. But I'm right, and I was trying to help him improve his work. Look at the images after I worked on them. The contrast is more natural, with actual midtone contrast. See the difference in Bokeh between the lenses NOW?

Chris, thanks for adding the contrast to a "real" level
Frankly, yes, I do see a clear difference between the ASPH's and the IV.
The background spots are fuzzier on the ASPH designs, sharper on MR. IV.
Yo may argue that under different conditions, the results may vary, but for this test, I clearly prefer the ASPH's

 

[AhtoT]

UPDATE: I have now adjusted the images with similar curve to add moderate contrast. Also I tried to crop images to be similar so they don't "jump around" so much because different film alignment and feed position when scanning different frames.

P.S. I have never argued that my original images were not flat. In fact the page had note that these were flat scans without post-processing.

 

[Tim Gray]

Yeah no big deal about the contrast. While they were obviously flat (and stated to be so), all of the differences were still apparent. Nice test.