am i expecting to much (summicron 50 on mp set to infinity)?

[el.nino]

hi,

i have a 50 summicron (the last version without the built in hood) on my mp.
from time to time i recognized photos taken with focus set to infinity look kind of unsharp.

i had it serviced by leica just a few weeks ago because i couldn't focus to infinity - just could not turn the focus ring enough. they changed the focus-gear and two lenses (everything but cheap).

last weekend i took some test-shots because most infinity-photos still looked slightly unsharp.
so i compared the 90 summicron and the 50 summicron.
both were focused on the antenna of the building, but with the 50 summicron i still wasn't able to get the viewfinder-images perfectly covered - no problem with the 90.

below you will find the whole "test-scenery" and the corresponding 100%-crop - starting with the 50 summicron.
both photos were taken with 1/1000 at f5.6. the film was fuji sensia 100.
scanned with a nikon coolscan V at 4000dpi without any corrections. ICE was switched off. no photoshop-corrections at all - just cropping and saving.

thanks for your help

[wls.shanghai]

please make the test again and use
a solid tripod and cable release.
maybe change also the film to VELVIA 50 or VELVIA 100

regards
wls

[el.nino]

indeed i used a very solid tripod and a cable-release.
there should not be much difference in resolution between sensia and velvia since sensia is the same as astia (except the aging-factor). i don't think that more contrast and saturation will help me here.

here is one that lead me to do the test. film was kodak E100G.
1/250 at f8 (as far as i can remember). tripod and cable-release again.
the uncropped version is slightly sharpened in photoshop (usm: 50, 0.4, 2).

[el.nino]

may this be the scanner that just can't get the details?
though the photos look slightly unsharp through a 8x-magnifier on a light table as well.

[rob_x2004]

Can you shoot something that is perfectly in focus and not quite out at infinity to compare?

[el.nino]

this one should be pretty well in focus.
f2 1/125 handheld.

no corrections at all. only cropped.

[rob_x2004]

Its an interesting one. Comparing the first 50C image, and the last 50C, there appears a similar level of detail recorded, though on the last it is difficult to find something as small or fine as say the railings on the first. I looked at the shadow gap between the wire and the screw head where the wire is bent round to fix the grill.

I was also wondering if it is a matter of contrasts. The second image appears clearer, but it is all pretty much an even and lower light exposure. The first contains extremes in highlight and adjacent shadow and maybe something of the lenses writing qualities come in to play.

My feeling is that the lens is not quite on the money at infinity focus, but I wonder if that is rangefinder coupling accuracy and whether lenses at infinity stop are tuned just passed focus, or whether there is something in the lens optics that dictates best focus is in a zone where the lens is designed to be used, and to get that the thing becomes tuned away from perfection at infinity.

Hopefully there is someone who understands lens design here at forum.

[el.nino]

thanks, unfortunately i have no slide projector to have a more detailed look at the photos.
i can only evaluate by scanning and with a magnifier on a lighttable. but as i said before, the 50 summicron infinity images look slightly unsharp through a mangnifier as well. at least not like what i expect from a leica summicron that was serviced 3 weeks ago - and they exchanged about have of the parts of the lens.
the next thing is, that by turning the focus ring i just can't match the images in the viewfinder (like the antenna in my first example). it feels that i would have to turn the focusring just a little bit more towards infinity to get a perfect match in the viewfinder. with the 90 summicron and 28 elmarit it's no problem to get a match.

by the way, as far as i read all lenses are optimized for infinity nowadays.
so they should perform best set to infinity.

i am pretty much aware that it is difficult to evaluate all this by looking at slides not scanned with a decent drum-scanner, but i thought i give it a try.
maybe someone else can have a look at those images and give me a hint.
or just upload a nikon coolscan scanned photo at inifity as well. just to have a comparison.

normaly i am not the "testing-lenses"-guy but that really bothers me. the next days i'll repeat the test on a friends m8 and compare the summicron to his 50 summilux asph at infinity. that seems easier than having another test-film developed just to be restricted to my scanners boundaries again.

[el.nino]

as announced before i did the summicron vs. summilux-asph. comparison with a M8 today.
through all f-stops (1.4/2 - 8) the summilux is far superior.
i know the summilux is aspherical, but isn't the summicron said to be one of leicas sharpest?

all on tripod. M8, ISO160, f8, 1/60

[Olof]

...i know the summilux is aspherical, but isn't the summicron said to be one of leicas sharpest?...

That was in the past, actually the Summilux 50mm ASPH is the better one

[el.nino]

but THAT better?

[delander]

My year old Summicron 50mm is very sharp on the M8, I have also recently got the Summilux ASPH, dont yet see a real difference at f5.6 between the two.

See Summicron photos especially the crop. These are RAW from the camera using default settings in C1. I must admit when I first saw your post I thought there was something wrong with your copy of the lens, but was not sure as you were using film. Hope this helps,

Jeff

[ndjambrose]

I'd suggest there's definitely something wrong with your lens.

I have the latest version of the 50 summicron and it is acutely sharp. Even shooting wide open at infinity the sharpness is not dissimilar to the image you shot above with your summilux at f8. Being unable to align the RF at infinity would also trouble me. There's no reason why this should happen except for poor calibration in the RF mechanism or the lens gearing; it's certainly not something that happens on any of my cameras.

[el.nino]

thanks everbody. contacted cs today. haven't got a reply yet, but
will send it in on monday. hopefully will be back by the end of the week.

[marknorton]

I agree, the Summicron is out of adjustment. Did I understand you correctly that you had some of the lens elements replaced? If so, they may not be seated properly. Otherwise, it's possible that the lens is focussing beyond infinity, meaning that when you set the lens to infinity, it is actually too close to the focal plane and then nothing is in focus, anywhere.

You might try backing the lens off from the infinity setting by small amounts to see if the focus improves. This is exactly the problem I had with my Noctilux - set to infinity, nothing in focus.

In a test like this, it's worth avoiding using the rangefinder and just bracket focussing. As soon as you bring the rangefinder into the equation, that's one more thing which might be out of adjustment to complicate the test. The Summilux is working well though.

[BTW, handy infinity reference for you to have available with lots of detail to check on the spire and edges against the sky to show up any fringing...]

[darkstar2004]

Something does appear to be "not right" with your lens.

If it were me, I'd send it in for a clean/lube/adjust and then see how it does.

[el.nino]

i called cs today and sent it in.
so i have to wait now.

thanks everyone