The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Summicron M 35mm ASPH - soft center image at mid-aperture

Frits

Member
I recently got a 35mm Summicron ASPH for my M9 (2012 serial number).
I am surprised and disappointed to see definite softness in the image center at the f 4 - 5.6 range. The image is fine away from center all the way into the corners. At larger apertures it seems OK and it is fine again at f8.
The MTF chart does not seem to reflect this softness at all, nor did I expect it at what is rated to be an excellent lens.

I would appreciate comments on your experience with this lens, as I suspect (hope?) that the issue is an isolated one.
 

glenerrolrd

Workshop Member
Frits

This is most likely a misfocus . Most M lenses have some field curvature so the edges may focus somewhat closer than the middle . So if you slightly backfocus ..the edges may look good and the middle maybe back .

Easy to test ..take a hat and place in a large flat surface...grass , concrete etc ..but something where you can easily see the plan of focus . You may have a camera or lens that requires calibration .
 

Shashin

Well-known member
Can you post some images showing this effect at different apertures? It would be odd that your lens is better both at larger and smaller apertures.
 

Frits

Member
Thanks all for your comments.
Here are some images to illustrate:

Overview, followed by some 100% crops from the left, center and right of the image from f 2.0 through f 8.0.

Center: Wide open is as expected and gets better at f 2.8. But then notice how soft the center is at f 4.0 and still some at 5.6. From f 8.0 onward all is well.

I don't think that focusing is an issue here, as it is at a bit of a distance at a somewhat wide angle (35mm). That combined with the DOF at 4.0 and 5.6 makes me think so.











 

Jeff S

New member
Your lens is exhibiting normal focus shift. Sean Reid's review sample showed some focus shift at f4 and f5.6, then fine again by f8. My own lens showed identical results after some careful tests.

The 35 Summilux has received notable attention for its focus shift, enough to warrant a new lens with floating element. But this is a common phenomenon for fast(ish) lenses, and the Summicron is no exception. The real issue IMO is whether prints suffer, and in most cases (at least at my print sizes) there is no practical issue.

Jeff
 

edwardkaraa

New member
Your lens is exhibiting normal focus shift. Sean Reid's review sample showed some focus shift at f4 and f5.6, then fine again by f8. My own lens showed identical results after some careful tests.

The 35 Summilux has received notable attention for its focus shift, enough to warrant a new lens with floating element. But this is a common phenomenon for fast(ish) lenses, and the Summicron is no exception. The real issue IMO is whether prints suffer, and in most cases (at least at my print sizes) there is no practical issue.

Jeff
Focus shift is caused by spherical aberrations, an aspherical lens is not supposed to have focus shift, unless the ASPH is just for marketing purposes.
 

Frits

Member
It seems to be front focusing to me. Is there any plane of sharp focus in the foreground?
Spot on, issue resolved!
Your remark got me thinking and I went and shoot a few. It front focuses by a fair bit.

Instead of near infinity, I need to focus at approx. 30ft / 10m when targeting the building front door.
When I focus on the white van (center, just across the street) at f4, most everything is soft. The van is in perfect focus when the focus ring is offset the same as mentioned before, with the far items then being soft (out of focus really, which is perfectly normal).

That said, I have a lens that exhibits a good bit of front focusing, which really annoys me.
I bought it used from a very reputable dealer, with a Leica passport and much of the original guarantee (the original owner purchased it in June of this year).
I can return it and buy a new one (and pay the difference), or have it sent in for calibration (and lose it while it is gone).
Rats!!
 
Last edited:

Jeff S

New member
Focus shift is caused by spherical aberrations, an aspherical lens is not supposed to have focus shift, unless the ASPH is just for marketing purposes.
Funny then that the 35 Summilux asph needed a second version with floating element to reduce focus shift, huh. A complex lens can have one or more aspherical elements, and a variety of other design considerations.

Jeff
 

Jeff S

New member
Spot on, issue resolved!
Your remark got me thinking and I went and shoot a few. It front focuses by a fair bit.
But it only does this at 2 mid-apertures, right? If so, you're experiencing focus shift as I described.

Jeff
 

Shashin

Well-known member
But it only does this at 2 mid-apertures, right? If so, you're experiencing focus shift as I described.

Jeff
That does not make sense as why does the focus shift again at smaller apertures? Focus shift because of SA simply gets better as the lens is stopped down with the maximum aperture being the worst. Focus calibration makes a great deal more sense here.
 

Jeff S

New member
That does not make sense as why does the focus shift again at smaller apertures? Focus shift because of SA simply gets better as the lens is stopped down with the maximum aperture being the worst. Focus calibration makes a great deal more sense here.
You better write Sean Reid and tell him that his review is incorrect regarding the 35 Summicron asph. And what are the odds that my lens matches his erroneous findings?

To repeat, Sean's test lens (and mine) is fine at f2 and f2.8, then exhibits shift at f4 & f5.6, then returns to normal. (I believe the OP's findings were consistent with this, unless I misunderstood.) Of course distance to subject and depth of field remain variables. And, as I said, it's not a practical matter in prints.

Jeff
 
Last edited:

edwardkaraa

New member
Funny then that the 35 Summilux asph needed a second version with floating element to reduce focus shift, huh. A complex lens can have one or more aspherical elements, and a variety of other design considerations.

Jeff
The purpose of floating elements is to improve close focus performance.

Anyhow, if this is the summicron normal behavior, I would prefer a coke bottle instead. My Biogon 35 is light years ahead.

But I insist, the lens needs calibration. It is front focusing.
 

Jeff S

New member
The purpose of floating elements is to improve close focus performance.
Yep, it's called focus shift. And if calibration could solve the issue on the first version of the 35 Summilux asph, why do you suppose Leica created a second version with floating element to solve the problem?

The Summicron asph is no different in principle, just in degree, and in the apertures affected. It's an inherent characteristic of many fast lenses. I can assure you that the Summicron is far better than a coke bottle, and you'd never notice the issue in my prints, which is all I care about. Leica checked it out for me and confirmed that it's perfectly calibrated...and that the shift is normal.

Jeff
 

bubuli

Member
Yep, it's called focus shift. And if calibration could solve the issue on the first version of the 35 Summilux asph, why do you suppose Leica created a second version with floating element to solve the problem?

The Summicron asph is no different in principle, just in degree, and in the apertures affected. It's an inherent characteristic of many fast lenses. I can assure you that the Summicron is far better than a coke bottle, and you'd never notice the issue in my prints, which is all I care about. Leica checked it out for me and confirmed that it's perfectly calibrated...and that the shift is normal.

Jeff
Jeff is correct: Aspherical element is not for correcting focus shift for the 35 Summilux; the floating element does the focus shift compensation.

That said, this may be a case of field curvature as Graham noted...since the center and sides seem to be in the same plane of focus.

I have a 35 Summicron ASPH and this is an interesting discovery (at least for me)...I may have to conduct my own tests to see if this field curvature exists for my copy.
 

jaapv

Subscriber Member
It depends on the calibration of the lens whether focus shift improves or degrades the image at middle apertures. If one shoots wide open most of the time it is best to have the lens optimized wide open and avoid (or compensate at) middle apertures. Focus shift continues all through the aperture range, but it falls within DOF at the smaller apertures.
 

Frits

Member
I understand the field curvature an that it will cause uneven sharpness, especially at wide apertures.
I did some more testing with this lens and at the image center the actual focus point simply lies a fair bit behind where I focus with the rangefinder. It is not the camera, as I have verified with a few other lenses where it is spot-on.
I can deal with fall off, as long as I can focus correctly at image center. This lens does not - it is hosed...
 
Top