M8 behaves wierd with certain lenses

Makten said:
Sure, but this is still not about accuracy. I get the same misfocus everytime.

----------------------

Anyhow, it's pitch black outside here now (8 PM) so I can't do the infinity check for the 28/2. I'll forget about the Summicron for now and first try to confirm that the camera is OK.

Thanks for the advice so far! 🙂
Click to expand...
No, but because of the steep curve the whole tolerance is less. The infinity end may be within tolerance for a CL, but still wildly off for an M8 - those two cameras are at the far ends of the scale for focussing tolerance.
If you really want to go forward with this lens I would advise you to send it to Will van Manen for calibration. He can put it right in the middle of its tolerance span.
 
Last edited:
Makten said:
Maybe, but it's wierd. DOF is usually larger towards the corners due to the cosine^4 rule.
Click to expand...

With all due respect, you guys are mixing everything up here.
The COC is the same all over the image, as it is a convention, not a physical circle. The Airy disks thrown by the lens, however, are more ellipsoid and larger towards the corners because of geometrical considerations. On film even more than on a sensor, as film has a thickness and the Airy disk will be projected skewed in three dimensions. But normally lens resolution is high enough to let the Airy disks be much smaller than the COC, so this effect is of no relevance for this problem.The articles you refer to are about vignetting, not about sharpness and do not apply here.

Anyway, DOF and the size of the COC is ruled (nearly) solely by enlargement, nothing else. Assuming the sensor resolution is larger than the lens resolution (when the other way around -let's not go there - it gets complicated then), if you are looking at the image at 100% the COC is identical to the pixel size, 6.8 micron in this case, and the DOF is the "native" DOF of the sensor. So it will not have any bearing on the problem being discussed here.

Normally this lens is fairly even from centre to corners with the performance evening up at about 5.6. So it should not have extreme sharpness falloff as is you are seeing. Something must be wrong, maybe an element decentred or what have you. Btw the image you linked to is so oversharpened that is it really useless for judging the lens with. Don't you have another that is less heavily USM-ed?

I would advise you to dump this lens and buy an Elmarit 2.8/90, which is a beautiful little lens with an excellent performance. In fact one of the best 90ies ever built, and they are really not that expensive.
 
Last edited:
jaapv said:
With all due respect, you guys are mixing everything up here.
The COC is the same all over the image, as it is a convention, not a physical circle. The Airy disks thrown by the lens, however, are more ellipsoid and larger towards the corners because of geometrical considerations. On film even more than on a sensor, as film has a thickness and the Airy disk will be projected skewed in three dimensions. But normally lens resolution is high enough to let the Airy disks be much smaller than the COC, so this effect is of no relevance for this problem.The articles you refer to are about vignetting, not about sharpness and do not apply here.

Anyway, DOF and the size of the COC is ruled (nearly) solely by enlargement, nothing else. Assuming the sensor resolution is larger than the lens resolution (when the other way around -let's not go there - it gets complicated then), if you are looking at the image at 100% the COC is identical to the pixel size, 6.8 micron in this case, and the DOF is the "native" DOF of the sensor. So it will not have any bearing on the problem being discussed here.
Click to expand...
Sorry, but I don't understand why you are talking about Airy discs. This is not a diffraction issue. The CoC (or "blur disc" when out of focus) will of course be smaller (non-circular) off-axis if the projection is flat. This is one of the factors that gives natural vignetting.

Anyway, the reason I mentioned it was because of your suggestion that the CoC:s could be larger towards the corners behind the plane of focus. I don't understand why they would be that for other reasons than curvature of field.

Normally this lens is fairly even from centre to corners with the performance evening up at about 5.6. So it should not have extreme sharpness falloff as is you are seeing. Something must be wrong, maybe an element decentred or what have you.
Click to expand...
Funny that the CV 35/1.4 did the exact same thing. But it isn't impossible of course.

Btw the image you linked to is so oversharpened that is it really useless for judging the lens with. Don't you have another that is less heavily USM-ed?
Click to expand...
The image is almost not sharpened at all. 25%, 1 pixel radius in ACR. Noting more.

I would advise you to dump this lens and buy an Elmarit 2.8/90, which is a beautiful little lens with an excellent performance. In fact one of the best 90ies ever built, and they are really not that expensive.
Click to expand...
90??? Why on earth would I buy a 90 mm lens as a substitute for a 40? 🙄
 
Last edited:
Makten said:
Sorry, but I don't understand why you are talking about Airy discs. This is not a diffraction issue. The CoC (or "blur disc" when out of focus) will of course be smaller (non-circular) off-axis if the projection is flat. This is one of the factors that gives natural vignetting.
Click to expand...
No. The COC is a defined parameter, not an actual physical circle, so geometry does not apply. And it has nothing whatever to do with vignetting. You are really confusing it with the resolution computation of the lens.
Anyway, the reason I mentioned it was because of your suggestion that the CoC:s could be larger towards the corners behind the plane of focus. I don't understand why they would be that for other reasons than curvature of field.
Click to expand...
I never made that suggestion😕

Funny that the CV 35/1.4 did the exact same thing. But it isn't impossible of course.
Click to expand...
That lens is known for CFA, the Summicron not. What it is known for, however is loss of contrast in the field, so maybe you are seeing that.
The image is almost not sharpened at all. 25%, 1 pixel radius in ACR. Noting more.
Click to expand...
I believe you, but it looks very sharpened on my monitor; maybe a different subject matter, the proverbial brick wall, could tell us more.

90??? Why on earth would I buy a 90 mm lens as a substitute for a 40? 🙄
Click to expand...
Well, maybe you shouldn't. I was indeed confused by post #8. But substitute " any Summicron 2.0/50" for that, and you cannot go wrong.
 
Last edited:
jaapv said:
No. The COC is a defined parameter, not an actual physical circle, so geaometry does not apply. And it has nothing whatever to do with vignetting. You are really confusing it with the resolution computation of the lens.
Click to expand...
Well let's call it "blur disc" then, so that you understand what I'm talking about.

I never made that suggestion😕
Click to expand...
You did: "Maybe because you are close to the hyperfocal distance ? COC being larger in the corners ?"

That lens is known for CFA, the Summicron not. What it is known for, however is loss of contrast in the field, so maybe you are seeing that.
Click to expand...
There's still no explanation to how the tree branches in the corners can be sharp, while the edges further away are very unsharp, combined with a sharp middle at infinity.

I believe you, but it looks very sharpened on my monitor; maybe a different subject matter, the proverbial brick wall, could tell us more.
Click to expand...
Obviously it's a sharp lens where it's sharp. At least something good about it! 😀 Yes, I'll have to do the brick wall test.

Well, maybe you shouldn't. But then, maybe I should hit the 5 instead of the 9 on my keyboard 😉.
Click to expand...
Hahaha!
 
Last edited:
To clarify (from Wikipedia): "In optics, a circle of confusion is an optical spot caused by a cone of light rays from a lens not coming to a perfect focus when imaging a point source. It is also known as disk of confusion, circle of indistinctness, blur circle, or blur spot....In photography, the circle of confusion (“CoC”) is used to determine the depth of field, the part of an image that is acceptably sharp."


I meant it in the optics sense. In contrast to vignetting, size of the blur spots is not related to the cos^4 law, AFAIK. Blur spots usually get eliptical and somewhat longer when moving away from picture center.

The Summicron could be de-centered as Jaap suggests. But it could just be mis-colimated, and normal behavior too, not sure about that. It could also be a mis-aligned camera flange or sensor.

To exclude the latter, any luck with your 28 at infinity ?

Roland.
 
Last edited:
OK, let me clarify this thread with some facts:

Wikipedia is usually wrong.

For Leica lenses, the CoC is strictly defined to an accurate .023mm, not a "pixel" or any other but the constant .023mm.

CV, Minolta, the Summicron C hybrid lenses are what you have is what you get.

To the OP: most likely someone realized this lens had issues and dumped it. Get it fixed, or get another lens.
 
RFF is really becoming more and more like pnet used to be .... What does the OPs problem have to do with Wikipedia being wrong (I used it to clarify my terminology, apparently also used by others), Leica lenses being better than CV lenses, etc.

Before dumping the Summicron, I would check the camera. The OP had problems with other lenses, too; that have been successfully used by other M8 users. Again, this would be my process of analysis:

- check camera sensor and flange for alignment and registration distance accuracy. Use a reference lens (28).
- If OK:
...- check alignment and registration distance (collimation) of Summicron; correct if necessary
...- If OK:
......- check RF coupling mechanism,
.........- camera first; correct if necessary
.........- lens second; correct if necessary

Roland.
 
Last edited:
Just trying to help

Just trying to help

The M8 is relatively new compared with these lenses, and rarely if ever is the flange or sensor distance broken (or have I just not heard of these issues)?

It's common knowledge, possibly even in Wikipedia 😉, that often samples of the CV 35/1.4 has focus shift, and that the Summicron C 40/2 is not supported by Leica (note - no 6-bit code, no entry in table for M9, etc.).

It's also known that while Leica published it's CoC spec constant for design, that it's not clear if that covers the Summicron C. Also, I don't believe CV has ever said what CoC spec it designs for.

To me, it seems very likely the OP picked up a couple bargain bin lenses that are slightly off. His M8 is probably perfect wrt mount and sensor distance.

ferider said:
RFF is really becoming more and more like pnet used to be .... What does the OPs problem have to do with Wikipedia being wrong (I used it to clarify my terminology, apparently also used by others), Leica lenses being better than CV lenses, etc.

Before dumping the Summicron, I would check the camera. The OP had problems with other lenses, too. That have been successfully used by other M8 users.
Click to expand...
 
Geeze. The lens doesn't work on the camera. Why it doesn't work doesn't seem to be the main problem if you want to take photos with it. Fix it or throw it away. Strangest debate I've read in a while.
 
re: Wikipedia:

My 5th year old learned this at school about Wikipedia: - Are his teachers wrong??

1. Anyone can make or edit entries.

2. The person who made or edited the entries can be mistaken, or have motives to purposely provide wrong or biased information.

3. The information can be outdated.
 
Pickett Wilson said:
Geeze. The lens doesn't work on the camera. Why it doesn't work doesn't seem to be the main problem if you want to take photos with it. Fix it or throw it away. Strangest debate I've read in a while.
Click to expand...

Both 40 Summicron and 35/1.4 did not work at infinity on the OPs camera. Infinity excludes possible 40 RF coupling issues. And the aperture that the OP used on the 35 excludes focus shift issues. For me this points to one ore more possible camera problems. Not like the M8 is the most reliable Leica out there. Trying yet another lens on the same possibly broken camera might turn out even more frustrating, don't you think ?
 
Last edited:
ferider said:
To exclude the latter, any luck with your 28 at infinity ?
Click to expand...
Yes, I think so. It definitely reaches infinity wide open, even if the DOF is so deep that I don't know if it's sharpest a little closer to the camera. And since this lens also shifts focus, it's definitely at infinity at f/2.8.
On the other hand, the borders are quite unsharp. But I wouldn't expect them to be sharp when wide open at infinity. The lens is alot better at close distance regarding overall performance. The border unsharpness also looks very different to that seen in the Summicron shot. It's just hazy, not blurred.

So, I think the Ultron is OK at any distance, but it's not a perfect lens.

ampguy said:
To the OP: most likely someone realized this lens had issues and dumped it. Get it fixed, or get another lens.
Click to expand...
The CL, Summicron-C and Elmar-C was my grandfathers. The camera might have been dropped, because there is a tiny crack in the RF patch window on the front. But no dents or anything indicating it would have been a hard drop.
The Summicron has also been in for lubrication very recently, but I don't know if they checked anything else.

--------------------------

And please, let's cut the crap with CoC:s and all that. It doesn't belong in this discussion. 🙂
 
Last edited:
Ok

Ok

So given all the info. provided, I think it would be best to take the camera to a dealer and to try other known good M lenses.

If the known good lenses don't focus perfectly near and at infinity on your M8, your M8 needs to be looked at.

If the known good lenses focus near and far on your M8, then you should have your lenses looked at.

I agree that it is interesting that your 28 Ultron semi-works, but if you use this anecdotal piece of information, to assume your camera flange and/or vf/rf assembly is fine, you could be chasing your tail in fixing a lot of things, other than the root problem, the first time around.

In a nutshell, I would not use a 28 Ultron as a "reference" lens.


Makten said:
Yes, I think so. It definitely reaches infinity wide open, even if the DOF is so deep that I don't know if it's sharpest a little closer to the camera. And since this lens also shifts focus, it's definitely at infinity at f/2.8.
On the other hand, the borders are quite unsharp. But I wouldn't expect them to be sharp when wide open at infinity. The lens is alot better at close distance regarding overall performance. The border unsharpness also looks very different to that seen in the Summicron shot. It's just hazy, not blurred.

So, I think the Ultron is OK at any distance, but it's not a perfect lens.


The CL, Summicron-C and Elmar-C was my grandfathers. The camera might have been dropped, because there is a tiny crack in the RF patch window on the front. But no dents or anything indicating it would have been a hard drop.
The Summicron has also been in for lubrication very recently, but I don't know if they checked anything else.

--------------------------

And please, let's cut the crap with CoC:s and all that. It doesn't belong in this discussion. 🙂
Click to expand...
 
Last edited:
ampguy said:
So given all the info. provided, I think it would be best to take the camera to a dealer and to try other known good M lenses.
Click to expand...
The problem is that I live in Sweden, and the few Leica dealers that we have, seldom have any lenses. You have to buy one to get one.

If the known good lenses don't focus perfectly near and at infinity on your M8, your M8 needs to be looked at.

If the known good lenses focus near and far on your M8, then you should have your lenses looked at.

I agree that it is interesting that your 28 Ultron semi-works, but if you use this anecdotal piece of information, to assume your camera flange is fine, you could be chasing your tail in fixing a lot of things, other than the root problem, the first time around.

In a nutshell, I would not use a 28 Ultron as a "reference" lens.
Click to expand...
Nor would I, but that's the only one I got at hand. I've ordered a Zeiss 28/2.8 already, and I will test it before actually buying it. Hopefully it works fine.

Regarding the wierdness of the problems, I've seen pictures from the CV 35/1.4 with the exact same blurred borders and corners on M8 as I got with mine. Also, if the bayonet or sensor would be out of alignment, it should really be possible to see the actual tilt of the plane of sharpness in the pictures. I mean, you can't get something looking like curvature of field from misalignment. But perhaps you could from a decentered lens.
 
You must log in or register to reply here.
Back
Top Bottom