75mm summarit problem?

woodleica · 12-05-2009

Hey guys, I've never seen anything quite like this before. This looks likes purple fringing. Is it the lens, the sensor or what? Shot with my new M8 and 75mm summarit lens. I'm not using a UVIR filter. Also, I don't have a lens hood yet ( on order ). Any input appreciated. Attached is shot and 100% crop of an area. Also, I had looked at Sean Reids review of the lens before purchasing, and now that I look again, notice that the chromatic aberration portion of the review is missing.

payasam · 12-05-2009

No lens should do this. Know nothing about the M8.

35mmdelux · 12-05-2009

try the IR filter to eliminate this problem -

f 1,0 · 12-06-2009

i think you use Lightroom
try Capture One, it can reduce the purple fringing

swoop · 12-06-2009

I'd guess UVIR filter is the solution.

Fujitsu · 12-06-2009

Pretty normal.

woodleica · 12-06-2009

Quote:

Originally Posted by payasam

No lens should do this. Know nothing about the M8.

I agree with you which is why I was asking

woodleica · 12-06-2009

Quote:

Originally Posted by Fujitsu

Pretty normal.

Can you explain what you mean by "pretty normal" ?

fbf · 12-06-2009

Quote:

Originally Posted by woodleica

Can you explain what you mean by "pretty normal" ?

"normal" output directly from m8 w/o using uv/ir filters. Try uv/ir filters like people have mentioned above, see if there is any improvement.

woodleica · 12-06-2009

Quote:

Originally Posted by f 1,0

i think you use Lightroom
try Capture One, it can reduce the purple fringing

Thanks, I know software can help and that's not my issue, I'm more concerned about the lens exhibiting this in a situation like this. It should NOT happen.

woodleica · 12-06-2009

Quote:

Originally Posted by fbf

"normal" output directly from m8 w/o using uv/ir filters. Try uv/ir filters like people have mentioned above, see if there is any improvement.

I had not thought this to be an IR issue and it looks like you are saying it is. I am still waiting for my IR filter to arrive so until then, I cannot try it out, and also, I have to hope for another sunny day in Seattle in winter.

payasam · 12-06-2009

"I'm more concerned about the lens exhibiting this...".

You shouldn't blame the lens unless other lenses don't do this.

woodleica · 12-06-2009

Quote:

Originally Posted by payasam

"I'm more concerned about the lens exhibiting this...".

You shouldn't blame the lens unless other lenses don't do this.

Depends on your point of view - you could then blame all the lenses. However, to add I would like to think that I am not "blaming" the lens at all. I have observed an issue and would like input from the experts on this issue.

payasam · 12-06-2009

I meant the known facts, woodleica, not anyone's point of view. If similar photos taken with other lenses show the same effect, then indeed one could blame all lenses. However, that would involve standing the science of optics on its head. If you have used only the 75mm on the M8, you do not know enough to come to any conclusion and it would be best to stick to the filter hypothesis for the present.

jaapv · 12-07-2009

It is not the IR filter, it is not chromatic aberration and it is not sensor blooming, although all those factors can intensify the effect . It is an effect of high contrast edges in RAW conversion. ACR (Lightroom) is notorious in this repect. The best conversion software I kinow for this is C1 5.0.1 with purple defringing swiched on.
The problem is that in demosaicing the near-100% contrast jump from one pixel to the next screws up the Bayer colour interpolation, and as green is the most number of sites it shows up as magenta. The severity depends on the demosaicing algorithms. Perversely, the better the lens and the sharper (no AA filter) the sensor, the worse the problem. I usually lasso the problem areas and desaturate magenta, if the RAW conversion cannot cope.

woodleica · 12-07-2009

Quote:

Originally Posted by jaapv

It is not the IR filter, it is not chromatic aberration and it is not sensor blooming, although all those factors can intensify the effect . It is an effect of high contrast edges in RAW conversion. ACR (Lightroom) is notorious in this repect. The best conversion software I kinow for this is C1 5.0.1 with purple defringing swiched on.
The problem is that in demosaicing the near-100% contrast jump from one pixel to the next screws up the Bayer colour interpolation, and as green is the most number of sites it shows up as magenta. The severity depends on the demosaicing algorithms. Perversely, the better the lens and the sharper (no AA filter) the sensor, the worse the problem. I usually lasso the problem areas and desaturate magenta, if the RAW conversion cannot cope.

Ahh, THAT makes sense - thank you. It's been a while since I read about RAW conversion. I'm assuming then that RAW images from the Foveon don't suffer from this since interpolation isn't needed.

f 1,0 · 12-07-2009

Quote:

Originally Posted by f 1,0

i think you use Lightroom
try Capture One, it can reduce the purple fringing

???????????

woodleica · 12-07-2009

Quote:

Originally Posted by f 1,0

???????????

You had the solution right and but I was responding to understanding the cause in the earlier mail - i.e that it's a problem with RAW conversion, not the lens or the sensor. Having understood that, clearly the solution you prescribed earlier is correct. I don't like applying solutions without knowing the cause.

ruslan · 12-09-2009

Yes, it is purple fringing and it is normal for no AA, pretty easy fix in any RAW converter, but why no ir/cut filter? But, Ir/cut filter will not fix fringing though, only software.