Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Horrible CA in Contax Sonnar 135 f/2.8
Post new topic   Reply to topic
Home   Manual Focus Lenses
View previous topic :: View next topic  


PostPosted: Tue Dec 09, 2008 3:51 am    Post subject: Reply with quote
For a rather thorough explanation of chromatic aberration, see http://www.vanwalree.com/optics/chromatic.html.

Veijo


PostPosted: Tue Dec 09, 2008 5:31 am    Post subject: Reply with quote
What is your opinion about the posted samples, Veijo?


PostPosted: Tue Dec 09, 2008 1:02 pm    Post subject: Reply with quote
In the review of the zeiss 85 1.4 (sony mount) at photozone.de one can read.

"LoCAs (non-coinciding focal planes of the various colors), sometimes called "bokeh CAs", can be a problem in the field. Similar the Canon 85mm f/1.2 USM L and Nikkor 85mm f/1.4 the Zeiss does also suffer from this optical defect. As you can notice below the halos have different colors - magenta (red + blue) in front the focus point and green beyond. The problem is rather pronounced at f/1.4. It decreases slowly the further you stop down but it is still visible at f/4. The Zeiss ZA 135mm f/1.8 is probably a little better here thanks to its 2 ED elements - the ZA 85mm f/1.4 is a more conventional design. "

Look at the review here with detailed pictures.
http://www.photozone.de/sony-alpha-aps-c-lens-tests/374-zeiss_za_85_14?start=1

It seems to be the exact same thing that you experience.

According to this page at least, it is an optical problem and not a sensor problem. (A problem shared by many exspensive optics, it seems.)

/Jan


PostPosted: Tue Dec 09, 2008 6:09 pm    Post subject: Reply with quote
lulalake wrote:

Hi, Wariag,

I respectfully disagree with parts of your statement. (and agree wholeheartedly with others)

First, "bloom" can be considered a "chromatic aberration" as it involves color and color errors however the "chromatic aberration" of which you and I are speaking is that which you accurately described; the inability of a lens or lens system to focus different colors of light at the same point.

The problem here is that the sensor bloom is not that at all . .but the result of a miscommunication, overloading, and inability do handle high contrast of signal in and between individual photosensors, not completely unlike noise in a sensors.

A great lens on a digital camera will cause no less blooming than a poor lens under the same circumstances. You are correct, all digital systems have some bloom, from the best Canon L Glass, to the best Leica glass, to adapting ones camera to a great Cassagrain telescope.

Again, there is NO bloom using film, only the type of chromatic aberration that comes with no achromatic adjustments.

Cheers

Jules


I understand what the "bloom" is, and AFAIK that it's the main cause of magnification (or amplification) Smile of CA effects on dSLR comparing to film film SLR.
As the i.e. "blue fringe" CA can be described as not wanted peaks of the light intensity in the blue channel, sensors responsible for this channel gets easily overloaded - thus the range of the CA disturbed area in digital image can spread more than on film -of course green and red channels can be affected too with result of different tint of the aberrated area.
I think, that it's just enough to know that dSLR sensor is just much more sensitive for such not desired special effects as chromatic aberration, and that 35mm film is far more forgiving.
From the other hand, digital post processing software provides tools to nicely correct any kind of "fringing" - usually the stronger is CA - the easier is to correct it Smile,


PostPosted: Fri Jan 23, 2009 7:37 pm    Post subject: Reply with quote
lulalake wrote:
no-X wrote:
lulalake: so please explain to us, why the reflections between lens and focus are purple and the reflections between focus and infinity are green, if you say, that the lens is not the cause of this (d)effect Wink


The color green should be a clue in this issue. Chromatic Aberration, which I studied in optical theory (way back in the day) involves red and blue light not focusing on the same point. There is no green involved.


Well, no: if red and blue are focusing together in one place, then clearly that will produce magenta, while green will be on the other side.



edited to be a bit more polite

Last edited by ChrisLilley on Sat Jan 24, 2009 2:16 am; edited 1 time in total


PostPosted: Fri Jan 23, 2009 7:41 pm    Post subject: Reply with quote
uuhh, that's a cold one Chris....


PostPosted: Fri Jan 23, 2009 9:38 pm    Post subject: Reply with quote
The Sonnars are known to show what I've seen called "secondary" CA, in green/magenta, and if you want to demonstrate it for yourself then just open your Sonnar wide and shoot something with nice bright edges in front and behind. Like I did with my 4/300 just now to get this:



That is a daylight-balanced flourescent bulb at 100% from a 5D and the CA is obvious. However, this is a very, very good lens. I just have to be careful with difficult lighting.

The original examples look very extreme so maybe the lens is a poor copy. I don't know if the degree of CA varies from lens to lens or not.

It is certainly not true to say that modern lenses do not suffer from CA. It is generally better corrected than in the old Sonnars but it is still troublesome on the Canon L zooms, including the 24-70 f2.8 and the 17-40 f4, and under some conditions can be quite horrible. If you have modern lenses and you don't see it, then you are probably not studying every pixel in your pictures at 100%.

I think the owner of this lens should try it out under different conditions before deciding whether to ditch it or not. Backlit bare tree branches are a challenge for any lens.

Last edited by PaulC on Sat Jan 24, 2009 12:49 am; edited 1 time in total


PostPosted: Fri Jan 23, 2009 11:42 pm    Post subject: Reply with quote
Yes, all these samples are definitely bad cases of longitudinal chromatic aberration (LCA) and have nothing to do with sensor bloom. LCA is also called axial CA or bokeh CA, because it is mostly visible as green fringes in out of focus background. LCA affects many lenses that enjoy a very good reputation, like the Zeiss Sonnar 135/2.8, the Takumar 135/3.5 or the modern Pentax FA 35/2 for example. "Ugly green bokeh" bothers me a lot, that's why I don't like the three aforementioned lenses

If you want a lens free from this aberration, look for a good APO lens. For example, the Voigtländer APO Lanthar 125/2.5 and 180/4 lenses are known to be free from these aberrations.

Cheers!

Abbazz


PostPosted: Sat Jan 24, 2009 2:11 am    Post subject: Reply with quote
kds315* wrote:
uuhh, that's a cold one Chris....


You are right. Sorry to be snippy.


PostPosted: Sat Jan 24, 2009 2:29 am    Post subject: Reply with quote
vilva wrote:
For a rather thorough explanation of chromatic aberration, see http://www.vanwalree.com/optics/chromatic.html.

Veijo


Thanks for the link - and the references contained within it.


PostPosted: Sat Jan 24, 2009 12:00 pm    Post subject: Reply with quote
PaulC wrote:
The Sonnars are known to show what I've seen called "secondary" CA, in green/magenta, and if you want to demonstrate it for yourself then just open your Sonnar wide and shoot something with nice bright edges in front and behind. Like I did with my 4/300 just now to get this:

(...)
It is certainly not true to say that modern lenses do not suffer from CA. It is generally better corrected than in the old Sonnars but it is still troublesome on the Canon L zooms, including the 24-70 f2.8 and the 17-40 f4, and under some conditions can be quite horrible.
(...)

Backlit bare tree branches are a challenge for any lens.


The example you have is indeed exactly what my experience is, I was just surprised to see it this pronounced in a lens with such a reputation. Doesn't mean I don't like it for the Zeissness of less difficult subjects though Smile

I also own some L-glass, like the aforementioned 24-70/2.8 and this one does suffer from CA, but not so much the green/magenta type. Most of its CA is quite correctable with Lightroom's defringe option.

Thanks for all the info to all, quite informative thread methinks.

Crispian


PostPosted: Sat Jan 24, 2009 1:18 pm    Post subject: Reply with quote
Doubtlessly you already know the page, but coincidently with the problem that is analyzed here, the question of CA in zeiss 135mm len appears in

http://www.lupomesky.cz/czj_vs_cz/comp135.html

I am conscientious of the relativity of its data, and that the base is not scientist, but the images are clear.

I believe that the question would be to ask if the CA is a bloom question, why only the Zeiss shows it and not the CZJ.

Rino


PostPosted: Fri Feb 06, 2009 10:15 pm    Post subject: Reply with quote
Hi,

The effect seen on my example pictures is definitely not sensor blooming, as it does not appear as much on other lenses as much as here.

Even though I was quite disappointed with this lens at first, I think I will keep it now, because if you use it under better circumstances, its rendering is as fabulous as my Planars.

Just took some lucky mugshots of my youngest daughter and the Zeissness is really there in the Sonnar, especially if you magnify to 100%.

http://picasaweb.google.com/c.a.stones/IsabellaPhotoshoot#5299644536789472850


Regards,

Crispian


PostPosted: Sat Feb 07, 2009 8:16 am    Post subject: Reply with quote
I'm glad to hear it has won you over. As I understand it, Zeiss were able to get the Sonnar sharpness only at the expense of allowing the secondary CA, though they did manage to eliminate first-order CA. In the Tessar, they were able to get a sharp central region (wide open) and no CA, but the price for that was softness at the edges. The Sonnar bokeh was very good, with only slight brightening of the bright circle edges, the Tessar bokeh has a pronounced central bright spot and creates "clumpy" backgrounds (that some people like), so it's all swings and roundabouts.


PostPosted: Fri Apr 30, 2010 1:11 pm    Post subject: Reply with quote
Commonly advocated methods of avoiding purple fringing include:
1)Avoid shooting with a wide-open lens in high contrast scenes.
2)Avoid overexposing highlights (e.g. specular reflections and bright sky behind dark objects).
3)Shoot with a Haze-2A or other strong UV-cut filter.
4)Post-processing to remove purple fringing (or chromatic aberration in general) usually involves scaling the fringed colour channel, or subtracting some of a scaled version of the blue channel.


PostPosted: Wed May 12, 2010 7:51 pm    Post subject: Reply with quote
in photoshop it easy to remove the ca...
you must do the following steps:
1) Duplicate the principal layer
2) Set to Colour the option of fusion (of this new layer)
3) Apply the Gaussian Blur filter with 2 or 3 pixel of radius. (3 is a good value) You must use a radius in order to delete the more big level of CA.
4) In the new layer, create a new Level Mask.
5) Use the Paint Bucket Tool (with black colour) on the Level Mask.
It is important that the filling colour of this Level Mask must become Black!
6) Use the Brush Tool (with white colour) on the CA to remove.

These steps are useful for the sensor bloom (purple fringing) artifacts...like those in the guinea fowl image.
I tried with 3 pixels of radius...and It works !


PostPosted: Thu May 13, 2010 2:34 pm    Post subject: Reply with quote
Orio wrote:
If this is the case, then the lens is innocent, and the sensor is the culprit.

If you still see the problem, it's CA, else, it's sensor bloom.
Another test you can make is to shoot a roll of slide film and compare.


Exactly Orio!
I have accused some of my lenses for showing CA (under nearly the same conditions), later on, when I switched to slides, I've never seen this "CA" again, so I guess it was sensor blooming.
I think it's to easy to say "CA", you have to compare it film, as Orio suggested.


PostPosted: Thu May 13, 2010 2:41 pm    Post subject: Reply with quote
Sensor blooming is always neutral in color. It can make the CA more visible, but it doesn't cause it. The secondary red/purple fringe is caused by sensor microlenses.

Different behaviour of film is caused by lack of microlenses and the fact, that common films aren't as sensitive to some wavelengths as digital sensors.


PostPosted: Thu May 13, 2010 3:14 pm    Post subject: Reply with quote
no-X wrote:
Sensor blooming is always neutral in color. It can make the CA more visible, but it doesn't cause it. The secondary red/purple fringe is caused by sensor microlenses.
Different behaviour of film is caused by lack of microlenses and the fact, that common films aren't as sensitive to some wavelengths as digital sensors.


I quite agree and I have to say, I verified personally, that in my 5D Mark II, lenses show more purple fringing than in the 5D.
One of the things that I always wanted to test, is to compare the same lens on 5D, 5DII, and film, for purple fringing.
I never found the time, and by the way, it would change nothing in the end, because I'd still be using these cameras and lenses... but it would satisfy a curiosity.
I am inclined to think that density of cells in a sensor may be directly related to how much purple fringing.
Post new topic   Reply to topic
Home   Manual Focus Lenses
Goto page Previous  1, 2
Page 2 of 2