Jump to content

CL cyan drift with wide-angle lenses?


Recommended Posts

A question for those using the CL: Do you find excessive cyan drift in image corners when using lenses as wide as or wider than 35mm? Tom

 

 

No miscolouring seen on my side, based on 15mm Voigtländer M version III, 18mm TL, 23mm TL, 21mm Super-Elmar-M, 28mm Cron-M (latest), and some 35mm M-lenses (Leica, Contax, Zeiss). Based on the characteristics of these lenses on Leica FF sensors (M240/M10/SL), nothing should be seen either... Other lenses may possibly show miscolouring - among those Zeiss 21mm Biogon f4.5 and Zeiss Contax 16mm Hologon f8 - but for less exotic glass, one should be pretty much safe.

 

("Cyan drift": I guess this is the Italian flag phenomenon?)

Edited by helged
  • Like 2
Link to post
Share on other sites

"Cyan Drift" is a phrase invented by Sean Reid.  It is meaningless, so please don't use it.  The phenomena seen in the M8 and very rarely in the M240 is actually a vignetting of the red part of the image.  It's is fairly symmetric.  The Italian flag problem was seen on the M9, and is not symmetric, but has a loss of red on one side and excess red on the other.

  • Like 4
Link to post
Share on other sites

[...] Other lenses may possibly show miscolouring - among those Zeiss 21mm Biogon f4.5 and Zeiss Contax 16mm Hologon f8 - but for less exotic glass, one should be pretty much safe. [...]

 

Sadly, one can not use the Contax G Hologon 8/16 on the Leica crop bodies. It would have made such an amazing 24mm, too.

Link to post
Share on other sites

Advertisement (gone after registration)

I have no information on the CL - depends on whether it reads M lens coding and makes similar corrections.

 

I won't repeat scott kirkpatrick's correct information, just add the following. Both the Italian Flag effect and cyan drift/vignetting are caused by 1) short-focus (i.e. mirrorless, close to the sensor) wide-angle lenses, with 2) no coding or no corrections/lens type selected from the lens menu. But through different mechanisms, so they are not quite the same thing - and look different.

 

Cyan vignetting** is purely due to filtering to remove infrared light contamination, with a wide angle lens, and no coding/corrections. A wide-angle "sees through" a filter at strongly different angles at the center of the picture - and at the edges. This makes the filter effectively stronger and more heavily tinted around the edge of the picture, as the lens looks through more glass, "slantwise." See the diagram below. The bottom line is that the IR filters cut into visible red light a bit, as well as infrared light, thus producing a visible color shift.

 

Cyan vignetting can occur with either a cropped-sensor camera (M8) or a FF sensor (M9). In the case of the M8, with the external IR filters required for clean color, and in the case of the M9, due to the thicker internal filtering on the sensor itself (0.8mm in the M9 vs. 0.5mm in the M8).

 

If no external filter was used on the M8, no cyan vignetting occurred - even with an uncoded 15mm Voigtlander lens, let alone a 28mm.

 

Italian Flag syndrome is cause by - something - in the sensor (silicon) architecture itself, or perhaps the microlenses on the surface - not the filter glass, although with an uncoded WIDE lens, both may be present, emphasizing the red/green shading of IFS.

 

In the attached image, you can see cyan vignetting in both M8 and M9 images, taken with uncoded, semi-wide lenses of similar fields of view on the two sensors. A green/cyan tint overlaid on the water or paving progressively stronger at the picture ends. With a diagram showing how a wide-angle lens will "see more filtering" looking through the flat internal or external filters at a slant angle near the edges than straight through the glass at the center of the scene. A basic question of geometry.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

________

**Scott is correct that what gets blocked in "cyan drift/vignetting" is red light - vignetting of red. But since the visible effect is so clearly a "cyan" stain, I still find that more descriptive and recognizable, if not technically accurate. If we named Italian Flag syndrome after its missing colors (the inverse), we'd have to call it "Ivory Coast Flag syndrome" instead. ;)

Edited by adan
  • Like 8
Link to post
Share on other sites

 

 

Italian Flag syndrome is cause by - something - in the sensor (silicon) architecture itself, or perhaps the microlenses on the surface - not the filter glass, although with an uncoded WIDE lens, both may be present, emphasizing the red/green shading of IFS.

 

In the attached image, you can see cyan vignetting in both M8 and M9 images, taken with uncoded, semi-wide lenses of similar fields of view on the two sensors. A green/cyan tint overlaid on the water or paving progressively stronger at the picture ends. With a diagram showing how a wide-angle lens will "see more filtering" looking through the flat internal or external filters at a slant angle near the edges than straight through the glass at the center of the scene. A basic question of geometry.

 

attachicon.gifcyandrift.jpg

________

**Scott is correct that what gets blocked in "cyan drift/vignetting" is red light - vignetting of red. But since the visible effect is so clearly a "cyan" stain, I still find that more descriptive and recognizable, if not technically accurate. If we named Italian Flag syndrome after its missing colors (the inverse), we'd have to call it "Ivory Coast Flag syndrome" instead. ;)

 

'Italian Flag' is probably caused by the asymmetry of the colour filters on each square (ie. 4 pixels) of the bayer grid  ........ which lie in wells on the sensor. As rays become more oblique at the periphery, more of the blue in each bayer square on one side of the sensor is occluded......  and red on the other..... leading to progressive colour casts the further out you go. I proposed this as the only logical explanation years ago ..... and Leica did confirm this somewhere in discussions about the changes in the M240 sensor ....... which has specially designed wells which were much shallower ..... 

Edited by thighslapper
Link to post
Share on other sites

Italian flag's assymetry in the M9 may have its roots in a decision (about which I have no facts, only a suspicion) to fit a full-frame sensor into the M9 by going a little bit off-center in its position.  Sandy McGuffog, in writing CornerFix to handle this, observed that algorithms that assumed everything was centered and symmetric simply did not work.

  • Like 1
Link to post
Share on other sites

Italian flag's assymetry in the M9 may have its roots in a decision (about which I have no facts, only a suspicion) to fit a full-frame sensor into the M9 by going a little bit off-center in its position.  Sandy McGuffog, in writing CornerFix to handle this, observed that algorithms that assumed everything was centered and symmetric simply did not work.

 

Bizarre ..... I find that hard to believe ..... I think the reasons normal algorithms didn't work was because it was not a normal situation ......  :rolleyes:

 

The only asymmetry in the whole optical pathway/system is the arrangement of the colour filters in the bayer array, so I am convinced this is the cause ...... and the reason all subsequent M series sensors are custom made to reduce the problem. 

Link to post
Share on other sites

'Italian Flag' is probably caused by the asymmetry of the colour filters on each square (ie. 4 pixels) of the bayer grid  ........ which lie in wells on the sensor. As rays become more oblique at the periphery, more of the blue in each bayer square on one side of the sensor is occluded......  and red on the other..... leading to progressive colour casts the further out you go. I proposed this as the only logical explanation years ago ..... and Leica did confirm this somewhere in discussions about the changes in the M240 sensor ....... which has specially designed wells which were much shallower ..... 

In that case, why don't we see this on other Bayer-filtered cameras? The Italian Flag aberration can be corrected by slightly off-centering the lens (for instance by sticking a bit of Sellotape to the flange.

Link to post
Share on other sites

In that case, why don't we see this on other Bayer-filtered cameras? The Italian Flag aberration can be corrected by slightly off-centering the lens (for instance by sticking a bit of Sellotape to the flange.

 

A fair point - but what "other Bayer-filtered cameras" are you comparing to, that also replicate all the other unique characteristics of the Leica M digitals, particularly the M9?

 

- short lens register/back-focus (< 40mm) - eliminates all mirrored SLRs

- no AA filter

- thin (0.8mm) cover-glass/IR filter - eliminates Sony Alphas (1.5mm)

- no "designed for digital" telecentric wide-angles

- full-frame (since IFS did not regularly occur with the cropped M8) - eliminates Fuji and Olympus bodies

 

I note that Hasselblad warned about using the non-retrofocus "not optimized for digital capture" Super-Angulon-like 38mm Biogon/SWC on their digital CFV backs with the larger, non-square sensor (no link available, CFV discontinued - again :( ). And Phase One has a special module in Capture One exactly for correcting asymmetrical color casts with Phase One backs.

 

https://www.phaseone.com/en/search/article.aspx?articleid=2047&languageid=1

 

I tend to go with the simple asymmetry of the Bayer pattern as the root cause - combined with those other unique factors of the Leicas. No need to drag in theories about Leica mounting the sensors off-center - they are inherently "off-center" at the pixel level. But since no one at Leica (or elsewhere in the industry) has ever really made a definitive statement as to the technical cause(s), we are (still) left to make educated guesses.

__________________

 

@ thighslapper - the M9 sensors (and the M8s) were custom-made. Bespoke for Leica M (microlens radial offsets, thin IR filters, etc.) and never used in any other camera. Part of the reason it is such a pain (4 months now and still waiting) to get the M9/MM sensors replaced for corrosion. The CMOS sensors just have (perhaps) additional custom work in the silicon (shallow pixel wells).

Edited by adan
  • Like 2
Link to post
Share on other sites

 errr .... I think that is just a very long winded version of what I said ......  :huh:

 

my original reply ...... :

 

https://www.l-camera-forum.com/topic/153464-towards-an-explanation-of-the-italian-flag-phenomenon/page-11?do=findComment&comment=1759579

 

anyway ...... hopefully it is a historical oddity that will never again be a significant issue in current and future Leica cameras 

Edited by thighslapper
Link to post
Share on other sites

errr .... I think that is just a very long winded version of what I said ......  :huh:

I think you said it was caused by the arrangement of the CFA while the explanation I linked to appears to be saying that the shapes of the sensor pits differ.

  • Like 1
Link to post
Share on other sites

Seeing as there seems to be no consensus as to any issue here, I'm going to continue happy in the knowledge that there's no problem worth worrying about.

Rewinding to the core intention (I believe) of the thread:

 

Please chime in if there are lenses that show edge/corner miscoloring on the CL.

Link to post
Share on other sites

So far, no obvious issues with:

- WATE 16 to 21 tri-Elmar

- 21mm Summilux

- 21mm Super Elmar

- 35mm Summilux FLE

 

I don’t happen to own any 24mm or 28mm M mount Lenses—traded my 28mm for a Q a couple years ago.

 

All the above were connected with the T Adapter M and all brought up the appropriate profile automatically. I don’t own any non-Leica M mount lenses, so can’t provide feedback on CV or Zeiss wide angles. Oh, and all of the above appeared to have at least reasonable corner sharpness, as one would expect with an APS-C sized sensor.

  • Like 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...