M8 (2.005) + Coded 35 'Cron (IV) + UV/IR Filter = Magenta Sides?

[Jark89]

Advertisement (gone after registration)

Hi all, I've had my Leica M8 for almost a year now, and I absolutely love it to death. Before I purchased it, I read all about its quirks...or so I thought.

 

Anyway, I'm using the latest firmware with a Leica 35mm Summicron pre-ASPH (IV). I coded it as the same lens, and photos shows the "35mm" EXIF data. I also have a genuine B+W UV/IR filter on it.

 

Only recently have I just noticed some "reddish" (magenta?) cast towards the left & right "sides". Is this to be expected? I don't recall reading about this, is this some sort of "over correction" for the cyan corners? Any insight would be greatly appreciated!

 

Here's a recent picture that shows the issue, more easily seen on the left side.

 

[wda]

It is worth double-checking the menu. Sometimes I find inadvertent changes switch off camera recognition of the coding. Not often, but the symptoms suggest that cause.

[WestMichigan]

Ola,

you might have already ruled out what I'm about to suggest, but couldn't the 'magenta' in the image you shared be from multiple light sources in the scene with different color temperatures? Possibly the surface of the 'table' picking up something off camera and faintly reflecting it back to the lens? Recall, with a Rangefinder, reflections can line up differently for the lens than for the viewfinder and you won't know it until you see the Film or the File from the shot. Ah, one more of the bountiful joys of Parallax Error. :-).

Personally, I really lean towards it being a stray reflections problem. Then again, maybe you've got plenty of examples of the magenta problem in different settings and all I'm doing is pushing great theories!

 

Sincerely

Richard in Michigan

[Jark89]

Thanks for the suggestions!

 

wda: Yes I double checked the menu and its set to Coding + UV/IR. :/

 

WestMichigan: You guessed right in that I do have other samples that show the issue, these pictures contained gray overcast skies, and the reddish sides show up slightly as well.

[chris_livsey]

Are you aware that the specification of the Leica brand filter is different to "other" brands ?

 

The software is adjusted to match the Leica filter.

 

IR cut filters for use on the Leica M8 | photostream

 

This thread discusses the issue at more length. It appears that wider then 35mm are more affected. It could also be that your sample (filter) is at the far end of the QC limit and is thus showing the problem more.

 

http://www.l-camera-forum.com/leica-forum/leica-m8-forum/27330-b-w-486-filter-official-schneider.html#post285665

 

Could you not borrow another filter (Leica) to test against ?

[Jark89]

Chris, thanks for that info! I've never come upon it before. However, from reading the thread, it seems that only lenses wider than 28mm are affected more? Not sure though. A bit ironic, because I actually sold my Leica filters, and replaced them with B+Ws since not only did I read they were of higher quality, but also from my own unscientific testing which showed weaker ghosting with the B+W. However, I never took into account to check for color discrepancies between the two.

 

Your thought of maybe I got a bad quality filter is plausible though, unfortunately I am well past the return/exchange date with B&H. None of my friends use Leica unfortunately...Oh well.

 

Again, thanks for the info.

 

Still open to other thoughts too!

[chris_livsey]

Advertisement (gone after registration)

Can't be that much to buy a Leica filter, test it then sell whichever looses. I mean buy S/H, all the dealers have plenty since the M9rs dumped them, inadvisably it appears for some work

 

You should be able to re-sell for a modest loss if at all.

 

Do a google search there was a lot of info in the early days on filters. I mean really early before Leica brought them out then later when the software was changed to accommodate them.

 

BTW my V4 Summicron shows no problems, Leica filter though.

 

Looking at that shot I wonder if the reflections fooled the sensor that gives the software an estimated F stop. The corrections applied vary with lens, f stop and colour temp. If you can remember what you actually used it would be interesting to compare with what the metadata says was estimated. It could be the other shots fall into the same category, the wrong correction applied because of a bad guess.

 

Cornerfix will do this.

 

CornerFix | Download CornerFix software for free at SourceForge.net

[Jark89]

Hi Chris, regarding the above test shot, I set the aperture at f/2 with ISO 640, the EXIF reports f/3.4.

[chris_livsey]

Not a bad guess

 

Can't see that doing it then on its own.

My money is on a set of circumstances conspiring. Odd/unusual colour temp, bad f stop guess with filter not Leica and/or/maybe on edge of spec.

 

You could try provoking it but I would check out the filter issue first good luck.

[ho_co]

I doubt that this is a filter problem, since the UV/IR-Cut filters are extinction filters, and the ray incidence angle isn't shallow enough to cause the problem.

 

There's a similar thread running from a new purchaser of the current 35/2 as well, but on M9.

 

I'm surprised to see the problem with a 35 mm on either camera, and would recommend checking with Leica Technical or Leica Customer Service.

[Guest V64]

Somewhere I recall reading about magenta colored flare (I think flare is normally seen as an uncolored loss of contrast) on some Leica lenses.

 

These two shots were taken when testing a new hood on a 28 cron. ASPH coded

 

I think you can see how a small rotation away from the light source (top right) has had a significant impact on the image. Leica UV/IR filter used.

 

My speculation is that this is due to a combination of the filter and the flat/concave lens surface on the first element. I have never seen this effect when using a Zeiss 25mm. or with an early Elmar 50mm f2.8, where the flare is never colored.

 

Would anyone care to explore this theory?

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!

[Jark89]

Chris, thanks for all the suggestions. I'm going to see if I can find a free Leica filter to test out, but I don't think I'm going to go the route of buying then selling, just too much of a hassle.

 

Howard, I'm not sure if my issue is persistent all the time, as I only noticed it on some shots beginning with a trip back in August. Looking at all the pictures taken before that date, I can't seem to find a single shot that contains these reddish sides...even shots that seem to have been taken in conditions which would show it - ie. shooting grayish overcast sky/clouds. Weird.

 

V64, thanks for the suggestion. I see flare in your first shot. I am however pretty familiar with the flare signature of my lens, and I don't think it is that. Also with shots that contain the issue, the light source is usually behind my back.

[Guest V64]

Jark89

 

Yes there is 'classic flare' top right - I should have emphasized my point, which is the magenta color cast in the gray stone wall and the grass, this like the magenta cast in your example, although it goes across the frame i.e. it is not 'at the sides'.

[WestMichigan]

Ola - Me Again,

not sure why it didn't occur to me before, but another cause might be that the UV/IR Filter in question is Defective. The 'obvious' test is to find a 'subject' which will exhibit the IR Pollution Color Defect (Testing types of Black Cloth might serve you best for finding an appropriate 'static' test subject.) Then without the UV/IR Filter shoot a 'frame filling' image which exhibits oodles of the 'Magenta' problems caused by the sensors design then shoot it again with the filter on. If the filter isn't blocking IR consistently across the entire frame it should be instantly apparent using this methodology.

BTW I see zero evidence of an IR pollution issue in the two outdoors images you shared, but I'm not an expert on the IR problem. Still for whatever it may be worth, on my monitor anyway, the color balance and color distribution looks consistent and even across the entirety of both frames. Except for the flared section, of course :-).

Richard

[leicashot]

Somewhere I recall reading about magenta colored flare (I think flare is normally seen as an uncolored loss of contrast) on some Leica lenses.

 

These two shots were taken when testing a new hood on a 28 cron. ASPH coded

 

I think you can see how a small rotation away from the light source (top right) has had a significant impact on the image. Leica UV/IR filter used.

 

My speculation is that this is due to a combination of the filter and the flat/concave lens surface on the first element. I have never seen this effect when using a Zeiss 25mm. or with an early Elmar 50mm f2.8, where the flare is never colored.

 

Would anyone care to explore this theory?

 

You've also moved the lens away from the sun. That may just have something to do with it too ;-)

[Jark89]

Ola - Me Again,

not sure why it didn't occur to me before, but another cause might be that the UV/IR Filter in question is Defective. The 'obvious' test is to find a 'subject' which will exhibit the IR Pollution Color Defect (Testing types of Black Cloth might serve you best for finding an appropriate 'static' test subject.) Then without the UV/IR Filter shoot a 'frame filling' image which exhibits oodles of the 'Magenta' problems caused by the sensors design then shoot it again with the filter on. If the filter isn't blocking IR consistently across the entire frame it should be instantly apparent using this methodology.

BTW I see zero evidence of an IR pollution issue in the two outdoors images you shared, but I'm not an expert on the IR problem. Still for whatever it may be worth, on my monitor anyway, the color balance and color distribution looks consistent and even across the entirety of both frames. Except for the flared section, of course :-).

Richard

 

Hi WestMichigan, thanks for the suggestion! However, I've never had any issues with the "blacks turning magenta" problem. So I think my UV/IR filter is working OK from that perspective at least.

 

P.S. The above 2 outdoors photos were not posted by me, so I'm not sure if you're replying to me or the other poster.

[WestMichigan]

Hi Jark89 -

I didn't catch that the 'outdoor' shots weren't from you. Reading posts on the forum at 2am isn't entirely conducive to clarity! Mia Culpa - Ah Well.

I also didn't catch that the magenta 'contamination' was also on the right in your posted image. I'd even pondered the photo trying to think of reasons for the magenta area. Guess I'm getting a head start on Senior Moments - I'm only 41! Practice makes perfect I guess. :-)

It just occurred to me that there might have been a failure somewhere in your imaging chain software. ie: raw converter, raw converter settings, editing program issue, or editing program settings. As in the DNG had the correct tags to fix magenta corners, but it didn't get done by the programs for some reason.

I still am puzzled because 35mm isn't all that wide and my 'understanding' is that the problem correctly coding a lens 'fixes' is progressively worse the wider the lens and the 'extent' of magenta showing doesn't quite look right compared to images I've seen from non-coded wide angle images.

I'm going to be a M8 owner and your inquiry has piqued my curiousity about what's going on with your image. Hope to find out what the answer ultimately is.

Richard

[Jark89]

WestMichigan, you just hit upon something I haven't thought of yet, the RAW converter. I think I will download a trial version of Capture One to test. I'm currently using ACR. Thanks.

[Jark89]

OK I took another shot trying to mimic the original problem shot, and I couldn't get it to come out as reddish, its just very very faintly there. Anyway, I tested this second shot in both ACR and trial version of Capture One Pro 5.2, and its basically the same. So I guess its not a RAW converter problem. :S

[Advertisement]

Hallo,

Look here: Leica M

[Jark89]

Update: Well I decided to "un-code" my lens, and there is no more reddish tint. I also contacted Leica NJ, and they said its not normal and I should send my camera in for inspection. I'm not too keen on sending my camera in though, so I think I'll just keep it this way.