Red Edge--Which lenses?

[photolandscape]

Advertisement (gone after registration)

Noah, I know you requested we not get off-track in discussing the "why" of the red edge - and just stick to a list of lenses that cause it.

 

Unfortunately, it does not seem to be linked only to specific lens optical designs, but also to either individual lenses or individual cameras (e.g. apparently Pascal gets red edges with a 35 f/2 ASPH and I don't. And I did not see red edges using my 21 on an M9 Sept. 9-10 - but I do get them with the identical lens on my early-production M9 received Sept. 17).

 

Which tends to indicate that it is not simply faulty corrections written into the firmware (which could be revised), but may be some kind of physical centering problem in specific lenses, or sensor placement within some bodies.

 

So there may be no simple list of lenses that do cause it or don't cause it.

 

That being said, certainly a list like this will reveal just that kind of inconsistency, which is good in its own way.

 

I'm beginning to think Andy is on to something--that there are widespread variations due to a whole range of factors, which makes this more problematic.

 

Please add the 25mm Skopar P to the list of red fringing lenses. I bought a brand new one, hoping for the best, and interestingly, it exhibits strong red fringing along the left hand edge of the picture frame, with more subtle red fringing on the righthand side.

 

I really need something for my M9 in the 21-25mm length, and I may have to take my M9 to a larger store with a large variety of lenses and try a whole range of possibilities. Having spent a small fortune on the M9, and having replaced my much-loved 28 Elmarit ASPH with a 35 Summicron ASPH, I'm hopeful that I can come up with a Zeiss or CV that will fill the bill. I use Cornerfix when I have to, but I'd really rather not have to.

[UliWer]

HI There

Is this the same thing? I think that this is just vignetting with a colour cast isn't it?

 

Yes, though it becomes more or less obvious with different settings of white balance. If you avoid too much of red tint you won't notice it in most cases. Though with tungsten lighting trying to remove it by adjusting white balance gives very greenish results, so CornerFix is the better general solution.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[adan]

Another data point for Noah's list - I had a chance to try a 24 Elmarit ASPH on my camera today - no red edges apparent on the LCD (haven't seen it big yet since I'm dealing with a HD failure on my main computer and won't have Photoshop running for a couple of hours).

[photolandscape]

Another data point for Noah's list - I had a chance to try a 24 Elmarit ASPH on my camera today - no red edges apparent on the LCD (haven't seen it big yet since I'm dealing with a HD failure on my main computer and won't have Photoshop running for a couple of hours).

 

Andy, thanks for the post. It gives me some hope that something will work.

[adan]

Sadly - upon further review in the computer, the 24 Elmarit I used (post #23) does show red edge(s) - but they were more symmetrical than what my 21 preASPH puts out. Not "red-left", but "red-all-around." I don't know why it was less visible on the camera LCD.

 

On blowing the image up I did notice that the "red" seems to be a preponderance of red noise, so the theory that it is vignetting correction gone awry (lightening the corners and increasing the noise) gains some credence.

[photolandscape]

Sadly - upon further review in the computer, the 24 Elmarit I used (post #23) does show red edge(s) - but they were more symmetrical than what my 21 preASPH puts out. Not "red-left", but "red-all-around." I don't know why it was less visible on the camera LCD.

 

On blowing the image up I did notice that the "red" seems to be a preponderance of red noise, so the theory that it is vignetting correction gone awry (lightening the corners and increasing the noise) gains some credence.

 

Andy, sorry to hear that. Do you feel it is likely that Leica will offer a cure of the red edge problem in the next version of its firmware? The camera has been on the market nearly 5 months, so I am hoping we'll be seeing a Firmware release soon, and hopefully one that will correct the problem with the 18-24mm focal length lenses.

[xtoph]

Advertisement (gone after registration)

i am sometimes seeing a definite red edge/corner problem on the left side using brand new 28 'cron. seems to be pronounced when the exposure is already strained--high iso low light levels--but is also easily perceptible on a medium iso shot of clouds.

 

so far, using 40/1.4 i don't see color shift, just heavy vignetting. i haven't tried to eliminate the vignetting in post yet; it may come out if i do.

 

btw, has anyone figured out what a good choice would be if i wanted to enter a lens code for the cv40/1.4 and let the camera reduce vignetting? any film folks out there want to hazard a guess on what character of leica lens corner shading it most looks like? is there a database already out there for best choices to code non-leica lenses on the m9? or is the consensus that it is better to turn lens code adjustments off until leica provides a fw update?

[noah_addis]

I've looked fairly hard and don't see any problems with my 35 Summicron Asph, 35 Summilux Asph or 28 Summicron.

 

Just the 21 is causing me problems.

 

I was talking to a repair guy at Leica NJ about my line problem and I mentioned the red edge. He claimed to know nothing about the problem:confused:

[adan]

Not too surprising that a line repair tech in the US might not know about red edges - lines have been showing up on M8s (and M9s) for three years, and are a definite service problem (requiring replacement in most cases), so I'd expect he's seen lots of those come through the door over 3+ years on their way to Solms.

 

The red edge thing is unique to the M9's bigger sensor, has only been evident for 5 months or less, and is not yet a "service" problem - i.e. I don't think anyone has sent in a camera to have that "repaired." Mostly because no one (yet - and including Leica, for all they've said on the matter) seems to know what servicing may be needed (FW improvements alone? Sensor replacement? Sensor positioning adjustment? Rebuilds of lenses that themselves aren't perfectly centered optically?)

 

xtoph - I'd start with one of the 35mm Summilux settings in the menus - pre-ASPH or ASPH.

[Nikkor AIS]

 

I tested a M9 in the local camera store and the 21 f4 Super Angulon was basically

 

unusable. Kind of bummed me out as it is so nice on film. The 15 Voigtlander was not bad and the 35 1.4 summilux was good as well as the Noctilux 50 1.0 .

 

It didn't matter what setting I used in the M9 profile to try to get rid of the problem, I recently tested a M8 and the results where just as disappointing with the 21 4 Super Angulon.

My theory is that lens whose back elements are close the sensor are not usable with the the Leica M8/M9.

Edited February 6, 2010 by Nikkor AIS

[Alnitak]

Here's what I have observed so far:

 

1) Voigtlander 12mm f/5.6: Medium red edges, left and bottom

2) Voigtlander 15mm f/4.5: Strong red edges, particlarly the left and bottom

3) Voigtlander 21mm f/4 Color Skopar: Strong left red edge, medium red bottom

4) Leica 21mm f/2.8 Elmarit-M ASPH, very mild red on the left and bottom

 

All of the other following lenses have not shown a red edge for me:

 

1) Voigltander 28mm f/2 Ultron

2) Voigtlander 28mm f/1.9 Aspherical Ultron

3) Voigtlander 28mm f/3.5 Color-Skopar

4) Voigtlander 35mm f/1.4 Nokton

5) Voigtlander 35mm f/2.5 P II Color Skopar

6) Leica Tri-Elmar-M 16-18-21mm f/4 "WATE"

7) Leica 28mm f/2 Summicron ASPH

8) Leica 35mm f/1.4 Summilux ASPH

 

I'm not listing lenses above 35mm because as far as I know, no one has reported this issue with lenses above 35mm (and even at 35mm it seems like there is only 1-2 reports). I've never seen it any of my Leica or Voigtlander lenses above 21mm.

 

Jeff

[frank_dernie]

I imagine it can happen with any lens. I just took a f22 shot of an OOF cloudy sky to check whether my sensor needs a clean (it does..) with 50mm f1.4 asph, and there is a reddish edge.

[okram]

This edge is always on the left, with all the lenses?

 

I have a question: does the lens show more vignette (no color shift just darkening) on the opposite side?

 

I suspect that the sensor has been decentered to the right (looking from the back of the camera) because of the battery...marknorton wrote about battery being in the way for ff sensor on the left side... maybe they just moved it a few milimeters to the right, pretending nobody will notice?

 

Can anybody have a look inside?

[jonoslack]

This edge is always on the left, with all the lenses?

 

I have a question: does the lens show more vignette (no color shift just darkening) on the opposite side?

 

I suspect that the sensor has been decentered to the right (looking from the back of the camera) because of the battery...marknorton wrote about battery being in the way for ff sensor on the left side... maybe they just moved it a few milimeters to the right, pretending nobody will notice?

 

Can anybody have a look inside?

 

HI Okram - we've been there and back again - the red varies in position from lens to lens, added to which Leica have said that they haven't de-centred the sensor, and with most lenses the red is also at the bottom (and with some it's at the top).

 

Scott thought that it might be light spillage at extreme angles, and that the asymmetrical RGBG on the bayer filter might cause the red shift on the left - I don't think that anyone has a definitive answer, but an off-centre sensor really doesn't fit the bill.

[jonoslack]

I imagine it can happen with any lens. I just took a f22 shot of an OOF cloudy sky to check whether my sensor needs a clean (it does..) with 50mm f1.4 asph, and there is a reddish edge.

 

Hi Frank - I think this must be something else. It happens with wide angle lenses which have a very steep angle of incidence of light, and where lens correction is applied to reduce vignetting - it's much more prevalent wide open than stopped down, and there is little or no vignetting on the 50 1.4 asph.

[frank_dernie]

I thought how easy it would be to do a quick check with a few lenses, so here we go.

Lenses:

50mm f1.4 asph, reddish tinge LH edge wide open, OK stopped down.

28mm Summicron, reddish tinge LH edge and bottom wide open, OK stopped down.

Tri-Elmar, OK all focal lengths wide open or stopped down

Wide angle tri-elmar, focal length corrected manually (rather than relying on the coding which does not know if it is at 16, 18 or 21mm) slight red LH and corners at 16mm and 18mm at f4, reducing stopped down. Slight red tinge 21 mm f4, OK stopped down at 21 mm.

WATE with auto lens recognition and using 16mm focal length there is a more reddish LH edge than when 16mm is manually set.

12mm f5.6 CV correction off, reddish left lower corner and pronounced vignetting. I did not set a correction since I have not found one which is better than off IMO.

In summary the reddish edge is always the left on my M9 and noticeable wide open even at 50mm.

My guess had been that the lens and sensor are not exactly centered on the M9, perhaps to fit in the M8 battery which, according to somebody on the forum a few months ago will not fit otherwise.

cheers,

Frank

[frank_dernie]

Hi Frank - I think this must be something else. It happens with wide angle lenses which have a very steep angle of incidence of light, and where lens correction is applied to reduce vignetting - it's much more prevalent wide open than stopped down, and there is little or no vignetting on the 50 1.4 asph.

 

My mistake Jono, see my other post. The red edge was at f1.4, not stopped down.

I was very surprised to see it on the 50 f1.4 and 28 f2, I had not noted it in normal photography, but did taking OOF pictures of a cloudy sky, hence the quick check in my other post.

[okram]

One thing is certain- the have to fix it in FW.

 

At what cost, I don't know. Sean Reid reports more noise in the corners...

 

 

 

P.S: Jono, thanks for the info.

[frank_dernie]

Hi Frank - I think this must be something else. It happens with wide angle lenses which have a very steep angle of incidence of light, and where lens correction is applied to reduce vignetting - it's much more prevalent wide open than stopped down, and there is little or no vignetting on the 50 1.4 asph.

 

I have also just checked the data and both the 50mm f1.4 asph and 50mm summicron have just over 2 stops vignetting wide open. The Noctilux around 3 stops, which is as I had remembered it.

I don't have data to hand for the CV lenses but seem to remember that the 12mm has less vignetting than the 15mm, and an exit pupil further from the focal plane, being a more retrofocus design.

 

FWIW,

Frank

[Advertisement]

Hallo,

Look here: Leica M

[russell]

I recently tested a M8 and the results where just as disappointing with the 21 4 Super Angulon.

My theory is that lens whose back elements are close the sensor are not usable with the the Leica M8/M9.

I shot the SA f4 on the M8 and it was ok-ish. If you set the meter right it was fine. Nothing corner fix could fix (assuming I set up the corner fix profile properly which I never got around to doing...)