M9 red edge question

[x-ray]

I'm getting close to purchasing a new M9 but have read mor about the red edge problem with wide angles.

The lenses that I will be using are the non asph 21 Elmarit, 25 ZM Biogon, 28mm CV f1.9, 35 asph Summicron, 35mm Nokton and 35mm Biogon, 50 asph Summilux and 50 Planar, 75 Summilux and 90 apo asph.

Question, how many of you have the red edge problem?

Tell me what you know about the red edge and is it easy to fix.

What I read no one seems to know why it happens. Some say selecting the correct lens setting in the menu corrects it but others say it doesn't. ( will not be coding lenses immediately ).

What is the percentage of people having the problem.

I'm not certain I want to spend $7k on a camera I have to work around problems constantly. I expect more from a $7k camera.

I appreciat any comments.

 

[jaapv]

I have no red edge problems with my Elmarit 28 and Summilux 24. I used to have some with the Biogon 21-2.8, but recoding it solved the problem. If you get it it is a 15 sec. fix in Photoshop, just run a grad layer over the red channel. Or use Cornerfix.
But I would strongly advise you to have the lenses coded.

 

[x-ray]

Is there a difference using the menu driven coding vs the lens coding?

I will be using this for some of my commercial work and often deal with a fair number of files. Each additional step at the computer can translate to additional hours.

 

[250swb]

No difference in the results between menu and lens coding, but I'm sure most people would forget to change the coding from time to time if doing it manually. So its good for the red edge and good for the Exif file to have the lens coded (which you can do yourself if you are at all practical).

Steve

 

[kbg32]

I have no problems with Leitz wide lenses 28, 21. Though lately, at high ISOs, 1250 upwards, I am noticing a bit of a red edge with my 28/2.8 Elmarit Asph. It is coded. My Zeiss 25/2.8 is awesome and does not produce the red edge. My Zeiss 21/4.5 Biogon produces the red edge, as do my CV wides, 28, 25, and 21. No problem with 50mm lenses or longer. My 35/2.5 CV is fine. My Konica Dual 21/35 is awesome as well. I do manually select the lens length in the camera's menu. I believe the problem is with focal lengths of 28 and wider.

 

[Nikkor AIS]

Taken with M9 and the 21 F4 1600 ISO

 

[jaapv]

Zeiss lens? Why should Leica correct for that? What did you code is as? Or is it the antique Leitz 21-4.0? That one is known for extreme light fall-off and even a hot spot up to f8.0 ( see Erwin Puts) so the result you show should not be surprising. 😉

 

[Ben Z]

I'm getting close to purchasing a new M9 but have read mor about the red edge problem with wide angles.

The lenses that I will be using are the non asph 21 Elmarit, 25 ZM Biogon, 28mm CV f1.9, 35 asph Summicron, 35mm Nokton and 35mm Biogon, 50 asph Summilux and 50 Planar, 75 Summilux and 90 apo asph.

Like cyan drift with IR filters on the M8, the degree to which red-edge shows varies with the composition. Shots with white or light-colored corners show it more. Dark shades in the corners tend to mask it.

Of those lenses you listed, I have the non-asph 21 Elmarit and the 28mm CV f1.9. Coded, the 21 Elmarit still shows quite a bit of red-edge even with the latest firmware. For shots where it shows I still need to use Cornerfix.

I had the 28 CV coded as a 28 Summicron with the M8, but that code is completely unsuitable for the M9 due to red-edge. Selecting the code for the non asph 21 Elmarit from the menu (or re-coding the lens, if you don't also own an M8) gets rid of red-edge at ISO 160, but it starts to creep back in as you up the ISO.

The latter is the biggest problem with coding. Red edge gets worse with increasing ISO due to the way cyan drift correction is computed in the firmware. (Shooting un-coded, red-edge doesn't change much). So if you code these lenses you need to either make a Cornerfix profile for each ISO increment, or else shoot un-coded (or turn off recognition) and then you can get away with a single Cornerfix profile for each lens.

Evidently, correcting for cyan drift of front IR filters is a lot less complicated than correcting for that of a sensor-mounted IR filter, because nearest-equivalent coding of CV lenses worked perfectly with the M8. It does not, on the M9. Leica was not pulling our legs when they said there were technical hurdles in making it full frame, and it would appear this one (red edge) was the most difficult, because it has yet to be completely eliminated.

In practice, however, it really isn't that big a deal to fix. I have to evaluate each of my Canon 5D shots as to the sharpening settings it will need, whereas I don't sharpen my M9 shots as a rule. So net-net, the occasional need for Cornerfix doesn't bother me much.

 

[JMQ]

The only red edge (or color shift) with my M9 is from the 21 f3.4 SA, similar to Greg's result of his 21 f4. This is due to the deeply protruding rear lens element. I have no issues with the 35 asph (pre FLE), tri elmar 28-35-50 v1, 50 nikkor, 50 pre-lux asph, and the various medium teles. None of my lenses are coded.

 

[x-ray]

Thanks for the information. I;ve come to the conclusion the M9 produces exceptional images but has a variety of issues, the main one that's a deal breaker for me is the red edge problem. I may have it or I may not but I'm not willing to put $7k into a camera that's not fully debugged. I expect more from a camera that costs this much. When and if they get things hashed out I may buy one but for the moment I'm going to buy more investments and grow the money.


I think at the moment I would rather double my money and buy a Hasselblad back. It's larger and heavier but the performance is better in line with my expectations. I'm not willing to spend additional hours to correct a problem that should'nt exist and doesn't exist in any other camera. I understand why it does but that doesn't make up for the need to do additional corrections.

Thanks all for the information. Some day down the road let's hope the M10 is free of such issues.

 

[akk2]

21SA , 21/4.5ZM = lot of red edge.

 

[x-ray]

I hate that Leica can't seem to get the digital going. I'm taking the wait and see approach for now. My reason was bringing the weight of my gear down when I'm out on assignments. Weight is one thing I can continue to deal with but quality is most important.

 

[peter_n]

... I'm not willing to put $7k into a camera that's not fully debugged. I expect more from a camera that costs this much.

I'm not willing to spend additional hours to correct a problem that should'nt exist and doesn't exist in any other camera. I understand why it does but that doesn't make up for the need to do additional corrections.

I hate that Leica can't seem to get the digital going. I'm taking the wait and see approach for now.

I totally agree. It really pisses me off that Leica have been unable to produce a reliable digital M. The M8 and M9 are completely unacceptable products for what they cost. I'm reduced to using a dSLR for color, but the camera is bulletproof and cost 10% of an M9. I don't think Leica will ever be able to produce a decent digital M and I've been buying Nikkor lenses back.

 

[x-ray]

It seem safe to say lenses from 35mm up to 90 are fine but there have been a few limited issues with 35mm's. 35-90 just seems too limited for any practical commercial application. Why spend $7k on something you can't use your 15-28mm lenses on. Again if I have to go to another software and spend a minute or so fixing red edges over the corse of a day when I process 100 or more images this adds considerably more time. I've added 1 to 2 hours of more work each day. I would guess the perfect M9 file is a sharper file than a Canon 1DsIII or 5DII file if viewed at 100% but how often do we reproduce that large. Consider how large a print would be relative to viewing at 100%. Pixel peeping can give a distorted view of the realities of practical photography. We rarely make 100 inch prints.

A new 5DII would be a winner for me if lenses were smaller and lighter. On a normal assignment I carry a 1DSII and another body as a backup. In the case is a 15, 16-35, 35 1.4, 50 1.4, 24-70, 85 1.2, 135 2 and 70-200 2.8 plus a 24 TSE and sometimes 90 TSE and three extra batteries. This is heavy! The drawback to the 5DII is lense size and weight. I might shave off a pound or so and that's about it. I think the solution is to reduce the load by taking only what I think I will need and no more. I rarely use the zooms so that can stay back in the studio but I'm still around 35 pounds of cameras. The M would have reduced this substantially but that's life.

 

[Alnitak]

I think this is getting blown out of proportion. Leica has done an excellent job dealing with the red edge issue with their own lenses. We simply can't expect them to deal with lenses from Zeiss and Voigtlander.

This is a design issue: The angle of incidence of light hitting a digital sensor will introduce problems. A sensor is much, much different than film in this sense. Leica has been saying this for years, and they know what they are talking about. Leica's in-camera profiles nicely correct for this issue, but Leica has carefully created these profiles to match their lenses. If a Zeiss or Voigtlander lens doesn't match, such is life. It's not Leica's problem.

As for work, its a very simple matter to create a batch process to correct images with Cornerfix. I regularly do this to correct large batches of images and it adds a few seconds per image. Not a big deal--and I only need to use it for my 15mm and 12mm Voigtlander lenses.

Jeff

 

[sig]

So, I guess putting a full frame sensor in a M body is pushing the limits of the sensor and computing power available today, and Leica has probably done as much as they can on this issue. But if I had paid 7000 dollars for the M9 that would not be a good enough excuse for me..... (not a problem for Leica, I would never spend that money on a camera)

 

[photogdave]

I think at the moment I would rather double my money and buy a Hasselblad back. It's larger and heavier but the performance is better in line with my expectations.

You should look at the new Pentax 645D. Outstanding IQ and superior handling/ergonomics to any other MF system, save the Leica S2. It's weather sealed, has fast accurate AF, an awesome viewfinder and is surprisingly light.

 

[jaapv]

I've heard this before, and sadly, one of the primary reasons I would consider a M9 is to make full use of my 21/3.4 SA. I love this lens for film, any film. The crop factor removes too much of its charm/character on the M8.2, I would just as soon use the 21/2.8 Biogon, not a bad lens in its own right.

The 3.4 SA will never make the grade on a digital camera, I fear. The weak points of that lens will get shown up mercilessly. Extreme vignetting, very soft corners, you name it. Stick to film for that lens.

 

[jaapv]

Happily Geoff on LUF has provided the final answer for this question. I hope he doesn't mind my reposting it here:

I wrote to Customer Service in Solms once again after these reports and received this information back. As when the broken cover glass reports emerged, my question was referred directly to Stefan Daniel who responded very promptly and openly. On this red edge question, technical staff provided this information to me. I have quoted the response verbatim. I would add my personal comment that I have never experienced any problem personally (from about 2,000 frames with my M9). I posted my examples from all of my lenses from 35 and wider.



...............................................................................................................

"Thank you very much for your information.

We have had some complaints about the red edges, you are right. I will give you the full technical explanation for that.

On the one side, we do have a very short distance from the lens to the sensor (for this reason the m-system is so compact), on the other side the pixels of the ccd sensor are not symmetric. This combination produces a totally colored and inhomogenious RAW image. To provide a plain and neutral image, we do a lot of complex compensation, sensor corrections and lens vigentting corrections. What we achieve is a nearly neutral image, but unfortunately just "nearly" neutral. The tolerances for the M9 are even tighter as for the M8 with IR-filter, but unfortunately there are some very slight deviations towards red, the obviously are not acceptable for some special applications.

We did an improvement in a previous update, using the vignetting correction, this improved the effect visibly. But unfortunately we had to recognize that even this was not enough. There are still some complaints, therefore we decided to apply additional correction algorithms for the critical lenses. But developing these need unfortunately some time, because some sequences in the internal data management have to be changed, this is always a difficult and risky modification. Of course correcting this on a PC is an easy thing. But letting the camera do this needs a very simple algorithm, because we would not accept any delay in DNG mode.

We are planning to offer this update as soon as possible, but it will need at least until spring 2011.

I hope this answer helps you understand the confusion about the red edges."

...................................................................