Red Edge--Which lenses?

[C_R]

Advertisement (gone after registration)

...May I suggest that you check some RGB readings around some of the images? Exact values aren't that important, but the difference in each of the three colors from that at the center.

....

 

 

Howard, excellent suggestion. I find RGB values not very helpful for this issue - why not use the PS LAB color info ? PS CS shows RGB as well as LAB values in the Navigator window (chose the "Info" tabulator, and look @ the "a" and "b" values). A positive "a" value (a value greater than 0) indicates magenta. A negative "a" value is green, a positive "b" value is yellow and a negative "b" value is blue. The higher the values, the more saturation.

An "a" value of 2, for example is a very slight magenta. With a "b" value not equal zero it is a violet (b<0) or red (b>0)

 

 

A zero value for "a" and "b" is perfectly neutral, "L" is the grey value then.

 

Peter´s 1.4 Lux examples here show a slight magenta cast LHS, and a slight cyan-blue cast in the center, accentuated towards the corners. The a/b readings are, for example:

 

f:1.4 in the UL corner violet-blue (2/-5), LL corner violet(ish) blue with 0 to -1/-4, center slight cyan (-2/-2), UR and LR corner cyan(ish) blue with -1/-5 to 0/-6

The blue values are ok, as the excellent Elmarit example shows. Probably the lighting and reflection of the sky.

All very slight magenta casts, but visible. In a snow scene, I would try to fix it.

 

f:5.6 and f:11 is better, UL corner 0 to 1 as "a" value, -3 as "b" value, less intense casts as with f:1.4 (similar for LL to LR corners)

 

The 24 Elmarit has almost no magenta, only a cyan(ish) blue for all apertures, this may be due to the lighting conditions here.

Leica has done a very good job with the firmware here

 

LAB values are very useful for evaluation of colors, independent of your monitor and vision.

[Advertisement]

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

[noah_addis]

I shot some tests today with my 21 Elmarit Asph and 28 Summicron.

 

The 28 didn't exhibit any serious flaws. There was some amount of measurable color cast but it was barely noticeable.

 

The 21 showed pretty much the same at all apertures tested: f/2.8, 5.6 and 11.

 

The results from my two M9s were extremely similar. All were quick conversions with ACR, as shot WB.

 

Here are the f/5.6 samples:

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!

[innerimager]

wow Noah, now I see what all the fuss is about!

[Alnitak]

Noah,

 

That looks about like my 21mm Elmarit-M ASPH. I am hopeful that it will be corrected with a FW update, as it seems pretty consistent on that lens across different copies of the body and lenses. Depending on what you are shooting, the effect can be invisible, but obviously not when shooting snow.

 

Jeff

[noah_addis]

wow Noah, now I see what all the fuss is about!

 

Yeah, just about the time I (a) started liking the 21mm focal length and ( started working in color again:mad:.

[Michaelberlin]

I have the red-egde problem with the M9 and the elmarit M 1/2,8 28mm and snow!

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!

[noah_addis]

Advertisement (gone after registration)

Noah,

 

That looks about like my 21mm Elmarit-M ASPH. I am hopeful that it will be corrected with a FW update, as it seems pretty consistent on that lens across different copies of the body and lenses. Depending on what you are shooting, the effect can be invisible, but obviously not when shooting snow.

 

Jeff

 

Glad to hear my copy isn't abnormally bad, I guess. It may be masked by some subjects and lighting conditions, but I often shoot in sort of flat overcast light where the problem is quite obvious, especially in the sky but also in other areas.

[Alnitak]

Yeah, just about the time I (a) started liking the 21mm focal length and ( started working in color again:mad:.

 

I just bought mine last week, so its very annoying. There were some reports from folks who said their copy did not have the red edge effect, but I suspect that they just aren't noticing it.

 

The solution at 21mm is either the WATE--which is slow at f/4--or the 21 'Lux, which is fast, but large and heavy and very expensive. I bought the 21/2.8 precisely as a compromise in speed, weight and price--so there had better be a fix for the issue forthcoming.

 

Jeff

[innerimager]

Yeah, just about the time I (a) started liking the 21mm focal length and ( started working in color again:mad:.

 

and here I am seriously considering committing to a year of B&W only. Who said life is fair? No one I guess......Peter

[ho_co]

... I find RGB values not very helpful for this issue - why not use the PS LAB color info ? ...

 

C_R--

Fantastic! Lab and I haven't met yet, though I have seen her around. Your introduction makes good sense.

 

Thanks!

[noah_addis]

I have the red-egde problem with the M9 and the elmarit M 1/2,8 28mm and snow!

 

Is this the latest ASPH version?

[poe-bear]

I have a leica m9 + summicron 35 mm. I Have also a soft red edge in my pictures.

 

Have tried cornerfix. It works fine, but i will NOT use it anymore. I've come to the conclusion that cornerfix changes ALL colors in the pictures. Not only the red edge. I've not paid 7500 euro to have the perfect picture, and then overlay it with a cyan colorcast.

 

my advice: use the graduated filter tool in lightroom. I've did 4 filters on one picture. 3 on the left side/ 1 on the top of the image. (180-200 degrees / 2 to 4 percent) One you have done that, save it with a preset.

 

It works!

[thrice]

I have a leica m9 + summicron 35 mm. I Have also a soft red edge in my pictures.

 

Have tried cornerfix. It works fine, but i will NOT use it anymore. I've come to the conclusion that cornerfix changes ALL colors in the pictures. Not only the red edge. I've not paid 7500 euro to have the perfect picture, and then overlay it with a cyan colorcast.

 

my advice: use the graduated filter tool in lightroom. I've did 4 filters on one picture. 3 on the left side/ 1 on the top of the image. (180-200 degrees / 2 to 4 percent) One you have done that, save it with a preset.

 

It works!

 

Thanks for posting that... in every thread about the red edge issue.

 

Cornerfix does nothing to the other colours if profiled properly. If you're getting a cyan cast I am assuming the WB on your M9 wasn't set correctly before creating your profile, or you're profiling from a cream/off-white wall.

 

The perfect picture isn't determined by the expense of the camera, it is the skill of the photographer with the tools at his/her disposal. But the M9 gives very high resolution files and you have a vast array of excellent lenses available.

[Michaelberlin]

Is this the latest ASPH version?

 

Yes, is is a brand new lens.

In the meantime i got a message from the Leica shop that there will be a softwareupdate for the M9 which should solve the problem. Leica knows this problem already and it seems to happen with every lens >35mm.

[Jerry_R]

I haven't observed red corners with new 35 Summarit... Maybe I need to make a photo of white wal.

 

Will try tonight, especially that I bought 15mm today.

[Englander]

This thread appalls me. It along with the threads on maladjusted rangefinders on brand new M9s should be made stickies! They should be sticky and appear at the top with a title "What Leica should fix."

 

After having gone through the same charade with the M8's inadequate IR filtration--Leica's denying there was a problem, suggesting that the problem was infrequent, then making the problem the photographer's (you shouldn't take pictures that require more IR filtration), finally admitting there was a problem, offering a slow fix for those who owned only three lenses and an expensive fix for those who owned more--I feel it was the outcry on this forum that finally persuaded the small slow moving company to fix that problem.

 

In many ways the current IR problem might be worse because, while it is there all the time, it isn't noticeable with all lenses all the time. But it is still a serious problem for any photographer who takes an important picture with light tones at the edges of the image. All the proposed post-processing solutions echo the post-processing and profile solutions offered by forum members for the M8 and ultimately indicate the M9 cannot be trusted to operate correctly in many normal photographic situations and is a camera for owners who want to play with their images after the fact. Perhaps there was an inkling of warning that such was the market Leica was seeking for the M9 by their selection of "testers" that we know about who were mainly physicians, surgeons and salesmen (not that such people can't be fine photographers, but rather by having more people who take pictures exclusively test the camera, the problem might have been highlighted before the gimicky 9.9.09 release). When one of the few photojournalists on this forum (Noah Addis) uses the camera--guess what? He notices there's a problem. A problem that has always been there, that was predictable since it is the same problem that Kodak itself had with its own full-frame cameras, that should have been fixed before the trickling release to the public. Leica may eventually provide a fix for this problem, but without an outcry from this forum, if it happens, it will be very slow in coming. The solution is already late in delivery, but then, so are most of the cameras.

 

The maladjusted rangefinders? Having had to send both my brand new M8s to Solms for adjustment after initial purchase and AGAIN after NJ "upgraded" the frame-lines and subsequently adjusted the rangefinder to ITS standard (apparently being with in a couple of inches is NJ's standard of accuracy), this is not a new or infrequent occurrence. The M9 delivered to me was several inches out of focus producing obviously blurry images right out of the box; I refused delivery not being willing to pay full price for a camera that has to go into months of waiting for warranty work in NJ and after that in Solms. Bill Pierce, another fine photojournalist, says this problem of new Leicas having maladjusted rangefinders has been endemic to Leica for years. Leica CAN get cameras out the door with correctly adjusted rangefinders that match correctly cammed lenses, but if there isn't an outcry from places like this forum, they are not going to do it.

 

Joe

[sandymc]

I don't know that we can place too much blame on the beta testers for red edges issue. So far as I know, the only Leica lens that the issue pretty much always shows on in a immediately visible way is the 18mm, which it's quite possible that none of the beta testers owned.

 

What I find difficult is the question of why Leica's internal testing didn't pick up on this - either (a) they didn't test an 18mm for vignetting, or ( they did, and decided to ship the M9 anyway. I find either explanation difficult to accept.

 

Sandy

[Michaelberlin]

I haven't observed red corners with new 35 Summarit... Maybe I need to make a photo of white wal.

 

Will try tonight, especially that I bought 15mm today.

 

The problem with the left side red corners is only to see with lenses below 35mm.

And it mostly appears on snow and white walls.

[Advertisement]

Hallo,

Look here: Leica M

[noah_addis]

This thread appalls me. It along with the threads on maladjusted rangefinders on brand new M9s should be made stickies! They should be sticky and appear at the top with a title "What Leica should fix."

 

I tend to agree for the most part. This problem should have been found. It's not only with the 18mm lens, but even if it were shouldn't the camera have been tested with one of Leica's latest lenses?

 

Look, if there are problems with Zeiss and CV lenses that's different. While it would be great if we could use those lenses easily, Leica seems to only support their own glass, which is their right. And everyone who bought an M9 knew that even with the option to manually set lenses, there are no specific settings for third-party glass. If you need to use cornerfix with your CV 15--that's the price you pay for not buying Leica glass and it's acceptable in my opinion.

 

On the other hand, I'm using a Leica 21 Elmarit Asph, factory coded. It's a lens in the current Leica lineup. Michaelberlin is using a brand-new 28 Elmarit Asph. The 18 is one of the newest Leica lenses.

 

It's inexcusable that Leica didn't know of this problem. Or worse, they did know and released the camera without fixing it or at the very least issuing a warning that, pending a future fix in firmware, certain wides don't work well.

 

The people who designed the M9 know that there are corrections going on for color and vignetting. Knowing that, I can't comprehend how they would not do some white-wall tests to check for these issues. It should have happened before the beta-test stage. To me the manufacturer should do scientific tests while beta testers should do real-world testing.

[thrice]

Truly, that is the crux of it Sandy (and Noah), they must have either done drastically insufficient testing with wide angles, or just wanted the product out the door warts & all.

 

They really can't afford to be too blasé about the problem and just rest on their laurels. A firmware should come quickly and fix at least the glaring problems with the camera. Performance optimisations can come second.