Noctilux 0.95 purple fringe

[vanhulsenbeek]

Advertisement (gone after registration)

.....

 

I also dont get his with my with my 50 1.0 Noctilux using leica film M.

Has anyone tried the 50 mm .95 version on Leica film body.

 

 

Very likely you will never see purple fringing on film, as it is a pixel problem. Film has no pixels, AFIK.

 

Chromatic aberation; yes. With distant lights in the night, or with stars. They may show up multicolored.

[Double Negative]

...When the much acclaimed Canon EF 85mm f/1.2L II USM lens came out, Chasseur d' Images published a review and showed the shocking purple fringing that lens caused.

 

Too much light and to much contrast and there you go.

 

Indeed, my 85mm f/1.2L II shows this in the right situations.

 

This picture doesn't clearly illustrate this, but at full size you can easily see it:

 

[mah]

New Noctilux shot:

 

Millennium Bridge | Flickr - Photo Sharing!

[neli]

As one of the relatively impoverished peasants, I can't help feeling some smug schadenfreude reading this thread.

 

John.

 

can we have a nocti forum please, keep out the riff raff ;-)

[mjh]

New Noctilux shot:

No need to spam all Noctilux-related threads.

[jaques]

I just got a f1 Noctilux and I see the same purple fringes in high contrast situations.

[Googaliser]

Advertisement (gone after registration)

I have shot several thousand frames with the 0.95. I have seen fringing occasionally - but it is quite predictable and easily fixed in post (assuming the high contrast situation hasn't resulted in a poor image anyway - regardless of fringing). To write this lens off for this , is like saying 'I would never buy a Ferrari because they handle badly in the snow'. Nothing is perfect - although for me, photographically, the 0.95 comes very close.

[mah]

New Noctilux shot:

 

St Paul's Cathedral | Flickr - Photo Sharing!

[57andrew]

After shooting 4000 frames with my M9, probably >75% with the 0.95 notci I finally experienced this phenomenon. The subject was 2 people in the shade on a boat with a backdrop of bright sky. The image was taken at F1.4 and the fringing was horrendous. I went into CS4 and desaturated blue and magenta in the hue/saturation option and it improved it significantly but not entirely. For a 1 in 4000 problem I can live with this especially now understanding the circumstances that cause this to occur. Useful thread.

[Washington]

Hmmm, unpopular as my comment may be: there is no need for a 0.95 lens unless

it is required for special purposes. A total waist of money otherwise. And, believe me,

I waist my money on sillier things so this is not a judgement-call.

Whatever you like is fine, but isn't it simple to create a few "Corner Fix" profiles

to correct this issue?

[57andrew]

Your comment is neither popular nor unpopular. You are entitled to your opinion.

[Olsen]

I have both the old Noctilux 50 mm 1,0 with the built in sun shade and the old Canon 50 mm 1,0L. Neither is good lenses and both show 'blue edges' along branches etc. To what extent and comparable the new Noctilux 0,95 is difficult for me to decide. I regard both these two lenses as curiosities - like 'the village idiots' of the lens world. Neither is particularly good nor sharp, but aparture 0,1 is hefty. Nice to have, sort of. I had a Carl Zeiss ZM 50 mm 2,0 which is far better.

 

The largest problem I have with these 0,1 lens is focusing. I simply don't have the eyesight anymore to be able to place a 2 cm DOF dead on at 2 meters. - If I ever had it. Too often it is the nose tip that gets sharp, not the eyes. Add to this that the noctiluxes front focuses slightly ( about 1 cm) at full aparture at 2 meters.

 

To what I can see of tests and pictures on the Net, the new Leica 50 mm 0,95 represents a wast improvement in sharpness and contrast (my Noctilux is absolutely 'dull'), and microcontrast. Jaap points out that it is lenses with high microcontrast that is prone to blue edges. The old Canon 28-70 mm 2,8L was very prone to this on the old 1Ds.

 

That this can be fixed in C1 is new to me. How?

[01af]

There are at least four different phenomena, and all have different intensities, causes, and remedies. In discussions like here, these usually get all mixed up and confused with each other.

 

1. Lateral chromatic aberration – Lens fault. Will occur on both film and digital. Cannot be reduced by stopping down. Manifests as complementry-coloured fringes (red/cyan, purple/green, or blue/yellow) along high-contrast edges; absent at the frame's center; gradually more pronounced towards the frame's corners. In digital captures or scans, can be eliminated or at least greatly mitigated using the chromatic-aberration controls of the raw converter or image processor.

 

2. Longitudinal chromatic aberration – Lens fault. Will occur on both film and digital. Can be reduced by stopping down. Manifests as complementry-coloured fringes (purple/green) along out-of-focus edges across the whole field; absent at the plane of focus; gradually more pronounced towards the out-of-focus areas. Purple fringes usually occur before, green fringes behind the plane of focus.

 

3. Purple fringing – Bayer sensor fault. Will occur on digital only. Can be reduced by stopping down (while keeping exposure constant). Caused by wide light cones hitting the sensor's photo sites. Manifests as magenta-coloured fringes along high-contrast edges across the whole field.

 

4. Blooming/halation – Sensor/film fault. Will occur mostly on digital but can happen on film, too. Cannot be reduced by stopping down as such but by reducing exposure. Caused by over-exposed areas spilling over into adjacent darker areas. On Bayer sensors, manifests as magenta- or violet-coloured fringes along high-contrast edges across the whole field.

Basically the problems #3 and #4 can occur on Foveon sensors, too, and also on film albeit to a lesser degree (the film's anti-halation coating is supposed to supress this effect). However their colour won't be purplish but they will manifest as (more or less) neutral-colour flares. The purple colour on Bayer sensors comes from the asymmetry of the colour filter array—twice as many green pixels as red or blue ones, and purple (magenta) is the complementary colour to green.

 

The original poster's problem is mostly #3, I guess, possibly with a bit of #4 thrown in.

[jaapv]

What Olaf says - different raw converters will handle 3 and 4 differently. Best results have been reported with C1.

[Washington]

Woody, I was not aware that Capture One did this and so I find this of interest.

Honestly, one thing I do not understand is folks reluctance to use software and

post processing. I guess they figure the camera has to do all the work...or something.

Sooting in RAW requires post processing no matter what camera one uses.

[tgray]

2. Longitudinal chromatic aberration – Lens fault. Will occur on both film and digital. Can be reduced by stopping down. Manifests as complementry-coloured fringes (purple/green) along out-of-focus edges across the whole field; absent at the plane of focus; gradually more pronounced towards the out-of-focus areas. Purple fringes usually occur before, green fringes behind the plane of focus.

 

 

Not to be pedantic, but longitudinal chromatic aberration occurs in the plane of focus. The lens can't bring different colors in focus in the same plane. I would imagine what you describe, color fringes in the out of focus areas ('bokeh chromatic aberration' or some similar name), is related, but probably technically not longitudinal chromatic aberration. Even if it is considered to be that, longitudinal chromatic aberration CAN occur in the plane of focus - it's just that most manufacturers have design it out of the lens effectively.

[adan]

01af makes the best point. There are multiple possible sources for color fringing - and they can occur simultaneously in the same picture, reinforcing one another.

 

Take a lens with:

 

high overall contrast - the light energy in the highlights tends to stay in the highlights and overexpose them, rather than bouncing around and being evenly distributed over the whole picture,

 

and chromatic aberrations (inherently more common in extreme lenses with very long or very short focal lengths, or unusually large apertures (e.g. f/0.95)

 

and use it to photograph a contrasty situation with lots of dark areas silhouetted against very light areas...

 

and the odds of color fringing will go up and up and up.

 

Additionally - keep in mind that the violet end of the visible spectrum is the most energetic (ultraviolet light will burn your skin; red light usually will not). So if there is an overexposure of WHITE light - the VIOLET wavelengths will be what spreads and flares most energetically. Giving these flares and fringes their characteristic violet tint. If the overexposed background is BLUE sky, that is also towards the violet end of the spectrum, and will add its tint bias to the fringes as well.

 

Even without "chromatic aberration" as such.

[Jeff S]

01af makes the best point.

 

You mean made...almost 3 years ago.

 

Jeff

[CheshireCat]

The purple colour on Bayer sensors comes from the asymmetry of the colour filter array—twice as many green pixels as red or blue ones, and purple (magenta) is the complementary colour to green.

 

Demosaic algorithms give G pixels half the weight of R and B. This means that when all sensels are saturated, the final color of the blooming will not be magenta but white, as most of us have experienced on CCD Leica cameras (CMOS sensors are far more resistant to blooming).

That said, how can you explain #3 and #4 above ?

 

P.S. This thread was cross-referenced from:

http://www.l-camera-forum.com/leica-forum/leica-m-lenses/366748-ca-problem-my-50-1-4-a.html

[Advertisement]

Hallo,

Look here: Leica M

[Guest NEIL-D-WILLIAMS]

On the Sony a7ii I have never seen this pneumonia when using my Noctilux 0.95...saying that mine is one of the newer ones