M262 - unexpected banding outdoors at ISO 1600

[Joey Pasco]

Advertisement (gone after registration)

Hi all,

 

I recently purchased an M262, and for the most part I absolutely love it. However, yesterday I noticed an image that has some fairly obvious vertical banding in a situation where I would not have expected it. I haven't noticed this in any other images, despite shooting indoors at ISO 1600 the majority of the time. Should I be concerned about the sensor?

 

Shooting DNG, attached is a screenshot of the issue SOOC. The banding is only visible around her neck/chin/face, but nowhere else.

 

 

 

 

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!

Edited September 19, 2017 by fotolorea

[Joey Pasco]

Here's the affected area with the exposure boosted so the issue is more apparent.

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!

[Advertisement]

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

[sblutter]

Looks like high ISO noise in an underexposed area to me.  There are several noise reducing techniques you can look up to improve the image. You also may want to reshoot it in stronger lighting.  The 2nd shot has been clearly brightened, which unfortunately enhances noise when lightened few stops in all digital cameras.

 

You'd think 1600 should be fast enough but as you can see... not so much

 

Being back lit didn't help either

 

I suggest you do tests to see what lighting conditions your gear can handle with ease

[jdlaing]

Me too.

[Gregm61]

I've only noticed obvious banding in images recorded in the extended ISO range above ISO 3200 with my M262. I cannot tell how radically you had to raise the shadow area to reveal that banding, but I'm guessing it was more than just a little.

 

I recently purchased the Topaz Denoise to use in Photoshop and noticed a De-banding option that I've not yet tried. The noise reducing slider is quite excellent compared to the noise filter in ACR, so I'm figuring the de-banding feature should be pretty good.

Edited September 19, 2017 by Gregm61

[Joey Pasco]

I've only noticed obvious banding in images recorded in the extended ISO range above ISO 3200 with my M262. I cannot tell how radically you had to raise the shadow area to reveal that banding, but I'm guessing it was more than just a little.

 

 

 

The first image is straight out of camera, no editing whatsoever. The banding is visible without making any adjustments to the shadows. The second image I posted has been adjusted to make the banding more visible.

[lct]

Advertisement (gone after registration)

Typical underexposed image. Pushed by +2 EV or more it shows noise and banding like a 6400 iso pic or more. Nothing to worry about IMO, suffice it to avoid 1600 iso outdoor and/or underexposure and/or pushing too much in PP, or choose an M10 or another cleaner camera.

[LocalHero1953]

Looking at the first, SOOC, shot, I am surprised, even though it is underexposed. It is clearly visible. I would not expect this on my M240. I do not have the M10, but understood it performs better.

Sure, in shadows pushed in PP you will see banding, but the first shot is not pushed.

 

(And I have LOTS of experience of poorly exposed images )

Edited September 19, 2017 by LocalHero1953

[Joey Pasco]

Looking at the first, SOOC, shot, I am surprised, even though it is underexposed. It is clearly visible.

 

This is my concern, and the reason I posted. I've seen banding in high-ISO images before, especially after being brightened in post-processing, but this banding is actually visible in the image as shot on the back of the camera (and in the DNG file, as seen in my screenshots). 

 

It just seems very unusual to me that I'd get such pronounced banding under these circumstances. Yes, it is slightly underexposed, but the banding shouldn't be this obvious (I don't think) until I try to address that underexposure in post.

 

I'm going to assume it's a fluke and try to nail my exposure better in-camera. I've been shooting mostly film over the past few years, so I'm still adjusting to digital exposure methods. Everything I've read told me it was better to underexpose in-camera, so as not to blow the highlights. I guess I have to figure out some sort of happy medium between underexposure and blown highlights.

[Luke_Miller]

All of the (rare) banding I've experienced has been horizontal rather than vertical.  I wonder if the very overexposed background isn't introducing some other phenomenon. 

[jmahto]

I am coming from M240 experience. I won't expect to see banding at ISO 1600 without pushing. If I were you, I would run a controlled test with a verity of scenes, at different ISO values.

[Mark II]

On my M 262, I can sometimes see banding but only in shadow areas with extreme (3 stops or more) pushes at ISO 3200 - and it would be impossible to see on the back LCD as even then it is very faint. Banding performance is roughly on par with the 5D III I used to shoot - annoying but rarely an issue.

 

And yes, the banding is horizontal (with the image in landscape orientation), not vertical...

[adan]

Well, I have seen that "shot through a screen door" look from some (mostly, but not always, earlier) CMOS sensors (e.g. Canon 5D Mk1). Horizontal and vertical bands overlapping. Usually it is "fixed pattern noise" (FPN), particularly the type caused by pixel response non-uniformity (PRNU) as opposed to dark-signal FP noise (traditional "hot" pixels in long exposures).

 

https://1.img-dpreview.com/files/p/E~forums/55391061/1c05327134b04fec82581f81c503fa7f

 

https://www.eecs.tu-berlin.de/fileadmin/fg144/Courses/10WS/pdci/talks/noise_sensors.pdf

 

At any rate, it seems a bit excessive for the M240/M262 sensor, based on my limited trials with the 240 - where with correct exposure I only got horizontal bands, and only above ISO 2500. I'd second Jayant's suggestion to do some tests in varying scenes/lighting/ISOs.