Dark banding on M11?

[dante]

Advertisement (gone after registration)

Is there an issue with the M11 shutter (both in mech and hybrid mode) with dark banding and LED room lights? I am getting a dark band about every third or fourth shot. It's almost like it's related to the frequency of the AC power (60hz). Sometimes it's the top of frame, sometimes the bottom, sometimes the middle.

ISO 1600, 1/320 sec

Thanks,

D

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

[lct]

•

Edited April 6 by lct

[pedaes]

5 hours ago, dante said:

banding and LED room lights?

Banding is well known with electronic shutter and some artificial light - common to all makes of camera. Not sure that what you show is usual banding though as it tends to be multiple bands and not that dark, but who knows from specific light you were under.

Edited April 6 by pedaes

[jaapv]

This is weird, the edge is so unsharp that it does not look like a sensor/shutter problem, but more like something dangling in front of the lens. 

[dante]

2 hours ago, jaapv said:

This is weird, the edge is so unsharp that it does not look like a sensor/shutter problem, but more like something dangling in front of the lens. 

It's not anything in front of the lens (and it would be very difficult to get something 1/3 of the way down the frame like that - or randomly through the middle like on other shots). I do know what banding looks like in various forms (electronic shutter, focal plane, card writing) and for various reasons, and this one is puzzling (and irritating). I have sent some samples to Leica to get some understanding. Here's one that looked black but with the exposure cranked up.

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!

[dante]

3 hours ago, pedaes said:

Banding is well known with electronic shutter and some artificial light - common to all makes of camera. Not sure that what you show is usual banding though as it tends to be multiple bands and not that dark, but who knows from specific light you were under.

Not an electronic shutter here. This is the mechanical.

Banding on electronic shutters (which I have observed M11M and Sonys that I suspect have the same base 61mp sensor) manifests as a "venetian blind" look, not this strange, symmetrical, hemispheric shape that changes position from shot to shot (and sometimes does not show at all). Take a look at the sample I posted above (obviously abused in post to make the shape more clear - the raw is super dark). This looks very much like what would happen when you would photograph a television set with film in the old days - but that would be a straight or diagonal band across the frame.

D

[jaapv]

Advertisement (gone after registration)

55 minutes ago, dante said:

It's not anything in front of the lens (and it would be very difficult to get something 1/3 of the way down the frame like that - or randomly through the middle like on other shots). I do know what banding looks like in various forms (electronic shutter, focal plane, card writing) and for various reasons, and this one is puzzling (and irritating). I have sent some samples to Leica to get some understanding. Here's one that looked black but with the exposure cranked up.

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!

I find it puzzling as well. My prediction: Leica will replace the main board. 

[Carlos cruz]

39 minutes ago, jaapv said:

find it puzzling as well. My prediction: Leica will replace the main board. 

No need for physical intervention, it could be done on software level, l LEDs with lack of standards tend to have frequencies all over the place not sticking to 50/60 Hz. There’s always a shutter speed possible for every Hz frequency but it could be a very precisely defined number like 1/147 of a second. One could possibly hunt for exact speed, but with lights that don’t adhere to any standards it would be next to impossible doing it in every situation. My solution for such situations is to go as low with shutter speed as possible.

[jaapv]

Yes, but normally the banding is much more defined and parallel to the long side of the sensor, not skewed. 

[PhotoCruiser]

As Carlos mentioned, the way to find out if the LED lights are the problem is to play with shutter speed.

But i agree with Jaapv, it does not look as a shutter problem and not LED banding either but i don't exclude that. I have a similar problem with one of my old lenses (Hector 9.5cm ) on the SL2 with a M39->L Adapter.
If i screw the lens tight in the adapter i get a dark stripe, if i unscrew the lens and turn it about 1/4 the stripe is away, but my stripe/dot is all on the bottom.
If you use a adapter then try to turn the lens slowly to see if stripe moves, just to eliminate this hypothesis, and do the slower shutter speed with various speed too.
Chris

[Carlos cruz]

1 hour ago, jaapv said:

Yes, but normally the banding is much more defined and parallel to the long side of the sensor, not skewed. 

Direction and shape of banding usually depends on the characteristics of the light source and order in which pixels are processed by camera. The skew is probably effect of pixels being read from upper left corner line by line, maintaining same direction for each line of pixels. So there’s some minuscule time difference between the moment when first and last pixels are “read”. Banding not having well defined edges means that light source probably isn’t operating only in 1 and 0 but has in between values. Ideally it could be “fixed” if the pixel reading order pattern didn’t follow the way we read(left to right, line after line) but was more of chaotic type, but I believe It’s much easier to have your lights powered with some higher Hz ballasts, than change the order in pixel queue.
@PhotoCruiserlenses having their own banding issues depending on mechanical compression?! That’s too much of guessing games for one Sunday afternoon I give up, no idea.
 

 

[J S H]

If you don't see the banding under sunlight, then it is tied to the artificial lighting. You can still see banding artifacts from artificial lights with a mechanical shutter - it's tied to the shutter speed. There is no set formula, as there is no guarantee that any two lights are using the exact same frequencies and have the same color accuracy/response. Longer shutter speeds will indeed get you past most issues with artificial lights, as you then catch more than one cycle during the exposure. 

[adan]

15 hours ago, dante said:

Is there an issue with the M11 shutter (both in mech and hybrid mode) with dark banding and LED room lights?

The short version - "Yes."

Longer - typically at any shutter speed above the mechanical "flash sync" speed (sensor uncovered all at once -no shutter "slit"). And yes, due to the AC current fluctuations.

Bottom line - old school electrical "hot tungsten/halogen" lights did not flicker, because the hot filament did not have time to change its output significantly (lose its high radiant temperature) in 1/60-50th of a second.

Electronic LED bulbs can and do change brightness much faster. At least the "basic consumer" ones - there probably exist specialized ones (for a price) with more "continuity" engineered in - a capacitor-diode or similar that serves as a "flywheel for electrons" and smooths the peaks and valleys of AC.

Generally, I try to stay below 1/250-180th sec. (M10) when (increasingly) taking pictures under LED illumination.

 

[PhotoCruiser]

vor 14 Stunden schrieb Carlos cruz:

@PhotoCruiserlenses having their own banding issues depending on mechanical compression?! That’s too much of guessing games for one Sunday afternoon I give up, no idea.

My Hector has a tripod attachment and the counter plate inside is visible depending how the lens is positioned, if i turn the lens or the adapter it disappears.
The black band on the first photo seems to be a little skewed to the right, right side goes a bit more down for this my guess.
Chris

[dante]

Leica got back to me in record time and told me that this is an interference pattern with a high-frequency LED light. They said the fix is to use 1/125 or slower.

One day, I would also like to see camera meters that do not have sampling rates that get tripped up by lighting. I noticed in the course of testing this that the M meter still has a sampling frequency that results in the reading fluctuating under modern light sources that look constant to the naked eye. 

Dante