Monochrom Banding Normal?

[MarkP]

Advertisement (gone after registration)

Monochrom as above

except correctly exposed and not pushed

 

There is just the faintest of banding in the shadows by #3 on continuous, but it needs a good screen viewing the original file to clearly identify (barely seen on jpeg).

 

So the problem is inherent to M9 and Monochrom I assume struggling to write in continuous mode, and exacerbated by significantly pushing underexposed areas.

 

I think it's just that people have been squeezing Monochrom files for all they are worth and with respect to this have identified their limits (and those of the M9).

 

This should put the issue to bed!

 

As I never shoot like this I'm not fussed, but I hope this is helpful to others.

 

 

 

#1 100% crop

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!

 

 

 

#5 100% crop

Edited June 9, 2013 by MarkP

[rfnoob]

This issue occurs to different degrees with different cameras (M9, M9-P and MM). I have not put my hands on an M-E so don't know but am quite certain it would occur on those as well. My suspicions as an electrical engineer is that this is the SD card writing is generating interference with the CCD sensor. It's highly unlikely that writing to the card is in error because there is error correction when writing to the card.

 

One is able to trigger this when one shoots while the buffer is being written to the card. Luckily my MM is one of those where the banding is very minor because I can't "downgrade" the firmware.

 

On my M9, I've gone back to 1.162 so the card writing is much faster and therefore I can work around waiting for the buffer to complete writing before shooting my next shot.

 

I would be extremely happy if there is a firmware fix but I'm not holding my breadth as this seems to be more of a hardware design issue than a software issue. But I would be more than happy to be proven wrong.

[Advertisement]

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

[jffielde]

I can confirm that I can generate banding like MarkP's (which I will call "fine" banding as opposed to the "gross" banding in my original Monochrom) under similar circumstances. If I push an underexposed file more than two stops, I can start to see it in some cases in very low light, and generally I can see it at 3+ stops push in low light. I don't notice it at all in my normal workflow, but lots of people here are much more attuned to that sort of thing than I am.

 

This behavior is repeatable (just as my catastrophic banding was before). It appears there are two work-arounds if I needed them: 1. Wait for the buffer to clear if I need an especially clean file to push several stops; and (ii) shoot at ISO 400, which seems to cure the issue altogether on my camera, even in continuous mode. If I need continuous mode, I'd choose the second option.

[jffielde]

One last note: I've examined files from cameras I've owned or used recently to see what banding they might produce when similarly pushed in LR4. My Monochrom and M9 don't show banding (that I notice, at least) until pushed 2-3 stops. It isn't obtrusive to me until 3 stops or more. The Canon 5D3 is worse than both Leica cameras by a full stop - maybe more. The Canon 1Dx looks to behave similarly to the two Leicas but with a cross-thatch pattern rather than all-horizontal banding. And again, I'm comparing only the shots recorded while buffering in the two Leica's since they seem to push to full white without banding on the first shot. Finally, I threw in my Sony RX100, which seems to band almost none. It's similar to the Leicas on their first shot, but not quite as good.

[raumplan]

I can confirm that I can generate banding like MarkP's (which I will call "fine" banding as opposed to the "gross" banding in my original Monochrom) under similar circumstances. If I push an underexposed file more than two stops, I can start to see it in some cases in very low light, and generally I can see it at 3+ stops push in low light. I don't notice it at all in my normal workflow, but lots of people here are much more attuned to that sort of thing than I am.

 

This behavior is repeatable (just as my catastrophic banding was before). It appears there are two work-arounds if I needed them: 1. Wait for the buffer to clear if I need an especially clean file to push several stops; and (ii) shoot at ISO 400, which seems to cure the issue altogether on my camera, even in continuous mode. If I need continuous mode, I'd choose the second option.

 

Thanks.

 

It sounds as though your new camera still has "fine" banding in consecutive shots and the only work around is to wait for buffer to clear and shoot at iso 400.

[jsjxyz]

Don't use Toshiba FlashAir 32GB in MM, it create 6 terrible horizontal bandings, 

[Lenshacker]

Advertisement (gone after registration)

This is an old thread, I've been using Sandisk 8GByte 4x cards with very good results. I do not see banding with the slower cards that I get with the 10x and faster cards. Speculation- bursts of data to the faster cards, buffers filling up, slowing down, then speeding up again- have an overall effect on the data still be read-out and digitized by the CCD. The image takes about 1/2 second to be shifted out and converted to digital. During this time the analog signal is subject to induced noise and shows up as banding. Keep everything flowing smoothly, even power draw- banding seems to disappear. "works for me", may or may not be correct, but my ISO10000 images are free of fixed-pattern noise.

[Lenshacker]

I posted this example on another thread, but it seems appropriate here as well.

 

100% crop, almost a 6 stop push of the shadows.

 

doorway_5p5stops by fiftyonepointsix, on Flickr

 

The original ISO 320 shot:

 

outside_building by fiftyonepointsix, on Flickr

 

This is what I've been used to since Jan 2013. I'm using 4x cards.

[Fireboy]

It's been a couple years since the last post. I'm discovering this on my recently acquired 2nd hand monochrom CCD.

 

I suspect that it's NOT the SD card - I tried, 8, 16, 32gb capacities and Lexar and Sandisk brands. I've been able to reproduce the banding in a consistent and repeatable fashion.

 

My tests included shooting one frame, waiting for the buffer to clear, then taking another shot. I also did shot after shot. I then opened the images in lightroom and raised the shadows from +.03 up to 5 stops.

 

What I've discovered has more to do with the sensor (and likely temperature of it) than the buffer or write speed or SD card. 

 

The first shot is ALWAYS consistently perfect, NO BANDING. Then the 2nd, 3rd, 4th contain banding. If I then wait a couple minutes, try again, I'll get the same results. 1st shot is clean, following shots have varying degrees of banding.

 

Here's the interesting part. Seems to only be at 320 as previous posters have pointed out. Switching to 400, 800, 1600 - the banding doesn't appear.

I reported my results and screenshots to Leica NJ asking for a goodwill sensor replacement once I receive my M10. However, if this is the nature of the CCD then I'll learn to shoot at 400 instead of 320. 

 

I'm running v1.002 firmware.

Edited January 21, 2017 by Fireboy

[Ken Abrahams]

Shocking banding appeared on some long exposure images when my battery was quite low using MM CCD. Pobably due to power available for noise reduction in camera.  In low light, banding appears when attempting to bring more shadow detail with processing however if you use slightly higher ISO and convert to TIff file before processing the banding is negligent.After a sensor replacement I have been having less issues with banding.  

Ken

[Alberti]

It's been a couple years since the last post. I'm discovering this on my recently acquired 2nd hand monochrom CCD. [  ]

 

I suspect that it's NOT the SD card - I tried, 8, 16, 32gb capacities and Lexar and Sandisk brands. I've been able to reproduce the banding in a consistent and repeatable fashion.

 

Here's the interesting part. Seems to only be at 320 as previous posters have pointed out. Switching to 400, 800, 1600 - the banding doesn't appear.

 

[  ] However, if this is the nature of the CCD then I'll learn to shoot at 400 instead of 320. 

 

I'm running v1.002 firmware.

On my M8 I have seen this same, due to the battery "resistance" as I called it myself or cold weather; the power buffer seems to have some (tens/hundreds? of) millivolts of juice drop when very active.

on my MM1 there is an occasional banding (low light)  only when shadows are pushed too far. I never shoot series. It is due to the ten years old design.

I like it 100% more than my other CMOS camera in this aspect.

I take it for granted.