Banding with the E-M5 and the 20/1.7: A potential work-around

Started Oct 19, 2012 | Discussions thread
Michael J Davis
Michael J Davis Veteran Member • Posts: 3,354
Re: Banding with the E-M5 and the 20/1.7: Before you'll got bogged down with electronics...

Anders W wrote:

Michael J Davis wrote:

I've just caught up with this thread, and spent a happy 80 minutes reading it all.

It occurred to me that there were some simple tests missing. So I've done a quick and dirty check all at AWB ISO8000 P-mode -1 stop exposure compensation looking at jpgs only.

Please note that I can create banding in the conditions that you have shown them - but never as severe as your seem to be.

1. What happens if the lens is in the same position but not electrically connected?

(i.e. is it electric/onic interference between body & lens - or emf effects?)

Using a paper disk behind the lens to give contact isolation and translucency. The PP to max contrast.

Answer seems to be:-

No banding with daylight source (in conservatory -has mains supply but not all round), but...

... V. slight detectable banding with tungsten lighting indoors light reflecting off a piece of mount card (1/125th second). However I can also see marks on the sensor as I can see them in the same position on all the photos to a greater or less degree. These sensor marks are of the same order of magnitude as the apparent banding.

I doubt if this amount would be noticeable in a normal photo (clearly I haven't been bothered by them up to now).

Thanks for helping out with the testing. That's very welcome for many reasons, in part because we don't know yet whether different copies of the camera and lens behave slightly differently.

Testing without electric contact is a nice addition and I have been thinking about trying that. But I am not sure I understand exactly how you did it ("paper disk behind the lens"?). Can you describe that in further detail? The lens wasn't really mounted or what? What I was thinking of doing was simply to put some tape on the lens contacts to isolate them from the pins on the camera.

Sure, after trying to cut some Scotch tape to fit over the pins, and worrying what would happen if a bit got on the sensor, I merely found a piece of glossy 'flyer paper' (ie. what flyers are printed on) complete with printing and cut a disk to fit in the opening. I chose glossy to minimise friction with the pins. It was also somewhat translucent, and so gave me a reasonable exposure at high ISO.

Masking disk

I cut a disk of paper as above that covered the contacts but wouldn't fall into onto the sensor.

I am also not sure what to think about your results in the second test ("marks on the sensor in the same position"?). Can you please say more about this as well and perhaps post some sample images from both tests?

Umm. Have you ever tried a pinhole lens, say f:160 with a digital camera - it will show you a whole new world of nasties adhering to the sensor.  Because the disk (and later the translucent, but oof layer in front of the lens, all one sees is the shadows of gunk on the sensor.

2. What happens if camera + lens is set to manual focus? Is there interference in the AF?

Here I used a translucent lens cap (from enlarger) to diffuse the light - so more uneven and thus slightly more difficult to identify. Then I tried to create banding on piece of grey card out of focus.

Both sets of shots were taken in the conservatory (see above).

Apart from the sensor marks referred to above I'm unable to create banding, but that may be because I didn't allow enough time (8mins) for the camera to cool down between sequences.

This is a test that I as well as others have tried. And yes, you get banding even with MF. The reason why you didn't get it was in all likelihood the one you mention: you didn't let the lens cool down sufficiently long. I try to leave it alone for a couple of hours before considering it "cold" again.

OK, I'll have another go when I get time.

As I say these were intended to be rough and ready - but designed to test a couple of factors that seem to have been missed so far. All it's done is to upset me about the condition of my sensor! The banding is nothing like as obvious as your examples - and only just distinguishable from the background noise. For this reason, I'm not posting examples right now.

OK. But please pody anyway. Perhaps we can help you out with the sensor marks you are worrying about.

I'll post them, but the distinction between the shadows and noise is only obvious when looking at several side by side. I need to let you know what degree of contrast enhancement brings them out. Need to do a bit of work on that.

I should say about the first sample in my OP that it is an unusually clear case. That's in part why I chose it.

OK, I can get that - like you it's usually against woodgrain. I'm unusually busy at present, so it may take a few days to get a simple and coherent posting together. (I really want to repeat the tests.)


-- hide signature --

Mike Davis
Photographing the public for over 50 years

 Michael J Davis's gear list:Michael J Davis's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix DMC-GX85 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +5 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow