M8 Banding Problem [Archive] - Rangefinderforum.com

jlw

11-03-2006, 09:34

The Nikon's problem was caused by the way it scans the CCD to get the data off it. One reason the D200 is so fast is that it has four separate channels (data paths) to scan the CCD; if I remember the explanation correctly, some types of images could cause these scans to get slightly out of sync and cause a banded appearance in the images.

I've had a few images from my R-D 1 that exhibit what I might call "banding" or "smearing" -- generally they happen when there's one very overexposed area in an otherwise dark photo, such as a bright window in the back of a dark room. (I've got a perfect example image at home, but can't get at it here.)

This causes a light streak, the same height as the overexposed object, extending lengthwise along the frame. It fades out eventually.

I've read that this is also caused by the way the imager is scanned to get the data out; the very strong signal from the photosites in the overexposed area "leaks" onto the adjacent sites. Apparently it's somewhat related to the phenomenon of "blooming" that you get with a video camera.

I still can't see the sample images, but does this sound like a description? If so, the problem can't be solved by using higher bit depths; it happens at the analog stage of the signal being formed on the photosites.

jaapv

11-03-2006, 09:39

I seem to remember a similar " issue " thread on FM when the 1Ds came out if memory serves. I never hear anything about it anymore nowadays... Welcome to the digital world, it has been said before.

nksyoon

11-03-2006, 11:00

I still can't see the sample images, but does this sound like a description? If so, the problem can't be solved by using higher bit depths; it happens at the analog stage of the signal being formed on the photosites.

Yes, it's a horizontal band same height as the light source stretching across the image.

See page 19 in the Italian article: http://rpo.eranet.tv/files/Leicapassion%203-2006.pdf (slow download - it's 51 pages).

petermcwerner

11-03-2006, 11:23

I never experienced or heard of a similar problem with the DMR. How come the older DMR is better in than respect than the M8?

Terao

11-03-2006, 11:42

On the D200 it was mostly contained in one of the channels (ISTR blue)
I've seen the issue JLW reports both on a D70 and R-D1. So I wonder if its a data-shunting thing (D200) or a blooming thing. Blooming can't be resolved whereas the data-shunting thing can be - Nikon repaired D200s under warranty.
This might also be due to the absence of an AA filter in the image path...
It really doesn't surprise me with fast lenses, dark scenes, and a few blown highlights - a combination that will show up any shortcomings a sensor has...

John Camp

11-03-2006, 13:00

I was looking at those photos and my thought is, I'm not sure you could hold anything, film or digital, with that much overexposure. The shot was taken at night, ISO 1250, f2, at 1/25. The apparent banding came from a row of very bright streetlights. So how many stops overexposure is that? Ten, maybe? More? And these are not point-source specular highlights, they are large streetlights. I suspect this might be beyond the ability of any sensor to hold without some kind of light spillover.

ywenz

11-03-2006, 14:19

That street light shouldn't be too difficult of a subject for any modern imaging sensor to handle. My GRD handled these lights without banding... The wonderful star effect is the result of a dirty front element.

http://static.flickr.com/48/133699194_0ac1f6dab8.jpg

John Camp

11-03-2006, 19:02

Ywenz,

I don't really know what I'm talking about here, but your shot is somewhat different than the ones that show banding in the M8 (I've seen two shots so far with banding.) In both, the banding came when the high light source was immediately adjacent to a dark color -- and only that way. When the high light source (in the same photos) grades into something else -- white curtains, then a wall, and so on, I don't see the banding. Your high light sources play out across a light-to-dark gradiant.

I also see a bit of a blooming effect in some M8 shots, similar to that lighter area in your shot, around the top of the low tan building in front of the Hancock; but that looks "okay'" to me -- like something I'd see with my eye.

Sean Reid has looked at a number of shots, and found the banding problem only in a couple (of many) that included bright lights at ISO 2500. At this point, I'd say it's a problem that resembles the moire thing -- it'll effect a few shots done in extreme conditions.

JC

jlw

11-03-2006, 20:53

Here's my banding example, made with an R-D 1 at a performance by a Sudanese choir. Note the purple streak running through the dark area on the left side, aligned vertically with the very bright window to its right.

I'm guessing that the signal is pulled off the sensor from left to right (remember that the image would be reversed in the camera) and if a really bright area oversaturates the imager in one area, the excess signal will be "smeared" horizontally across the adjacent areas of the image until it dissipates. The worst-case scenario would be a really bright area right at the edge that gets scanned first, with a deep black area all the rest of the way across the frame.

If my guess is right, I'd think it would be possible to create conditions where this might happen with almost any digital camera, but there could be interesting variations:

-- Would a CMOS sensor behave differently than a CCD?
-- Would a camera that pulls the data off the imager more slowly do better, by allowing more time for the excess signal to dissipate? Or would a fast-scanning camera do better, by getting past the "hot" area before the signal has time to cascade to the adjacent photosites?
-- Does a small sensor perform better than a large one, or vice-versa?

I know we've got some digital-sensor experts lurking about -- anyone care to provide an authoritative explanation?

StuartR

11-03-2006, 21:01

The DMR never did that in my experience. Not that it did not have several issues in the beginning.

jlw

11-03-2006, 21:05

I've seen the issue JLW reports both on a D70 and R-D1. So I wonder if its a data-shunting thing (D200) or a blooming thing. Blooming can't be resolved whereas the data-shunting thing can be - Nikon repaired D200s under warranty.

If it's blooming, it seems as if it would bloom symmetrically in all directions, the way a video-camera image does. In all the cases I've seen, the streaking has a definite horizontal direction, and always stays aligned vertically with the height of the bright light source.

What I need now is a rainy afternoon with nothing else to do, so I could piddle around with some simple experiments...

jlw

11-03-2006, 21:37

I briefly read the italian article and indeed there are a couple of images that show banding.
The reviewer (Roberto Piero Ottavi) stated that the banding occured only with his Voigtlander 28 Ultron and it doesn't show when using any Leica lens (either asph or pre-asph) maybe because of "compatibility issue" between microlenses and light angle from non-Leica lenses.

But you can easily see the vertical (actually horizontal giving the portrait orientation) banding in the Noctilux image, so maybe we have a real issue here.

Yes, I can see it too. It's not as evident in the Noctilux image because the bright streak (from the small street lamp in the background) lines up fairly closely with a light area in the background. Against a completely black background I suspect it would be more visible.

It's too late and I've got too busy a weekend coming up to do this myself, but here's a suggested test rig for someone who'd like to explore this phenomenon:

-- Get a large sheet of some black, opaque material such as black foamcore board. Given the typical RF lens minimum focusing distance of 1m, you'll probably need a sheet about 0.5m high; the US standard 20x30" sheet should be about big enough.

-- Cut a thin, vertical slot in the center of the sheet, tall enough to occcupy about 1/4 the frame height.

-- Set up the sheet vertically in front of your digital camera, focus on the slot, and then backlight the slot from behind with a very bright light source such as a quartz lamp. Make sure there's no spill around the sheet.

-- Take test shots at various apertures and with the slot at various positions in the frame. This should enable you to determine such things as:

1) Do streaks appear? If so, what amount of overexposure does it take to provoke them?

2) If you get streaks, do they expand uniformly (blooming) or do they run in one specific direction? (The examples I've seen run in a specific direction; I suspect this is because the data is scanned horizontally off the imager.)

3) Is the streaking independent of shutter speed, or not? (I'm guessing it is, since the electronic scan of the imager is much faster than the mechanical shutter curtain's travel.)

4) Is the streaking constant across the field, or does it fade out toward the far edge? (I'd think it would fade out as the excess signal dissipates, but it's difficult to tell on the examples I've got because the backgrounds aren't uniform.)

5) Is the effect independent of ISO setting, or not? (I don't know, but maybe higher ISOs are more likely to streak since this involves greater amplification of the analog signal off the imager. If using a lower ISO definitely reduces the risk of streaks, this would be useful knowledge!)

6) Do different lenses have any effect? (If it's purely an imager phenomenon, they shouldn't -- but lens flares might be mistaken for this type of streaking.)

7) If you've got several different types of digital camera available, comparing them would be an interesting exercise.

You film diehards, with your symmetrical, streak-free, aesthetically pleasing overexposure effects, have my permission to smirk smugly at this point...

Terao

11-04-2006, 01:30

Alternatively, just stop pixel-peeping and go out and shoot :)

However, if any M8 owner is genuinely unhappy with what they see as their fundamentally flawed digital camera I'll happily take it off their hands for a small handling fee :D

IGMeanwell

11-04-2006, 01:50

Banding can be fixed with firmware ... so I don't think anyone really needs to be too worried

plus I am sure that is a big thing controlled by the 6 bit encoding of the lenses...

but like the D200 many people learn to take it in stride, wait for the company to fix it. Or just take care of it post processing

petermcwerner

11-04-2006, 05:54

jlw

11-04-2006, 06:36

Alternatively, just stop pixel-peeping and go out and shoot :)

Yeah, but if your artistic conception would be spoiled by purple horizontal streaks running through the final image, it might be a good idea to be aware of the problem so you can plan ahead to avoid it.

This is no different from what we've always had to do in photography. For example, if you use a lens with poor flare control, and there are bright light sources in your picture area, you learn either to recompose to avoid them, or to take advantage of the aesthetic effect they produce.

jlw

11-04-2006, 07:06

There is a very interesting in-depth discussion in the Leica-Camera-User Digital Forum: M8 vs DMR, very different colors in Capture One (http://www.leica-camera-user.com/digital-forum/8514-m8-vs-dmr-very-different-colors.html).

It is worth reading it, although it is in part highly technical

Thanks for that link; lots of good info (although you have to skip past a lot of discussion of color balance to find the posts related to horizontal "bleeding" of bright areas.)

Mike Prevette's post provided what seems like a good concise explanation of what's going on:

"The banding however is very hard to correct via firmware. It is a result of the circuitry of the sensor, and how it passes a charge down the row of pixels to the output terminal. It shows up when the sensor has it's gain turned up (high iso) and part of the sensor is exposed to a very bright light source. In order for the sensor to get that pixel data out it has to pass a signal down a row of active pixels to the output port, this causes interference that the other pixels pick up. Hence a band."

As to why people are observing horizontal bleeding with the M8, while most DNR users report they don't observe it, I'm not so sure. One possibility is simply that the DNR people are mis-observing. The effect only shows up when you've got a small, very bright area next to a large, very dark area (it has to be dark for the bleeding to show up.) This is typical of indoor "photojournalism" shots -- the kind of thing that would attract an M8 user -- but not so typical of the types of outdoor and studio photography that seem to be more common subjects for DNR users.

Again, this isn't an M8-specific problem: I've seen it in R-D 1 shots, and people have reported it with various DSLRs as well.

However, there's an interesting speculation in the thread that the M8's offset microlenses may play a role in increasing the problem. This makes sense: the microlenses correct for vignetting by offsetting the incoming light rays, and the amount of offset increases as you get farther from the center of the chip (I base this on Leica's widely published diagram.) In effect, the microlenses are angled so they "look" toward the center of the sensor, and the outward ones "look inward" more than the inner ones.

Now, suppose that a subject includes an area bright enough that one of these little lenses flares a bit. The flare will have only slight effect on the adjacent lenses above, below, and to the inside of it -- because they're all "looking away" from it.

The lens next to it on the outside, however, will be looking toward it, and might easily pick up some of this flare. It, in turn, flares a bit... and, by the same process, passes the flare image along toward the next adjacent lens on the outside, and so on down the row. Result: a strip of horizontal "bleeding" visible in dark areas of the image.

This is all speculation, but it makes sense, no?

jaapv

11-04-2006, 08:36

.
-- Would a CMOS sensor behave differently than a CCD?

My idea is that the blur induced by the AA screen might make this effect less pronounced.

greggebhardt

11-04-2006, 11:08

It IS poop!

73 closely looked at images from my M8 and I see no banding. If this is the only problem being reported, I will be a happy camper.

jaapv

11-04-2006, 11:26

Thanks Greg. Common sense is better than pixel-peeping.

jlw

11-04-2006, 12:55

It IS poop!

73 closely looked at images from my M8 and I see no banding. If this is the only problem being reported, I will be a happy camper.

"I've owned my new house for 73 days and it hasn't burned down yet. Why should I waste money on fire insurance?"

StuartR

11-04-2006, 13:34

Well, this is definitely a real problem. Hopefully they will solve it with a firmware upgrade soon. I had a chance to use the M8 today at PhotoPlus and two of the shots at higher ISO show significant banding, particularly bad at ISO 2500. At low ISO's I do not see it. The camera itself is a joy to use and the overall image quality, banding aside is excellent. The 16-18-21 is also quite an amazing lens. Unfortunately they did not have the finder there for me to check out.

Anyway, here are some of the shots I took.

ISO 2500: pronounced banding from the light sources...both in the lower light that is in line with the man's head, and in the higher lights up above. Not good...
http://www.stuartrichardson.com/m8-banding.jpg

ISO 640: Banding in the man's jacket from the reflection of the bright overhead lights in the glass of the table.
http://www.stuartrichardson.com/m8-banding2.jpg

ISO 320: Very nice image quality. This was taken with the 35/1.4 ASPH.
http://www.stuartrichardson.com/m8-35lux.jpg

StuartR

11-04-2006, 13:36

P.S. The firmware listed in exif is 1.06. I do not know if this is the latest.

John Camp

11-04-2006, 15:17

I took mine out with a Noctilux, shot at 1250 and 2500 with a garage light in one corner, the rest of the photo being of the garage door, to see if I could make it band. I could. I also got some other artifacts (for one thing, a vertical band that neither began nor ended in a bright light; I also got an odd square shape in green opposite the light; it's so geometrical I think it must be some kind of internal lens reflection. Damn. I may have to figure out how to load a photo here.

John Camp

11-04-2006, 16:58

Additional note on banding: I took my worst banding and pretty much fixed it in Photoshop in a minute or two...which would suggest that a firmware fix may be possible, if somebody should need it. I don't think I'd hardly ever need it, but some might.

After spending an hour or so outside shooting garage lights (mostly because I got the camera late in the day and didn't have much light to shoot with) I've come to these conclusions:

There is banding; you get it when you have large very-bright light sources that are extremely over-exposed in photographs that are otherwise correctly exposed. In my experience, it seems that the light has to be large within the frame -- small ultra-bright lights won't do it. And the light has to be really bright RELATIVE to the correct exposure for the rest of the photograph, so that the light is REALLY over-exposed. It doesn't make any difference whether the light is a 100 watt bulb or a 500-watt; it's the relative difference that seems to count.

You can also reduce banding if the lights themselves are slightly out of focus.

There may be some flare involvement, but I'm not sure. I got a couple of weird-looking artifacts when I was pushing the ISO.

On a practical level, it would seem to me that you'd get banding when you are trying to force night-time shots into daylight colors and illumination, but must include head-on shots of very bright lights within the frame. But if you let the night-time shots go darker than daylight-normal (so they look like they're shot at night) then the over-exposure isn't as great and the banding isn't so likely to happen. I should point out that we're really talking about shots that would be impossible with film; the film would simply be fogged in these conditions.

JC

stevenrk

11-04-2006, 17:25

After spending an hour or so outside shooting garage lights (mostly because I got the camera late in the day and didn't have much light to shoot with) I've come to these conclusions:

There is banding; you get it when you have large very-bright light sources that are extremely over-exposed in photographs that are otherwise correctly exposed.

JC

John, curious if you or anyone else has found that there is banding in day time shots -- for example a landscape or street shot where the sun is directly visible or focused light is (such as sunlight through clouds, reflection from a window, etc).

Would be very interested in what anyone might find -- and here you would normaly be working in lower ISOs, although in the nightime shots the lower ISO seems to reduce but not eliminate the problem.

John Camp

11-04-2006, 18:48

Stevenrk,

There are some other shots on this forum (the ones in Central Park) that show bright skies with a range of bright color on the ground, and no banding anywhere that anybody has reported. It seems to take a really peculiar set of circumstances. I took a couple of shots outside my local supermarket, with quite a bit of ambient light around, and no head-on bright lights, and saw no banding there, either. I really do have the feeling that this is showing up at the extremes; I pushed mine so hard that I began getting other kinds of artifacts that I think are lens related, rather than sensor related. Ultimately, I guess, my concern is diminishing...

JC

rvaubel

11-04-2006, 19:07

Stevenrk,

........ I really do have the feeling that this is showing up at the extremes; I pushed mine so hard that I began getting other kinds of artifacts that I think are lens related, rather than sensor related. Ultimately, I guess, my concern is diminishing...

JC

Not mine. Ironically, almost all the shots I have seen posted from PhotoPlus had banding. It appears that all those overhead lights provided almost perfect conditions for banding to breed. But those overhead light conditions are common for the type of photos the M8 should be best at.

Rex

haagen_dazs

11-04-2006, 21:30

check out a ton of banding photos at
http://www.rangefinderforum.com/forums/showthread.php?t=30775
even at iso 320, it bands :(

rvaubel

11-04-2006, 22:18

Haagen_Daaz

I would go out tommorrow and take some regular pictures....after I sobered up.

Although a real concern, I think this banding thing is overblown. I have reviewed all my pictures from way back to Photokina and the problem has always been there but nobody noticed it. Thats not to say that Leica doesn't have a responsibility to correct the problem. Both Canon and Nikon have had similar problems and they addressed them.

Rex

DaveSee

11-04-2006, 22:25

check out a ton of banding photos at
http://www.rangefinderforum.com/forums/showthread.php?t=30775
even at iso 320, it bands :(
All I really see of these shots is poor WB... what's with the Apple banner shifting from violet to magenta? And from these we are to truely know accurate conditions where banding occurs? Hmmm... seems moree like "my experiences with blown highlights and color captures nearly impossible with film". Banding is not the fault of the tool, but its user.

Take a trip to Clay's garage attic elsewhere in this forum.

rgds,
Dave

rvaubel

11-04-2006, 22:40

......... Banding is not the fault of the tool, but its user.........

Dave

What??! That's a truly unique perspective on sensor banding :eek:

Rex

J. Borger

11-04-2006, 22:54

Thanks Greg. Common sense is better than pixel-peeping.
Unfortunately you do not need to pixelpeep to see the issues.
In fact the M8 as it is now is as good as UNUSABLE for low-light photography, where lamps or candles are included (church, bar, concert....)
It does not bother me all that much ... because i am not in that kind of photography ... but there are people who use a Leica especially for this!
The C1 profiles are also NOT GOOD ... it takes me quite some time to tweek the colors. In fact the colors look better if i use any other profile except the one for the M8.
(Auto) white balance is even worse than on the R1-D!. White balance in mixed light is the worst i have ever seen with a camera.
The only situation where the M8 functions as it should is outside in normal daylight. And in that situation it is glorious!
Let's hope it all is firmware related and can be fixed.
JUst wonder if i was really that lucky getting my M8 so quick:( .....

Toby

11-04-2006, 23:41

I'm not really surprised the M8 has some early production issues. They are, after all, a tiny company compared to nikon or canon, trying to produce a pro-spec camera when they don't have a lot of money. The first batch of users are really only paying deposits to be beta testers. The only thing I can't work out is why anyone is surprised by this given Leica's well publicised troubles.

pfogle

11-05-2006, 03:37

I've had this problem once with the R-D1 (I posted the pic on that forum), but never with the Canon CMOS cameras (10D, 20D, 1Dm2 and now 5D).

I think it is specifically a CCD problem, due the the way the pixels are scanned in the electronics, as others have mentioned.

I doubt if there's a firmware fix, but I'd be very happy to be proved wrong. Sorry for the gloomy reply.

jaapv

11-05-2006, 04:10

Unfortunately you do not need to pixelpeep to see the issues.
In fact the M8 as it is now is as good as UNUSABLE for low-light photography, where lamps or candles are included (church, bar, concert....)
It does not bother me all that much ... because i am not in that kind of photography ... but there are people who use a Leica especially for this!
The C1 profiles are also NOT GOOD ... it takes me quite some time to tweek the colors. In fact the colors look better if i use any other profile except the one for the M8.
(Auto) white balance is even worse than on the R1-D!. White balance in mixed light is the worst i have ever seen with a camera.
The only situation where the M8 functions as it should is outside in normal daylight. And in that situation it is glorious!
Let's hope it all is firmware related and can be fixed.
JUst wonder if i was really that lucky getting my M8 so quick:( .....

I just mailed you a dedicated colour profile:) I have not gotten my camera yet, but let me know if it works.

Larry Kellogg

11-05-2006, 05:38

Larry Kellogg

11-05-2006, 06:40

I just mailed you a dedicated colour profile:) I have not gotten my camera yet, but let me know if it works.

Hello,
Could you email your dedicated color profile to me? You can mail it to mac.hive@mac.com

Thanks!

Larry

John Camp

11-05-2006, 09:20

Unfortunately you do not need to pixelpeep to see the issues.
In fact the M8 as it is now is as good as UNUSABLE for low-light photography, where lamps or candles are included (church, bar, concert....)
.....

Doesn't seem that way to me. Because I got my camera late in the day, and had to charge the battery for three hours, I was messing around with it in the dark. I could make it band (and I could mostly clean up the banding without too much trouble) but I took a lot of shots outside in the dark, around a well-lit supermarket, that had no banding whatsoever.

I said someplace else, but I repeat it here, it seems to me that to get banding, the bright light has to be massively over-exposed, and must be relatively large compared to the size of the frame, and, I think, fairly much in-focus. Out-of-focus lights don't seem to band so badly. The shots inside the convention center that show all the banding probably could have avoided it if he hadn't taken pictures of the ultra-bright lights on the ceiling. And ceilings are something that I often don't take pictures of. 8-)

I think, but I'm not sure, that if you were shooting film in, say, a concert setting, where you were in the crowd looking at the front of the band, with the spotlights behind you, you would not get banding, no matter how brightly lit the band was. If you took pictures of the spotlights themselves, you'd get banding. If you took the same shots with film, you'd get fog or light contamination. Some levels of brightness can't be held by anything; as somebody on the Lecia forum said, he can force the 5D to band in extreme conditions. You simply work around it.

JC

J. Borger

11-05-2006, 12:07

Doesn't seem that way to me. Because I got my camera late in the day, and had to charge the battery for three hours, I was messing around with it in the dark. I could make it band (and I could mostly clean up the banding without too much trouble) but I took a lot of shots outside in the dark, around a well-lit supermarket, that had no banding whatsoever.

I said someplace else, but I repeat it here, it seems to me that to get banding, the bright light has to be massively over-exposed, and must be relatively large compared to the size of the frame, and, I think, fairly much in-focus. Out-of-focus lights don't seem to band so badly. The shots inside the convention center that show all the banding probably could have avoided it if he hadn't taken pictures of the ultra-bright lights on the ceiling. And ceilings are something that I often don't take pictures of. 8-)

I think, but I'm not sure, that if you were shooting film in, say, a concert setting, where you were in the crowd looking at the front of the band, with the spotlights behind you, you would not get banding, no matter how brightly lit the band was. If you took pictures of the spotlights themselves, you'd get banding. If you took the same shots with film, you'd get fog or light contamination. Some levels of brightness can't be held by anything; as somebody on the Lecia forum said, he can force the 5D to band in extreme conditions. You simply work around it.

JC

John,

I hope you are right ... i love the M8 ..... and the more i use it the more i like it.
I just was surprised how easy it was to get the banding ... by just shooting 3 candles .... no banding at iso 160 ... but at iso 1250 it is there big time.
Like i said this is no issue for me .. because it is no part of my daily photography ..... and yes the M8 is worth to work around some possible issues. The files are glorious! THe best i have ever seen from any camera!

skimmel

11-05-2006, 12:29

John,

I hope you are right ... i love the M8 ..... and the more i use it the more i like it.
I just was surprised how easy it was to get the banding ... by just shooting 3 candles .... no banding at iso 160 ... but at iso 1250 it is there big time.
Like i said this is no issue for me .. because it is no part of my daily photography ..... and yes the M8 is worth to work around some possible issues. The files are glorious! THe best i have ever seen from any camera!

I love your gallery pics! Do you have any pictures using the M8 up on the M8 gallery? (They may be there and I just didn't see them).

J. Borger

11-05-2006, 12:55

I love your gallery pics! Do you have any pictures using the M8 up on the M8 gallery? (They may be there and I just didn't see them).
Thanks for the nice words ....... not yet .. just testing a bit and getting a feeling for the files!

IGMeanwell

11-05-2006, 13:21

Does it make it worse or better when using the 6 bit coded lenses?

Or is there no difference whatsoever?

skimmel

11-05-2006, 14:14

Thanks for the nice words ....... not yet .. just testing a bit and getting a feeling for the files!

Thanks. I look forward to seeing more of your work.

DaveSee

11-05-2006, 20:26

What??! That's a truly unique perspective on sensor banding :eek:

Rex
Yes, out of the context of my post, this statement seems, ah, dumb. However, I was commenting on a "banding problem" reported in snaps made at an exposition where even the color balance variance among them tossed all of this evidence in the waste bin.

While there is a possibility that banding(sic) may occur, there is--and I and my poor skill atest--much more likely poor user craft contributing to this "problem".[1]

For those who have enjoined the hunt (eg ClayH) for root cause, and to you too Rex, it is especially good to know that some folks see this "banding problem" as an opportunity to correct it... or bother vendors to do so, rather than sit on the fence and pitch stones at passers-by.

On that note:
- which color film(s) do folks use with ISO greater than 640/800 ISO? Not that I am seeking a banding corollary, but whether shooting color with ISO >800 is common amongst us.
- does this banding occur in BW mode?

rgds,
Dave
[1] Added to film emulsions and "grain clumping", lens flare and all those technological problems that make my images worse than they already are ;)

jaapv

11-05-2006, 23:55

I have not seen anybody reporting on the use of the banding tool in Capture One. One might assume that banding can be corrected in RAW conversion.

Ben Z

11-06-2006, 06:52

As a matter of fact ClayH mentioned here that the banding correction in C1 didn't seem to make much difference, and on the LUF I see where he tried turning it to maximum and it didn't help. They're saying over there that Leica has been informed of the issue and is promising a fix is in the works as top priority. Of course there are those also saying Leica must have known about it and shipped the M8s anyway, and some saying a firmware update won't fix it, it will need the sensor to be replaced.

BTW, I know it's heresy not to shoot RAW :D but has anyone shot any .jpg with the M8? Just wondering if some of these issues may be addressed by the in-camera processing algorithms.

Scary stuff anyway, sorry for the guys who have to deal with these issues, and hope this banding thing isn't something that can't be fixed. I wonder if that's the worst-case, will Leica have to give everyone their moneys back.

J. Borger

11-06-2006, 07:32

I have not seen anybody reporting on the use of the banding tool in Capture One. One might assume that banding can be corrected in RAW conversion.
That's because what we are discussing as banding on all forums in fact is not banding at all ........ therefore the specific tool in capture one has no effect.
Everybody calls it banding .. some call it bleeding, sensor striking etc. ........ we all know how it looks, but do not know how to call it ....so what's in a name .. it ads nicely to all the confusion:D

ClayH

11-06-2006, 08:01

Maybe we should amend the phrase and call in streaking or orthogonal highlight bleed. In any case, it is going to need to be fixed.

Ben Z

11-06-2006, 08:32

Ben Z

11-06-2006, 11:14

Here's an interesting reply to someone over on the LUG who has found the banding/streaking problem...maybe good news and worth checking your images to corroborate: http://www.leica-users.org/v33/msg06089.html

jaapv

11-06-2006, 11:28

Clay, have you tried shooting JPEGs, just for the heckuvit? Also, have you discovered any means in CS2 or other post-processing software to eradicate or at least diminsh this phenomenon?

BTW if it isn't the traditional banding that affect(ed) some earlier dslrs like the Canon 1D, maybe it will be successfully fixed with a firmware patch. Here's hoping.

Band Aide (Quantum) seems to resolve the problem in PS

anyway, and some saying a firmware update won't fix it, it will need the sensor to be replaced.

I'm looking forward to that.. Two sensors at the same price- it makes the camera positively cheap :D

truando

11-06-2006, 12:22

supposedly there was a meeting on friday at leica about this problem, and it was declared top priority. We'll get some official news soon, I assume.

Ben Z

11-06-2006, 14:14

Let's hope it's not "soon" in Leica's historical interpretation of the word :D And then let's hope it's good news.

John Camp

11-06-2006, 21:18

Band Aide (Quantum) seems to resolve the problem in PS :D
I tried this, and it seems to make no difference. Maybe I'm using it incorrectly? Just seemed to make everything a little fuzzy, and I could sstill see the bands. Any tips on technique? (I've got CS2 and I did go down to 8-bit.)

JC

willie_901

11-07-2006, 06:58

John,

Any software approach to digital-artifact attenuation or thermal-noise reduction (or grain reduction in film) uses some sort of data averaging algorithm. Whenever you average data, resolution will be lost. The value of Band Aide, and all the tools available for digital-image filtering, is they can achieve an acceptable balance between resolution degradation and artifact suppression.

If the vendor's model for the artifact(s) you wish to suppress does not coincide with the true – but unknown – artifact characteristics in the data, then all you will observe is a loss in resolution.

Band Aide was designed to model banding artifacts in the Nikon D1 at ISO > 800. You may have empirical evidence that the causes of the D1 artifacts and the M8 artifacts are different.

willie

MP Guy

11-07-2006, 07:45

I shot a picture of my office today at iso 1250. the only banding I saw was originating from a bright lamp. Regardless, the banding is there. :(

egpj

11-07-2006, 07:53

I shot a picture of my office today at iso 1250. the only banding I saw was originating from a bright lamp. Regardless, the banding is there. :(

Have you heard if Leica plans to do anything about it?:confused:

jaapv

11-07-2006, 07:58

Have you heard if Leica plans to do anything about it?:confused:
Yes- by chiqua, on the Leica Kunden Forum:
Guten Morgen.
Ich habe soeben mit dem Leica Support telefoniert, da hat sich einiges geklärt.
1. 6-bit Erkennung von Objektiven: Wenn ein Objektiv das keine 6-bit Version ist, an der M8 mit eingeschalteter Objektiverkennung verwendet wird, kann(!) es zu Fehlbelichtungen kommen. Das sagen die Leicas aber nur aus Sicherheit, weil sie es noch nicht genau wissen. Bislang sind noch keine Probleme bekannt geworden. D.h. kommt es zu keinen Fehlbelichtungen und sind die Vignettierungen nicht störend, gibt es keinen Grund das Objektiv ein zu schicken. Super 100 € gespart :-)
2. Mein M8 Absturz. Da konnte der nette Leica-Mann natürlich nichts zu sagen, er hat es weiter gegeben. Interessant war in dem Zusammenhang das er meine Einschätzung, dass die Firmware (1.06) sicherlich nicht die Finale ist, sehr deutlich teilte. Ich schätze, da haben wir recht bald was neues.
3. Banding (oder was es auch immer ist): Ja, da wäre er am Freitag noch in einem Meeting gewesen, das wäre ein "ganz heißes Thema" bei Leica. Hätte höchste Priorität und würde schnellstmöglich gelöst. Auf mein Nachfrage ob das auch Softwareseitig gelöst würde, sagte er, dass es da mehrere Ansätze gebe und man noch prüfe. Aha, das kann ja noch interessant werden.
So, mir reicht das erst mal. Ich werde mich jetzt weiterhin an meiner M8 erfreuen und der Dinge harren die da kommen. Ich glaube nicht, dass irgendwer die M8 zu früh gekauft hat, schlimmstenfalls müssen wir sie einfach noch mal einschicken. Bis dahin haben die einen Spass am Nörgeln und ich Spass am fotografieren. Nix für Ungut ;-)
Schönen Tag noch, Frank

egpj

11-07-2006, 08:06

I guess that is German so with Alta Vista I got the gist of it.

Alta Vista translation

Good mornings. I telephoned just with the Leica support, there some clarified myself. 1. 6-bit recognition of objectives: If an objective is not 6-bit version, at which M8 with switched on objective identification is used, kann(!) to false exposure come. The Leicas says however only from security, because they do not know it exactly yet. So far are still no problems admits become. I.e. it comes to no false exposure and is not not disturbing the vignettierungen, enters it no reason the objective to send. Super 100? saved: -) 2. My M8 crash. There the nice Leica man could say to naturally nothing, he it continued to give. Interesting was in the connection which it my estimate that the firmware (1.06) is surely not the final, divided very clearly. I estimate, there have we quite soon which new. 3. Banding (or which it also always is): Yes, there he would have been on Friday still in a meeting, that would be a "completely hot topic" with Leica. Highest priority would have and was as fast as possible solved. On my demand whether also the in terms of software was solved, said it that it gives several beginnings there and one still examines. Aha, which can become still interesting. So, that is enough to me only times. I will enjoy now further to my M8 and to the things will await to come there. I do not believe that possibly who bought the M8 too early at worst must we it simply still times return. Up to then have the one fun at the Noergeln and I fun to to photograph. Nix for bad; -) Beautiful day still, franc

jaapv

11-07-2006, 08:25

noergeln=grumbling. Other posters in the same thread have even more firm statements by the Leica service department: "the issue will be resolved definitly in the near future."
It is interesting that Leica is talking to its private customers on a personal basis and giving decent information this way instead of going in denial mode or a secretive huddle, like some other camera manufacturers with similar problems did in the past. One advantage of dealing with a small company with a dedicated staff imo.

Ben Z

11-07-2006, 10:19

How soon the cause and a fix can be determined is only one part of the question. The second part is how quickly it can be implemented. If it's a firmware thing and they can put it on their website as a download with instructions for user updating, then it shouldn't be a big hassle. If it isn't something user-fixable, then it could be quite a logistical challenge. AFAIK there is no service personnel or equipment anywhere but at Solms so the cameras would need to be shipped back there, most likely via the regional distributor. That could translate to quite a long wait. I would hope in that case Leica would simply ship out new ones in exchange but wouldn't expect it. At least if it takes 6 months I hope they re-start peoples' 2-yr warranties from that date rather than the purchase date. Some kind of goodwill gesture would be reasonable. Like a spare battery perhaps. They cost $109!

gogopix

11-07-2006, 11:05

you bet they should give us a new battery. $109! wow, and they don't even exist! (at least I couldnt' buy one)

And whie we are at it, why is coding taking so long?

you would think they could also do better on the wide end than a SLOW f4 lens with an ugly viewfinder

and where is the FF?

and autofocus?

at leasdt it should have IS, my hands shake something fierce...

.....

HEY, this 'Noergeln" stuff can be fun! :D

petermcwerner

11-09-2006, 06:01

From the German Digitalforum (http://www.leica-camera-user.com/digitalforum/8753-leica-antworten-zu-6-bit-kodierung-7.html#post88624):

A German user who had sent in yesterday some files with banding was called back today by Leica CS and told the following: (my translation)

1. The problem is known and is not being minimized
2. There will be a solution. This will also apply to those who have already bought the camera.
3. Expect December, might be January.
4. Tendency seems to be sending the camera in to the factory, but this is not yet decided. If so, the fix should be fast. Same day if you can bring in the camera to Solms, previous appointment, others add shipping time.

David Murphy

11-10-2006, 01:41

With the proviso of not being able to download the specific images in question I have a couple of comments. First of all "banding" is probably a read defect revealing what is more commonly called pattern noise in digital imaging *or* an electronic bias which is not being removed from the raw image by electronics/firmware in the camera (or both). Digital imaging sensors should ideally show only a random noise distrubution. Such a systematic read defect indicates to me that either these cameras were either inadequately tested by Leica or perhaps that they were sold with known defects. I think former case is more likely.

The origins of these sorts of problems are sometimes complex and require exensive testing and work to track down, and the fixes not always clear cut. The solutions are not necessarily realized by firmware changes, but sometimes require hardware modifications. The solutions are very specific to the actual causes and if they are addressable by firmware changes the camera electronics must be particularly flexible and reconfigurable.

As to remarks seen here that these are birthing pains that must be ironed out over time, I think there is some merit to that argument. Few digital imaging systems work as well in the field as well as they do in the lab and often fixes based on real-word results are necessary to acheive a satisfactory result. I personally hope the issues are resolved as I want to see Leica making cameras into the future.

While I could be wrong, I have been designing readout electronics for digital sensors for about the last 20 years so this has given me some limited insight into such problems.