jujube Posted December 25, 2006 Share #1  Posted December 25, 2006 Advertisement (gone after registration) I found black horizontal band in several pictures of M8 (firmware 1.09). What are these ?  1st pict. Exposure Time: 1/1502 sec ISO Speed Rating: 1250 Exposure Bias: -1 EV Lens: NOCTILUX 50mm f/1.0 Aperture: f/1.0  2ns pict. Exposure Time: 1/250 sec ISO Speed Rating: 1250 Exposure Bias: -1 EV Lens: NOCTILUX 50mm f/1.0 Aperture: f/1.0  3rd pict. Exposure Time: 1/30s ISO Speed Rating: 640 Exposure Bias: 0 EV Lens: NOCTON classic 40mm F1.4 Aperture: f/1.4 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/12047-black-horizontal-band/?do=findComment&comment=126136'>More sharing options...
khanosu Posted December 25, 2006 Share #2 Â Posted December 25, 2006 I can also reproduce this band on my new M8 v 1.09 (not a fixed one but the one that came form the factory with 1.09 preinstalled). Â The band is caused by bright light sources at the edge of the frame. The band starts form the light source at the edge and precisely ends in the middle of the frame. To reproduce this band you have to have a bright bare bulb light source right at the left edge or right edge of the frame. The position of the light source on the edge is very critical, there is a sweet spot for this to happen; it has to be at and slightly outside the edge. As soon as the light source moves into the frame this band disappears. This band can happen at all ISOs but is more evident at higher ISOs. Also it can happen with an IR cut filter or without. It happns with all lenses I have (28 summicron APO, 50 summicron, 90 summicron APO) When it happens it can look really ugly. It has a greenish bluish color. Â It looks blakish in tungsten lighting but in reality it is bluish greenish. Â I can reproduce the band at will. Â Furrukh Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #3  Posted December 25, 2006 Here is a sequence of four consecutive shots. I just moved the camera slightly between the frames to hit the sweet spot. When the offending light source moves into the frame then the banding disappears, also when the light source is sufficiently outside the source the banding disappears. This was at ISO1250, 50mm Summicron. No Filter.  Forgot to mention in my previous post that the banding occurs in Adobe RAW (Photoshop CS3) as well as C1 Pro.  Furrukh 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/12047-black-horizontal-band/?do=findComment&comment=126148'>More sharing options...
khanosu Posted December 25, 2006 Share #4  Posted December 25, 2006 Here is another particularly ugly example. The bare bulb light source is at the sweet spot on the edge.  If you follow my instructions above and if you have this banding in your camera then you will be able to reproduce it with no problem. So far I have been assuming that my camera is the only defective sample.  Furrukh 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/12047-black-horizontal-band/?do=findComment&comment=126161'>More sharing options...
atufte Posted December 25, 2006 Share #5  Posted December 25, 2006 I also have this horizontal band showing up from time to time, (very seldom) but i was thinking that the IR/CUT 486 filter caused this, if the lightsource hit's the lens/filter from a steep angle...? 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/12047-black-horizontal-band/?do=findComment&comment=126177'>More sharing options...
khanosu Posted December 25, 2006 Share #6 Â Posted December 25, 2006 Hi Alexander, Â I can reproduce it with or without a filter. If you follow my steps outlined in my post above then you shoudl be able to reproduce it at will. It is not a random thing. You do need a bright bare bulb as the offender. Â Best wishes, Â Furrukh Link to post Share on other sites More sharing options...
Venkman Posted December 25, 2006 Share #7 Â Posted December 25, 2006 Advertisement (gone after registration) I can confirm that this happens in the unpatched, 1.06 version as well. Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #8 Â Posted December 25, 2006 Yes Dirk, I think that it was there in the original un-patched version also but was being camouflaged by the original horizontal banding. Now that the new versions have the original horizontal banding fixed this new banding is standing out. Â This banding only happens with the light source is at the sweet spot (see my steps above). It is not random. Â I sent an email to Leica Technical Info USA along with the DNG files and steps to reproduce this new band. They responded and said that they have examined the DNG files and concluded that the camera is defective and needs to be sent to Germany for a fix. Â Again, my camera is a new one with v1.09 (one week old) not an upgraded one. Â Furrukh Link to post Share on other sites More sharing options...
scanner Posted December 25, 2006 Share #9 Â Posted December 25, 2006 This is very good !!! You must collect these images together with your very clear and solid test and send it to Leica customer service. Ask them what to do and if your M 8's can be shipped to Solms. A very concrete situation is very helpfull to solve the problem. Other customers can join you of course and see also the threat from Sean. Â I wish you good luck and I am convinced that there will be an answer and an solution !! Do you have the mailadress from customer sevice ? Â Â Greetings, Rob Link to post Share on other sites More sharing options...
rvaubel Posted December 25, 2006 Share #10  Posted December 25, 2006 Hi Alexander, I can reproduce it with or without a filter. If you follow my steps outlined in my post above then you shoudl be able to reproduce it at will. It is not a random thing. You do need a bright bare bulb as the offender.  Best wishes,  Furrukh  Furrukh  I noticed that in the original banding problem, that the light source had to be overexposed by at least 4 F stops. In other words, the highlights have to be completely blown to show a banding effect or, for the unfixed models, a negative mirror image (green blob)  It looks as if Leica has regionalized the problem to the area arround the sensor that is supposed to function as some sort of black frame gizmo. What I know is there is a perimeter band of pixels around the sensor that does not contribute to the picture directly. The photons falling on this area are used by firmware to somehow do something, but I don't know what. It sounds to me like it is doing something it is not supposed to be doing.  However, the good news is, how often are blown hoghlights falling on that narrow band? Plus it sounds like a firmware fix since this pixel perimeter is supposed to be there. The firmware is just not interprating what to do with blown highlights correctly.  Just my guess  Rex Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #11  Posted December 25, 2006 Hi Rex,  Your explanation is very plausible since there is a narrow sweet spot for the light source to be in. The photons from the source have to be falling almost perpendicular to the perimeter bands of pixel. However Leica Technical Info USA did not think that this is fixable by a firmware fix; they asked me to send the camera back to Germany for a fix. On the other hand this problem might be so new to them (Leica Technical Info USA) they do not know how to respond right now.  For my type of photography this problem is going to be rare but in situations encountered in Yoshinori’s (original poster) picture with stadium bank of lights at the edge it might not be possible to avoid it. The problem also is that if you hit the sweet spot then the banding really gets ugly.  Furrukh Link to post Share on other sites More sharing options...
steve_l Posted December 25, 2006 Share #12 Â Posted December 25, 2006 Can you all coordinate to send files to the same person, so they can start to put together a real explanation for it? Just in case they thought it was an isolated instance and therefore due to a broken M8....but in fact it may be something more in common to all M8's Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #13 Â Posted December 25, 2006 Rob: Yes I have sent the original DNG files to technicalinfo@leicacamerausa.com as well as cs@leica-camera.com . Leica technical info looked at the files and concluded that the camera was defective and needs to be sent to Germany. As steve says, they just assumed that it was an isolated incidence. It is good that the steps I have outlined always reproduce the problem and that it is not some random thing. Â Steve: Good idea. Hope Sean will not kill me for saying this (I am sure he is busy enough) but perhaps he could be this coordinating person? He already knows people to inform at Lecia. Â Furrukh Link to post Share on other sites More sharing options...
scanner Posted December 25, 2006 Share #14 Â Posted December 25, 2006 Dear Furrukh, Â Â Very good indeed and also that you described how it happens and how it is possible to re-call. Important because leica can try to recall this in Solms before and after the repair !!! Wish you good luck and a merry Christmas. Â Â Greetings, Rob (Netherlands) Link to post Share on other sites More sharing options...
rvaubel Posted December 25, 2006 Share #15  Posted December 25, 2006 Hi Rex, Your explanation is very plausible since there is a narrow sweet spot for the light source to be in. The photons from the source have to be falling almost perpendicular to the perimeter bands of pixel. However Leica Technical Info USA did not think that this is fixable by a firmware fix; they asked me to send the camera back to Germany for a fix. On the other hand this problem might be so new to them (Leica Technical Info USA) they do not know how to respond right now.  For my type of photography this problem is going to be rare but in situations encountered in Yoshinori’s (original poster) picture with stadium bank of lights at the edge it might not be possible to avoid it. The problem also is that if you hit the sweet spot then the banding really gets ugly.  Furrukh  Furrukh  I wouldn't send it back just yet for three reasons.  #1 A bird in the hand is worth two in the bush. Your lucky to have the camera in your hand. Anyway, they problem is not that bad. Better than not having the camera at all.  #2 All the banding problems, I think, are fundamentally related to the same issue. Leica has a bit of a learning curve to figure them all out...they are not done yet.  #3 Our dear friend from Eastern Europe may have stumblied onto something with his white balance comment. It is possible that the perimeter pixel band is used to analyse and compute white balance and maybe blown highlights in this region are screwing up the computation. Leica's fixing the white balance issue is more important for many people (not me=RAW) than whats left of the banding issue  In any case, Leica engineering seems to be doing a fairly good job in dealing with the teething problems as they come up. Marketing is another matter. I also thinks it helps alot for willing "beta testers" like yourself to keep on plugging away at the remaining glitches. Your systematic testing got to the issue in a much to be admired scientific fashion.  Rex ..arf Rex Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #16  Posted December 25, 2006 Thanks Rex. I am certainly going to hang on to the camera and not send it back soon. As I said for me it not an issue; for others it may very well be (stadium bank of lights or night scenes were it is hard to avoid bright light sources at the edges). I will also wait to see if it is a wide-spread problem or just effect a few isolated cameras. I am enjoying the camera and it is doing what I am asking it to do; sending it back so soon did nto even enter my mind  Furrukh Link to post Share on other sites More sharing options...
rvaubel Posted December 25, 2006 Share #17  Posted December 25, 2006 Thanks Rex. I am certainly going to hang on to the camera and not send it back soon. As I said for me it not an issue; for others it may very well be (stadium bank of lights or night scenes were it is hard to avoid bright light sources at the edges). I will also wait to see if it is a wide-spread problem or just effect a few isolated cameras. I am enjoying the camera and it is doing what I am asking it to do; sending it back so soon did nto even enter my mind  Furrukh  Furrukh  I'm glad you're keeping the camera.  Do you have problems with what balance? In wonder if the white balance changes when you include a blown highlight in the pixel perimeter? It stands to reason it could if that data is used to figure the white balance. But if it did, it could change the white balance to a value that is not representative of the scene.  Just a thought. If I wasn't still on a waiting list I would be experimenting myself  Rex Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #18 Â Posted December 25, 2006 Hi Rex. No change in white balance in the ones with banding and the ones without. Â To me it looks like the old timing issue. Seems like they fixed the timing for the bulk of the pixels but did something different with the timing for the edge pixels (if this makes any sense). Seems like the old banding problem still lurking in the pixels in the narrow edge? Â Furrukh Link to post Share on other sites More sharing options...
jujube Posted December 25, 2006 Author Share #19 Â Posted December 25, 2006 Thanks all, especially Furrukh. Â I understood that it was defect of my M8. I'll also send my DNG files to Leica (in my case Leica Camera Japan). Link to post Share on other sites More sharing options...
khanosu Posted December 25, 2006 Share #20  Posted December 25, 2006 Here is something very interesting! You can see this band here, although much wider caused by the flame at the edge. This picture is taken from Sean’s thread “Poll for Article – M8 Vertical Banding”   http://www.leica-camera-user.com/digital-forum/11998-poll-article-m8-vertical-band-2.html  The picture was posted by Eric (sorry I did not ask for permission from Eric), Post #35 in that thread.  I recall seeing other pictures posted here which show this broad band caused by a broad light source on the sweet spot at the edge. I will hunt around for them and post them here.  Furrukh 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/12047-black-horizontal-band/?do=findComment&comment=126256'>More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.