ho_co Posted November 8, 2006 Share #141 Posted November 8, 2006 Advertisement (gone after registration) I posted a shot of my RD-1 in IR ... I think it shows that black synthetic fabrics and the purple areas of Steven Kinberg's RD-1 really strongly reflect infrared light. David-- Very good comparison. Thanks for posting it. It demonstrates very well the assumption we've been working on. For those who haven't seen David's illustration, it's at http://www.leica-camera-user.com/digital-forum/8936-rd1-shot-ir-modified-d1x.html. --HC Link to post Share on other sites More sharing options...
Advertisement Posted November 8, 2006 Posted November 8, 2006 Hi ho_co, Take a look here purple RD1 shot with M8. I'm sure you'll find what you were looking for!
innerimager Posted November 8, 2006 Share #142 Posted November 8, 2006 I'm afraid Lloyd is correct, as I have been saying. It will almost certainly require a different sensor coating to address this. Hope we're wrong....Peter Link to post Share on other sites More sharing options...
carstenw Posted November 8, 2006 Share #143 Posted November 8, 2006 Carsten, that is a valid point. I dismissed one number and pulled another one out of thin air. But mine was based on shooting with the M8 for four days. Still an estimate, though. I'll revise my statement to saying that the 99% number is optimistic. I offer no competing estimate, and only assert that the success rate will be smaller than that, and will depend on what you are using the camera to photograph. For me, the problems occurred at a frequency that was unacceptable. Others may be perfectly content. It is a very well made and solid camera. I agree that it is a problem. I think that for many people it is a non-issue. Each person needs to carefully examine the issue though. It could be a 100% issue for wedding photographers, for example. The streaking which I saw with mine was not all that bad, and I really had to work hard to get it. The IR issue is larger for me, but I don't rely on colour accuracy of clothing, so ultimately I can swallow it if needed. I have also had colour accuracy issues with my 5D, so I am trading one issue for another. A little work in Photoshop is acceptable for me, to get the M8. It is still a very special camera, even if it isn't perfect. Others will have to weigh the issue for themselves. Link to post Share on other sites More sharing options...
gogopix Posted November 8, 2006 Share #144 Posted November 8, 2006 The 'magenta' issue is actually not new to the M8. In some of my P25 images a blue backpack no matter what I did came out PURPLE. I went and looked and it is ballistic nylon. The reason fabric is a problem(rather tha just a black matte surface) could be that it acts either as a reflective prism or a fresnel lens. The red is due to the longer wavelength somehow selectively being enhanced. Hence, blue became purple. Of course, that is also a Kodak sensor, and maybe has more sensitivity than Dalsa to these near IR/reds. The first two images (which I will not bore you with) I took (and only so far, have too much other work!) were of my screen with a black border no magenta cast and of the lights in our bathroom, 100 watts from a few feet away, but a dispersed tungsten filament, no banding. I do not take this lightly though, and the fact that there are very good understanding means that these will all likely be fixed. Rather than damning Leica for being 'elitist' we should let them know that the other features of this cameramake it outstanding. One thing was a surprise, and a bit of a disappointment - the shutter IS a bit noisy. Link to post Share on other sites More sharing options...
dhoelscher Posted November 8, 2006 Share #145 Posted November 8, 2006 David--Where is that posted? Sounds like the kind of proof of assumption we need. --HC Howard, It's right here in a new thread entitled "RD1 shot with IR modified D1x." Check out the black ski jacket - it's totally white in IR .... DH Link to post Share on other sites More sharing options...
carstenw Posted November 8, 2006 Share #146 Posted November 8, 2006 One thing was a surprise, and a bit of a disappointment - the shutter IS a bit noisy. I found that although I was surprised by the sound in a quiet room, in normal environments, it is nearly inaudible. I took some pictures in a business-customer-type restaurant, and no one batted an eyelid, even at neighbouring tables. Unfortunately the picture was out of focus, and after I tested a bit, I sent it back for focus adjustment. However, comparing side-by-side with an MP in the store, I found the levels pretty similar. Do you have a film-M to compare to? I find that with my 5D, also not a terribly loud camera, the most obvious sound resulting from taking a picture is not the shutter/mirror, but the winding, which is very piercing. The M8 I find different, with the shutter being an audible *clack*, but the winding being relatively silent. Link to post Share on other sites More sharing options...
roberth Posted November 8, 2006 Share #147 Posted November 8, 2006 Advertisement (gone after registration) Does anyone recall the colour profile view on LL review of the M8, it had strange spikes in the IR spectrum. Those are the changes we are seeing in some black materials. I believe the profile numbers can be pushed around to correct these spikes and would think a new C1 profile will get rid of it. No sensor coating required. Regards, Robert Link to post Share on other sites More sharing options...
carstenw Posted November 8, 2006 Share #148 Posted November 8, 2006 Robert, those spikes represent a sensitivity of the sensor, not the spectrum of some colour profile, as I understood it. In that case, the solution would be more difficult. Link to post Share on other sites More sharing options...
AndyME Posted November 8, 2006 Share #149 Posted November 8, 2006 I think it was already pointed out here: not every black item will look purple, it has to do with the material too. so how should a software know what material the coat is made of? I talked to a Leica support engineer today and actually I confronted him with what happened to me at a test shot in a store. The store was illuminated with fluorescent light and I made a pic of the sales clerk, one with the M8 and one with my EOS1Ds2. The lips of the man came out magenta with the Leica. Absolutely not acceptable...I would love to post the pics here, but i have no photosite...can send email though if requested. The support engineer explained to me the issue with the IR, and the filter, and the polyester...and asked wether this also explains the fluorescent light issue, he said it does.at least partially.....hm. And he said that the camera performs just great in daylight....hm... Link to post Share on other sites More sharing options...
pascal_meheut Posted November 8, 2006 Share #150 Posted November 8, 2006 As we said before, we may be facing 2 problems here even if they are related : 1) some black material are not blacks at all on pictures and this cannot be solved by software 2) under some fluorescent or incandescent lights, the colors are wrong and this can be solved by a different calibration Link to post Share on other sites More sharing options...
llc Posted November 8, 2006 Share #151 Posted November 8, 2006 I'm afraid Lloyd is correct, as I have been saying. It will almost certainly require a different sensor coating to address this. Hope we're wrong....Peter It’s not a sensor coating; it’s the sensor glass itself, which needs a hard cutoff at 720 or 750 nanometers. Lots of wishful thinkers on this thread. The problem is fundamentally impossible to solve with software. Leica will have to take the cameras back, and replace the sensor glass I've written up this issue in more detail in my blog (Nov 8 entry), along with links to the appropriate B+W filters, for those who find filtration via the lens an acceptable solution. Link to post Share on other sites More sharing options...
llc Posted November 8, 2006 Share #152 Posted November 8, 2006 Thomas--The problem is real and clearly the camera didn't get enough testing. But although your argument looks good on the surface, I think it's basically a red herring. The testers of the camera were serious photographers who shot the kinds of things they normally shoot. Almost everyone who has the camera today says that 99% of the time, the camera performs flawlessly. With no disrespect to the people who shot the pictures involved, I doubt that any of the M8 testers took pictures of the top plates of their R-D1s under tungsten illumination. I wish it were so. But I know from direct experience with the Nikon D2h (which also had a weak IR-blocking cover glass) how troublesome this problem can be, especially with pictures of people with fair skin. This rules it out for much portraiture, unless you like the sunburned look and bizarre color in otherwise neutral clothes. Forget using a flash, with its high infrared output. But it's not just flash, it can happen in all sorts of lighting. IMO the problem is very serious, so much so that every digital camera I've purchased since the Nikon D2h (many) I've tested for infrared response immediately. True, it many cases, it won't be a serious problem; subtle color shifts bother few people. But when it's off, it's really off, and can be extremely difficult to deal with. The solution (for now) is filtration. I go into this more in my Nov 8 blog entry Link to post Share on other sites More sharing options...
AndyME Posted November 8, 2006 Share #153 Posted November 8, 2006 "As we said before, we may be facing 2 problems here even if they are related : 1) some black material are not blacks at all on pictures and this cannot be solved by software 2) under some fluorescent or incandescent lights, the colors are wrong and this can be solved by a different calibration" that's quite right. but i am afraid i am not convinced of the practical use of a partial solution. obviously it would improve the situation for many shots, but you are never sure before most "accidental" (not arranged) shots, what kind of material will await you there...so you will always have to live with this uncertainty and I am not sure that this is anything one wants to accept. it is my feeling that the software solution sounds more useful than it would actually be, i think.... Link to post Share on other sites More sharing options...
Guest stevenrk Posted November 9, 2006 Share #154 Posted November 9, 2006 Has anyone yet seen a correlation between lighting temperature and the appearance of this magenta cast? I fully expect various replies from people who will misunderstand why I am asking this question but I want to focus on the replies from people who understand this question at face value. Once I get those replies, I'll move on to the next step of my idea. Cheers, Sean Sean, many thanks again for all your efforts. Here's a shot with a 2 strobe set up. WB is set to flash. FYI, both yesterday's and today's shots were taken using a 35 summilux. Yesterday's at f2 at about a 15th. Today's f9.5 at 125th. I've added in the WB card. Again processed in C1 3.7.6 with no adjustments made and no adjustments in PSCS2. Hope it helps in your search. Best, Steven 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/8734-purple-rd1-shot-with-m8/?do=findComment&comment=87979'>More sharing options...
llc Posted November 9, 2006 Share #155 Posted November 9, 2006 The testers of the camera were serious photographers who shot the kinds of things they normally shoot. Almost everyone who has the camera today says that 99% of the time, the camera performs flawlessly. And with the "wrong" subject, the pictures will have a strong, difficult-to-correct magenta/purple color shift—100% of the time, whether or not the photographer is serious. Link to post Share on other sites More sharing options...
Guest stevenrk Posted November 9, 2006 Share #156 Posted November 9, 2006 Some suggestions about whether different Leica lenses may produce better/worse results. One of the issues that Sean is planning to look in to -- tremendous of him to be doing Leica's job for them. The shots on this thread were taken with a summilux 35 ASPH serial no: 3962***, so post 2000 Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.