farnz Posted March 10, 2007 Share #21 Posted March 10, 2007 Advertisement (gone after registration) Rob, etairporpa erom eb yad gohdnourg t'ndluow tub won tiddeg I ho sdrawkcab yhw dnatsrednu t'nod I diarfa Pete. Link to post Share on other sites More sharing options...
Advertisement Posted March 10, 2007 Posted March 10, 2007 Hi farnz, Take a look here M8's Kodak Sensor. I'm sure you'll find what you were looking for!
Vivek Iyer Posted March 10, 2007 Share #22 Posted March 10, 2007 Yes - and a very simple solution it was: use a smaller sensor to cut out the edge and corner rays, where the problem lies, so one can use a more effective filer. Not that the RD1 filter is that effective, the Epson RD1 is -after the Leica M8, one of the most IR sensitive camera's around. The main difference is that they got spared the Internet hype. Jaap, You can use a telephoto lens with nearly perpendicular rays hitting the sensor, you will still get the magenta from M8 unlike from the Epson R-D1/s. By your assertion, a 75mm focal length lens should create no magenta problems on the M8. Is that so?! Link to post Share on other sites More sharing options...
stunsworth Posted March 10, 2007 Share #23 Posted March 10, 2007 Vivek, I think the point that Jaap was making was that will a smaller sensor and therefore the light rays hitting the sensor at a more perpendicular angle, there could be a thicker, more effective IR filter than that on the M8. Link to post Share on other sites More sharing options...
Vivek Iyer Posted March 10, 2007 Share #24 Posted March 10, 2007 Steve, I take exception to Jaap's "internet hype" characterization. The fact is that none of the Leica reps at photokina had any clues about IR and the M8 (yes, they were asked about it specifically). Leica only clarified the situation after this so called "internet hype". They still have not indicated if the M8 is UV sensitive. The Epson R-D1/s is. Link to post Share on other sites More sharing options...
jaapv Posted March 10, 2007 Share #25 Posted March 10, 2007 Vivek, Im sorry if you take the " internet hype" phrase personally - it was not, as it is a generalization. That the hype was there is beyond doubt, I think we have all seen that. My point is that there are many digital camera's that are IR sensitive, the RD1 more than most, but there are even more sensitive ones out there, the M8 a bit more, but it is a matter of degree.You might be interested to read THIS ARTICLE that adresses this. No camera maker ever mentioned the issue publicly.Yet only Leica was taken to task for it. Jaap, You can use a telephoto lens with nearly perpendicular rays hitting the sensor, you will still get the magenta from M8 unlike from the Epson R-D1/s. By your assertion, a 75mm focal length lens should create no magenta problems on the M8. Is that so?! No Vivek - I never said that. I said that the smaller sensor enabled the designers of the RD1 to use a slightly -but only slightly- more effective IR filter. The focal length has nothing to do with the intensity of IR contamination. It does affect the amount of cyan vignetting created by that IR filter though. Obviously a designer has to allow for the maximum angle the sensor will encounter. It would be a bit strange to have an M8 that only would produce decent images from 50 mm up. Link to post Share on other sites More sharing options...
Vivek Iyer Posted March 10, 2007 Share #26 Posted March 10, 2007 Jaap, No, it wasn't taken personally. Bear in mind that some the most vocal critics (not me) of this magenta madness are the ones who were also helpful about how to rectify the problem as well. The Epson R-D1/s is more sensitive to UV than it is to IR! I use my R-D1s almost exclusively for UV captures. See for example: Shards of Photography - Rangefinder photography (all the way to the bottom). At least on the Epson R-D1s, I can see an angle dependence when it comes to IR sensitivity. I will be shocked if that isn't the case with the M8. Link to post Share on other sites More sharing options...
jaapv Posted March 10, 2007 Share #27 Posted March 10, 2007 Advertisement (gone after registration) I would say that has to do with the IR filter in front of the RD-1 sensor. It is an interference filter, so the effectiveness is incidence-angle dependent. So it is logical that focal length has an effect on IR contamination in that case. The lack of an IR filter in front of the sensor on an M8 eliminates that effect. Link to post Share on other sites More sharing options...
MP3 Posted March 10, 2007 Author Share #28 Posted March 10, 2007 Enjoy all your views here. " 0.5mm... " " current technical limits... " If that is really the case instead of a design negligence, we shall all take it. Otherwise, an M9 is on the pipeline probably, and who knows. Sharing a recent wedding gallery with a Canonite (he moved from Fuji S2 to Nikon D2X to Canon 5D), he commented there is a problem on overall color balance. I have my 486 filters all time on already. You can imagine that is some frustration head on. My Canon friend had got substantial experience on digital color workflow from shooting to printing. You are welcome to view my gallery (the said gallery) at, Zenfolio | Matthew Yip's Zenfolio | Doris & Stephen and give me some comments. It is a wedding and nite banquet last week, all at available light, no flash. Just want to sort out a good workflow for accurate color work. Afterall it is a 5 grand camera. You know, I visited my camera dealer here yesterday, he said an M8 in southern China is priced upto USD 5,800 - 6,000.- while in northern China like Beijing, even upto USD 8,900.- I asked "do you know somebody is going to buy my 3100xxx at these price =) Cheers Matthew Link to post Share on other sites More sharing options...
AlanG Posted March 10, 2007 Share #29 Posted March 10, 2007 By way of some numbers, the angle from the center of the lens exit to the corner for a few camera formats is: Nikon full frame: 24.96 degrees Canon full frame: 26.19 degrees I have been looking at this for a few minutes now and can't understand it. Where does this information come from? Since I can and do use my Nikon lenses on my Canon full frame camera, how can the angle from the center to the corner be different if both cameras have 24x36mm sensors or film? We know it is impossible for the angle from the center exit to the corner to change simply by moving the lens from one body to the other. Is it that Canon and Nikon lenses are somehow different from each other in this regard? How so? Are we talking about wide angle retrofocus lenses of certain focal lenghts? Wouldn't a macro or telephoto lens have a much narrower angle than 25 or 26 degrees? In other words, doesn't this vary from lens to lens due to focal length and design characteristics? Link to post Share on other sites More sharing options...
stunsworth Posted March 10, 2007 Share #30 Posted March 10, 2007 Alan, the Nikon lenses sit further away from the sensor than the Canon ones (so do Leica's) that's why you can use them on the Canon with an adaptor - in effect a piece of metal that moves the lens further away from the Canon mount. Link to post Share on other sites More sharing options...
sandymc Posted March 10, 2007 Share #31 Posted March 10, 2007 Alan, In an SLR design, the restriction is that the back element of any lens cannot foul the mirror, which is why the film to flange distance for any 35mm SLR is approx 45 mm (46.5 for Nikon and 44 for Canon, 46 on OM, 47 on Leica R, etc.). The back element of a lens can be further out than that, e.g, for a classic long lens. But, any lens for a 35mm SLR shorter than 46mm technically has to to be a reverse telephoto, with the back element further out than the focal length. And in practice, in order to make long lenses as physically compact as possible, they all tend to be telephoto lenses. So nearly all SLR lenses have the back element right on the flange. Steve is correct on the adapter - the Canon mount is the closest to the film, so its possible to use an adapter for other lenses on a Cannon SLR, but not the other way round. Historically, one of the Leica M's big advantages was that there was no mirror to worry about, so the film to flange distance could be 27.95mm, and lenses could protrude backwards into the body. So Leica could build simple high performance wide lenses without the compromises that SLR reverse telephoto designs come with. Ironically, in the digital era, that advantage has now become a disadvantage. Sandy Link to post Share on other sites More sharing options...
ho_co Posted March 10, 2007 Share #32 Posted March 10, 2007 In regard to film, remember the days of orthochromatic vs panchromatic? Big breakthrough when the latter arrived. Why do we keep going over this stuff? Because there are people who haven't been watching the issues as long as we have. Why do I get annoyed to see these same issues hashed and unhashed and rehashed? I don't know. Look at the specs for the DMR Kodak sensor, and look at the specs for the M8 Kodak sensor. Leica and Kodak worked together to come up with a thinner coverglass for the M8. It was clear at that stage that the M8's sensor would be receiving about twice the amount of IR as that of the DMR. Surely you can't believe that "no one noticed till we told them!" At the time of Leica's first public admission of the IR sensitivity problem, Leica already had planned for the IR cut filters. They simply hadn't anticipated the degree to which IR would be a difficulty for general photography. (That is, they knew they had greater IR sensitivity than they wanted and they intended to make the filters available, but they had underestimated the importance of the issue. Part of that, as Sean Reid pointed out, arose from the fact that Leica had asked testers not to comment on color rendition since they were still working on that; so no one noticed IR contamination because Leica had said they were still working on the color. The instructions hindered the discovery.) As for why the Photokina reps didn't know about the IR problem, come on! You've got a bunch of salesmen with a hot product. You don't hold a sales meeting and tell them, "Oh, yes, and the camera is overly sensitive to IR and we're going to have some filters to cover for that fact and they'll be expensive and they'll be necessary but we don't want you to talk about this unless you are directly confronted with the issue." No manufacturer tells the salesmen more than they need to know. Fortunately, Leica gives their salesmen much better information than most manufacturers, but if you had your big product intro, would you have told your visible representives about this problem? The design compromises were made in order to make the M8 the instrument it is. Leica underestimated the IR problem but was aware of it and had the solution (filters) planned. This camera is state-of-the-art, an absolutely phenomenal achievement. Yes, there will be an M9, but not till after the R10 comes out. Celebrate the M8 for the accomplishment it is! --HC Link to post Share on other sites More sharing options...
Vivek Iyer Posted March 10, 2007 Share #33 Posted March 10, 2007 I would say that has to do with the IR filter in front of the RD-1 sensor. It is an interference filter, so the effectiveness is incidence-angle dependent. So it is logical that focal length has an effect on IR contamination in that case. The lack of an IR filter in front of the sensor on an M8 eliminates that effect. Jaap, All those filters (Epson R-D1/s, M8, etc) are broadly classified as "interference" filters. If you think it is a dichroic (reflective) filter then that is incorrect. The Epson's filter and the M8's filter are very much alike- both are cyan filters. Epson's is a bit lower end due to the absence of AR coatings on its filter unlike the M8's. Only, the M8's does not match its sensor properly and unlike Epson, Leica have not yet supplied a software that will take care of the magenta/cyan casts effectively. Link to post Share on other sites More sharing options...
Vivek Iyer Posted March 10, 2007 Share #34 Posted March 10, 2007 As for why the Photokina reps didn't know about the IR problem, come on! You've got a bunch of salesmen with a hot product. You don't hold a sales meeting and tell them, "Oh, yes, and the camera is overly sensitive to IR and we're going to have some filters to cover for that fact and they'll be expensive and they'll be necessary but we don't want you to talk about this unless you are directly confronted with the issue." No manufacturer tells the salesmen more than they need to know. Fortunately, Leica gives their salesmen much better information than most manufacturers, but if you had your big product intro, would you have told your visible representives about this problem? --HC I wish this were true! I was SHOCKED to learn they were just oblivious to this. I am not a company that is selling anything. Given the history of Kodak's ITO based CCD sensors and how they were implemented in cameras and digital backs, if I had been a developer, I would have certainly put in more effort to to address the magenta issue. It is truly surprising to see all this "new product", "clever implementation" (what is so clever about it?) characterizations. I suspect it amounts to saying such and such company or a product "sucks". I am all for the success and the longevity of a digital range finder- preferably with an M mount. Link to post Share on other sites More sharing options...
Riley Posted March 10, 2007 Share #35 Posted March 10, 2007 just like to return you to reality for a moment recall the flash metering issue, to exfill that for your memory flash was continually underexposed without the IRcut flash metering was good with the IRcut then there was WB WB was more accurate with the IRcut I think there is more evidence on the side of, this is how it was meant to be than somebody, or a 100 somebodies, totally screwed the engineering spec Link to post Share on other sites More sharing options...
mjh Posted March 10, 2007 Share #36 Posted March 10, 2007 All those filters (Epson R-D1/s, M8, etc) are broadly classified as "interference" filters. By whom? The filter on the M8’s sensor is an absorption filter, not an interference filter. As far as I know, the same holds for the Epson R-D2(s), it’s just that Epson’s filter is a bit thicker. If you think it is a dichroic (reflective) filter then that is incorrect. Indeed. A dichroitic filter is an interference filter. The Epson's filter and the M8's filter are very much alike- both are cyan filters. The M8’s filter is an IR absorbing filter, not a “cyan filter”, even when it also absorbs some red. Leica have not yet supplied a software that will take care of the magenta/cyan casts effectively. There are no magenta or cyan casts – a colour cast is uniform across the whole image and could easily be dealt with by the firmware. But the magenta and other IR-induced colour-shifts depend on the subject and in the general case they can only be eliminated by an IR-blocking interference filter. The red vignetting introduced by interference filters in front of wide-angle lenses, on the other hand, can be corrected by the firmware – and will be, once 1.10 is ready. Link to post Share on other sites More sharing options...
Vivek Iyer Posted March 10, 2007 Share #37 Posted March 10, 2007 By whom? The filter on the M8’s sensor is an absorption filter, not an interference filter. As far as I know, the same holds for the Epson R-D2(s), it’s just that Epson’s filter is a bit thicker. Please look up these terminologies. If the transmitted bandwidth is is long, it is a long-pass filter. Reflective or absorptive does not matter. That is what these are. Indeed. A dichroitic filter is an interference filter. Same as above, mjh. The M8’s filter is an IR absorbing filter, not a “cyan filter”, even when it also absorbs some red. It certainly is a cyan filter and most cyan filters do cut off IR. If in doubt, try an old Kodak Wratten (gel) cyan filter. There are no magenta or cyan casts – a colour cast is uniform across the whole image and could easily be dealt with by the firmware. But the magenta and other IR-induced colour-shifts depend on the subject and in the general case they can only be eliminated by an IR-blocking interference filter. The red vignetting introduced by interference filters in front of wide-angle lenses, on the other hand, can be corrected by the firmware – and will be, once 1.10 is ready. Yes, there is no magenta "cast". There is cyan cast present (Epson R-D1s) when an UV/IR cut filter is used. Link to post Share on other sites More sharing options...
skunkworks Posted March 10, 2007 Share #38 Posted March 10, 2007 IR is part of the EM spectrum (like microwaves and X-rays) but is outside the area we see. It may be that the magenta blacks the M8 sensor "sees" are actually "real" but our eye can't see them that way since we can't see IR. The "fault" is that the sensor sees too well. I've been trying for a while to understand something that this statement implies. It seems to indicate that after being recorded by the sensor the wavelength is being shortened when the data is processed for output; the camera "sees" infrared but then converts it to red for display, hence it becomes visible. If the output is not the equivilent of the input, why isn't this correctable in software? Why is the filter hack the only solution? Link to post Share on other sites More sharing options...
mjh Posted March 10, 2007 Share #39 Posted March 10, 2007 Please look up these terminologies. If the transmitted bandwidth is is long, it is a long-pass filter. Reflective or absorptive does not matter. That is what these are. There are absorption filters (absorbing IR) and dichroitic interference filters (reflecting IR). That’s the terminology as it is commonly used. And for an engineer designing a camera, this difference matters a lot. There is cyan cast present (Epson R-D1s) when an UV/IR cut filter is used. Only if the filter is of the absorption kind, and only if it is a filter in front of the lens. The absorption filter in front of the sensor does also absorb some red and thus favour blue and green, but that only countacts the sensor’s increased sensitivity for red. In other words, the IR absorbing cover glass also corrects for the red cast caused by the CCD. most cyan filters do cut off IR. That’s beside the point. For all intents and purposes, it is an IR-blocking filter: 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/18275-m8s-kodak-sensor/?do=findComment&comment=196196'>More sharing options...
mjh Posted March 10, 2007 Share #40 Posted March 10, 2007 It seems to indicate that after being recorded by the sensor the wavelength is being shortened when the data is processed for output; the camera "sees" infrared but then converts it to red for display, hence it becomes visible. Yes, the sensor sees IR insofar as the photo diodes are sensitive to IR. But it cannot see IR in the sense that IR could be distinguished from red, green, or blue. Of the red, green, and blue filters of the Bayer array, the red and blue filters have a higher IR transmission than the green filters, and thus IR is seen as magenta (red + blue). If the output is not the equivilent of the input, why isn't this correctable in software? Why is the filter hack the only solution? Because the colour shift isn’t uniform. It depends on the subject and is hard to predict. There’s no way telling real magenta from IR-induced magenta that should actually be black. And it’s not just that some blacks are rendered magenta; any colour can and will shift when IR is contaminating the image. For example (first the colours as they are, then as seen by the M8, and finally the Nikon D70 version): 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/18275-m8s-kodak-sensor/?do=findComment&comment=196211'>More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.