sandymc Posted October 22, 2009 Author Share #21 Posted October 22, 2009 Advertisement (gone after registration) Sandy- Thanks once again for the latest update. I will be posting some images later today with 12mm C/V & M9 using your even latest version. Just a question, what settings do you recommend on "preferences" in Cornerfix? Mike, Assuming you're using V1.3.0.2, which is up as of this morning, recommended base settings are: 1. Profile creation: "Multiple Equation" disabled, "Luminance Compensation" enabled, "Bidirectional Correction" enabled. 2. Compression: All settings disabled 3. Color: Both Luma and Chroma = 1 If the profile generated doesn't look good, then regenerate the profile with: a) First try enabling "Multiple equations" If that doesn't work try disabling "Bidirectional Correction" c) If that doesn't work try enabling "Multiple equations" and disabling "Bidirectional Correction" simultaneously. If that doesn't work........umm, take two aspirin and call me in the morning....... Sandy Link to post Share on other sites More sharing options...
Advertisement Posted October 22, 2009 Posted October 22, 2009 Hi sandymc, Take a look here CornerFix 1.3.0.0 available for download. I'm sure you'll find what you were looking for!
michali Posted October 22, 2009 Share #22 Posted October 22, 2009 Mike, Assuming you're using V1.3.0.2, which is up as of this morning, recommended base settings are: 1. Profile creation: "Multiple Equation" disabled, "Luminance Compensation" enabled, "Bidirectional Correction" enabled. 2. Compression: All settings disabled 3. Color: Both Luma and Chroma = 1 If the profile generated doesn't look good, then regenerate the profile with: a) First try enabling "Multiple equations" If that doesn't work try disabling "Bidirectional Correction" c) If that doesn't work try enabling "Multiple equations" and disabling "Bidirectional Correction" simultaneously. If that doesn't work........umm, take two aspirin and call me in the morning....... Sandy More likely a sleeping tablet! Thank you Sandy, until later....... Link to post Share on other sites More sharing options...
sandymc Posted October 22, 2009 Author Share #23 Posted October 22, 2009 Sandy You may remember I did a mock up of the M9 to look at the impact of mounting an FF "sensor" in an M8; I think the accuracy of my measurements was better than 0.5mm and I concluded that to avoid image clipping from the lens throat moulding, Leica was going to have to do one of three things - use a smaller battery, enlarge the casting to make space for the battery or move the lens mount across towards the shutter release end. In the event, they did none of these things and the castings for the M9 are the same as the M8's. The moulding is fractionally different and the intrusion into the lens throat slightly reduced. I'm wondering whether they also moved the sensor itself relative to the lens mount to avoid the clipping from the moulding so that the centre of the sensor is displaced horizontally from the optical centre of the lens/lens mount. Could that explain the asymmetric colour effects you and others are seeing with the 18mm lens? Mark, Yes, it could, and its been discussed a bit on one or two other threads. And in fact I asked both Jono and Tim to take a look down the throats of their M9 and see whether they could see an offset. (They can't see anything, btw, but they also don't think they'd be able to see any such offset.) Initially, I viewed it as unlikely, just because offsetting the sensor will give exactly these kinds of problems, and I think that a deliberate design decision to do so would be just plain stupid, and completely at odds with Leica's design ethos, "commitment to perfection", etc, etc. But there don't seem to be any better explanation for what's going on emerging either. So maybe it is the case. It should be fairly easy to tell in a well equipped lab without disassembly though, especially if someone has some laser measuring kit, so maybe someone will do the measurement. BTW, did you ever get your "sulking" M8 back up and working? Regards, Sandy Link to post Share on other sites More sharing options...
marknorton Posted October 22, 2009 Share #24 Posted October 22, 2009 Sandy, I'm sorry, I missed the discussion in the other threads. I've looked in my M9 and it looks OK, but difficult to judge accurately. Since Tim Ashley and Sean Reid are in contact with the Leica engineers, it would be good if they could ask the question: in the M9, is the lens axis aligned with the centreline of the sensor? I'm pleased to say my sulking M8 has cheered up. I removed a transistor when I was doing the work on the hardware delayed shutter wind and forgot about it! Link to post Share on other sites More sharing options...
carstenw Posted October 22, 2009 Share #25 Posted October 22, 2009 In the event, they did none of these things and the castings for the M9 are the same as the M8's. The moulding is fractionally different and the intrusion into the lens throat slightly reduced. I'm wondering whether they also moved the sensor itself relative to the lens mount to avoid the clipping from the moulding so that the centre of the sensor is displaced horizontally from the optical centre of the lens/lens mount. Could that explain the asymmetric colour effects you and others are seeing with the 18mm lens? I think this is very unlikely. When you measured, did you take into account that the rear element is a lot small than the bayonet, i.e. the image will be projected from well towards the center of the throat? Link to post Share on other sites More sharing options...
sean_reid Posted October 22, 2009 Share #26 Posted October 22, 2009 Sandy, Since Tim Ashley and Sean Reid are in contact with the Leica engineers, it would be good if they could ask the question: in the M9, is the lens axis aligned with the centreline of the sensor? Hi Mark, Done, question sent. Cheers, Sean Link to post Share on other sites More sharing options...
delander † Posted October 22, 2009 Share #27 Posted October 22, 2009 Advertisement (gone after registration) I think this is very unlikely. When you measured, did you take into account that the rear element is a lot small than the bayonet, i.e. the image will be projected from well towards the center of the throat? I also would be amazed if the sensor was offset, surely nothing short of optical suicide for Leica's reputation regarding optics. Looking at the M8/9 throats I wouild estimated that the M9 throat is 5mm wider overall maybe 6mm and the throat has been widened on both sides. Jeff Link to post Share on other sites More sharing options...
carstenw Posted October 22, 2009 Share #28 Posted October 22, 2009 A small bump does remain on both sides, so the view from straight in front is slightly obstructed. The lenses don't see the sensor from straight in front, however, but rather from a closer, more central place. Sadly, the significant bump on the bottom also remains. It doesn't block the sensor, but does prevent me from using my Dual-Range 50 Cron. Link to post Share on other sites More sharing options...
marknorton Posted October 22, 2009 Share #29 Posted October 22, 2009 I think this is very unlikely. When you measured, did you take into account that the rear element is a lot small than the bayonet, i.e. the image will be projected from well towards the center of the throat? Carsten, with my home made sensor carrier plate, I simply created a 24*36mm cut-out based around the same vertical and horizontal centrelines of the cropped sensor. Assuming my meaurements were correct (and I'll go back and re-check them), that caused an image from a long lens to be clipped by the lens throat moulding, which is why I used a 90mm Elmarit-M in the example. A wide-angle lens did not show the same effect because, as you say, the exit pupil is much further into the lens throat and can see "around" the obstruction. My original conclusion was that Leica would make the body wider, or use a smaller battery or move the lens mount (and sensor); we now know they kept the same battery and castings and changed the lens throat moulding. The question now is, did they have to move the sensor without moving the lens mount? If they did, not good. If they didn't, how else can the asymmetric effect with the ultra-wides be explained unless it's some sort of problem with the microlens array? I doubt it's anything to do with the lens centring, Leica goes to extraordinary lengths to centre the lens elements and I'd expect to see a more circular characteristic to the colour effect if this were the cause. The effect on the left hand side of the image, projected onto the right hand side of the sensor viewed from the back of the camera would be consistent with the sensor being to the right of centre. I hope I'm wrong. Link to post Share on other sites More sharing options...
marknorton Posted October 22, 2009 Share #30 Posted October 22, 2009 Hi Mark, Done, question sent. Cheers, Sean Thanks Sean, I hope they will answer, "of course" which eliminates one theory but leaves the banding unexplained. Link to post Share on other sites More sharing options...
Mauribix Posted October 22, 2009 Share #31 Posted October 22, 2009 Carsten, with my home made sensor carrier plate, I simply created a 24*36mm cut-out based around the same vertical and horizontal centrelines of the cropped sensor. Assuming my meaurements were correct (and I'll go back and re-check them), that caused an image from a long lens to be clipped by the lens throat moulding, which is why I used a 90mm Elmarit-M in the example. A wide-angle lens did not show the same effect because, as you say, the exit pupil is much further into the lens throat and can see "around" the obstruction. My original conclusion was that Leica would make the body wider, or use a smaller battery or move the lens mount (and sensor); we now know they kept the same battery and castings and changed the lens throat moulding. The question now is, did they have to move the sensor without moving the lens mount? If they did, not good. If they didn't, how else can the asymmetric effect with the ultra-wides be explained unless it's some sort of problem with the microlens array? I doubt it's anything to do with the lens centring, Leica goes to extraordinary lengths to centre the lens elements and I'd expect to see a more circular characteristic to the colour effect if this were the cause. The effect on the left hand side of the image, projected onto the right hand side of the sensor viewed from the back of the camera would be consistent with the sensor being to the right of centre. I hope I'm wrong. Mark, your conclusions make sense to me. The fact that today my dealer (I alread had my M9) said me that Leica is not delivering further cameras because of a "some sort of a sensor issue" is simply just scary. Link to post Share on other sites More sharing options...
carstenw Posted October 22, 2009 Share #32 Posted October 22, 2009 Carsten, with my home made sensor carrier plate, I simply created a 24*36mm cut-out based around the same vertical and horizontal centrelines of the cropped sensor. Assuming my meaurements were correct (and I'll go back and re-check them), that caused an image from a long lens to be clipped by the lens throat moulding, which is why I used a 90mm Elmarit-M in the example. A wide-angle lens did not show the same effect because, as you say, the exit pupil is much further into the lens throat and can see "around" the obstruction. The bump is smaller in the M9 though, I am pretty sure, and is now on both sides equally. Wouldn't that be enough? I am not going to take my M8 apart and put in a fake sensor to check, but looking at the back of my 75 Lux and 90 Cron ASPH, I would guess that they might make it without hitting, especially once you subtract the part of the projected image which fits within the rectangular 35mm format. It is also possible that the aperture construction means that not every last fraction of a millimeter of the rear element is used. In fact, I would be surprised if Leica didn't make these calculations already before making the M8, to get an idea of how much room they would have for the electronics in a FF camera. Link to post Share on other sites More sharing options...
george_b Posted October 22, 2009 Share #33 Posted October 22, 2009 . . . my dealer (I alread had my M9) said me that Leica is not delivering further cameras because of a "some sort of a sensor issue" is simply just scary. Eh? So when will this change? Link to post Share on other sites More sharing options...
Mauribix Posted October 22, 2009 Share #34 Posted October 22, 2009 Eh? So when will this change? Actually I don't know. It may be just rumour as well, call it vapourware. They said within a month. It would be nice to know if someone, right now, is getting is M9 order delivered... Link to post Share on other sites More sharing options...
george_b Posted October 22, 2009 Share #35 Posted October 22, 2009 Actually I don't know. It may be just rumour as well, call it vapourware.They said within a month. It would be nice to know if someone, right now, is getting is M9 order delivered... Very curious since currently there seems to be a complete and utter dearth of them in the UK. Link to post Share on other sites More sharing options...
michali Posted October 22, 2009 Share #36 Posted October 22, 2009 Sandy- Bingo! You guys are geniuses! I ran one of the problem test shots done with the 12mm C/V & M9 through your latest version of Cornerfix with settings as per your instructions earlier. I feel I haven't made the correct profile for the 12mm, so there's still room for improvement from my side. Before Cornerfix 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! After Cornerfix Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! After Cornerfix ' data-webShareUrl='https://www.l-camera-forum.com/topic/100856-cornerfix-1300-available-for-download/?do=findComment&comment=1085676'>More sharing options...
carstenw Posted October 22, 2009 Share #37 Posted October 22, 2009 Whoa, over-correction. Pipette that top-left corner; there is too much magenta. You need a new profile before continuing. Link to post Share on other sites More sharing options...
UliWer Posted October 25, 2009 Share #38 Posted October 25, 2009 I now tested the last version of CornerFix on files of photos taken with the 3.8/18 Super-Elmar and the M8 with "lens detection on - UV/IR". After trying some profiles for CornerFix I found out that the profile with "lens detection on" but without UV/IR filter works best. Here an example of a strong "red shift" - regular on both edges: 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! Now the corrected version from CornerFix: I think sandymc has done a great job and we have to be very grateful for his responsiveness and software craftmanship! I hope the people who do the Leica software will take him as a paragon. Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! Now the corrected version from CornerFix: I think sandymc has done a great job and we have to be very grateful for his responsiveness and software craftmanship! I hope the people who do the Leica software will take him as a paragon. ' data-webShareUrl='https://www.l-camera-forum.com/topic/100856-cornerfix-1300-available-for-download/?do=findComment&comment=1088884'>More sharing options...
michali Posted October 25, 2009 Share #39 Posted October 25, 2009 I think sandymc has done a great job and we have to be very grateful for his responsiveness and software craftmanship! I hope the people who do the Leica software will take him as a paragon. I fully agree! He's done us all a huge favor. Link to post Share on other sites More sharing options...
UliWer Posted October 25, 2009 Share #40 Posted October 25, 2009 Sometimes my CornerFix profile with "lens detection on" without UV/IR leads to a cyan shift in the edges, so the profile with "lens detection on with UV/IR" gives better results: 1. No correction with CornerFix: 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! 2. CornerFix profile with "lens detection on, without filter": 3. CornerFix profile with "lens detection on With UV/IR" Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! 2. CornerFix profile with "lens detection on, without filter": 3. CornerFix profile with "lens detection on With UV/IR" ' data-webShareUrl='https://www.l-camera-forum.com/topic/100856-cornerfix-1300-available-for-download/?do=findComment&comment=1088927'>More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.