 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
| PhotoKB HomeDiscussion Groups Digital Photography Digital PhotoDSLR CamerasZLR CamerasPoint & Shoot CamerasFilm Photography 35 mmLarge FormatMedium formatDarkroomFilm and LabsOther EquipmentPhoto Technique Nature PhotographyPeople PhotographyTechnique GeneralGeneral Photo Topics General TopicsAustralian PhotographyUK PhotographyDirectoryPhoto Clubs |
Photo Forum / Digital Photography / DSLR Cameras / September 2005Interesting Question about D2X sensors
Nikon are going to great lengths to explain how they capture 12 megapixels of data with their new "dual" sensor in the D2X. They are describing their sensor's photo collector points of the newly released Pro D2X cameras as being 2 different types/sizes of collector points... Sigma did the same with their Foveon sensor SD9 cameras. I think because Nikon collect RGB data from each site, they cannot be lumped in with the Foveon "invention" of sensor size. Just the same, the Nikon sensor is different from a traditional bayer sensor as we know it in a Canon camera. How different is not as clear as it should be. We are aware that smaller photo receptors have their own unique problems. Is using two different size receptors in the way inkjet printer makers get nicer looking images by using different size droplets of ink, have Nikon discovered how to make a high resolving sensor in a small size? So here's the question of the day for thinking people: How many sensors are in a D2X Nikon to achieve 12 megapixel images? Careful now... It's a loaded question!  SignatureDouglas, You never really make it on the 'net until you get your own personal Troll. Mine's called Chrlz. Don't feed him, he bites! > Nikon are going to great lengths to explain how they capture 12 > megapixels of data with their new "dual" sensor in the D2X. They are > describing their sensor's photo collector points of the newly released > Pro D2X cameras as being 2 different types/sizes of collector points... Where are they doing that, exactly? > Sigma did the same with their Foveon sensor SD9 cameras. I think because > Nikon collect RGB data from each site, they cannot be lumped in with the > Foveon "invention" of sensor size. Just the same, the Nikon sensor is > different from a traditional bayer sensor as we know it in a Canon > camera. How different is not as clear as it should be. It's a regular Bayer sensor. I don't know what you're talking about. > We are aware that smaller photo receptors have their own unique > problems. Is using two different size receptors in the way inkjet > printer makers get nicer looking images by using different size droplets > of ink, have Nikon discovered how to make a high resolving sensor in a > small size? Why would this have anything to do with a D2x? They aren't using two different sizes of photo receptors. > So here's the question of the day for thinking people: How many sensors > are in a D2X Nikon to achieve 12 megapixel images? Careful now... It's a > loaded question! There's one sensor, with 12.8 million photo detectors. SignatureJeremy | jeremy@exit109.com > There's one sensor, with 12.8 million photo detectors. Jeremy: if you want to get into a long, useless discussion... this is it. Dougie is off his freaking rocker and doesn't believe any published specs... nor common sense... nor appeals to his humanity. Next time he rattles off about the D2X sensor, just walk away! ;) Signaturehttp://www.pbase.com/bcbaird >>There's one sensor, with 12.8 million photo detectors. > [quoted text clipped - 4 lines] > specs... nor common sense... nor appeals to his humanity. Next time he > rattles off about the D2X sensor, just walk away! ; Ah, yes... Another poster so convinced he knows it all, he just presumes his square world is the only one existing. Pity poor old Brian when he gets to the edge. Jeremy is wasting his time... Been in my kill file because of his insults. You'll be the next to go Brian, right along with canongirlie. Prey tell, Brian, how do you (intentionally) manufacturer a sensor with different sized photo receptors? SignatureDouglas, You never really make it on the 'net until you get your own personal Troll. Mine's called Chrlz. Don't feed him, he bites! > Jeremy is wasting his time... Been in my kill file because of his insults. Not wasting at all. I don't answer your questions for *your* benefit, after all. If you want to go on believing that the camera has different sized photo sensors, that's up to you, but I'd hate to see anyone else think your delusions have any merit. > You'll be the next to go Brian, right along with canongirlie. Pity you're missing the answers you wanted when you posted. SignatureJeremy | jeremy@exit109.com > Prey tell, Brian, how do you (intentionally) manufacturer a sensor with > different sized photo receptors? Talk to freaking Fuji about that. As for what you MEAN (how do you make a dual mode CMOS sensor), it's relatively simple given the nature of CMOS technology. You can process the pixel individually so it's pretty easy. 12.8 megapixels uses the ENTIRE sensor. The 6 megapixel crop mode uses a center portion of the sensor. It isn't f.cking rocketry, you ingrate! Signaturehttp://www.pbase.com/bcbaird >> Sigma did the same with their Foveon sensor SD9 cameras. I think because >> Nikon collect RGB data from each site, they cannot be lumped in with the [quoted text clipped - 3 lines] > >It's a regular Bayer sensor. I don't know what you're talking about. I just looked quickly through the latest version of dcraw.c, and it looks pretty normal except for one thing; it seems to have four different color filters. The scaling factors for the sensitivities are pre_mul(n), where n is usually 0 for red, 1 for green, and 2 for blue, with green (1) usually unstated for each camera, assumed to be 1.0. The D2X has pre_mul() values for 0, 1, and 3. Very few cameras have 4 multiplier values. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >it looks pretty normal except for one thing; it seems to have four >different color filters. Perhaps it is like the Sony F828 sensor, which was an RGBE (E=Emerald) sensor, rather than the usual RGBG.. But anyway, this is not related to pixby/Douglas' delusion that the D2X interpolates from 6Mp to 12Mp (grin). As is well known to everyone but him, the sensor uses a high-speed *crop* mode, where it simply does not read data from about half of the sensor to speed it up - at the cost of image size. SNIP > But anyway, this is not related to pixby/Douglas' delusion that > the D2X interpolates from 6Mp to 12Mp (grin). Which is verifyable with a resolution test target, if one has a doubt. Bart > I just looked quickly through the latest version of dcraw.c, and it > looks pretty normal except for one thing; it seems to have four [quoted text clipped - 3 lines] > D2X has pre_mul() values for 0, 1, and 3. Very few cameras have 4 > multiplier values. Is what you're looking at the same as the TIFF CFAPattern? If so, 3 would be cyan. However, according to the metadata in my DNG files, the CFA layout is given as 0,1,1,2 (R/G/G/B), suggesting a standard layout. But if the values you're talking about are something else, I'm not sure what to look for; the difference could be in the color matrices. SignatureJeremy | jeremy@exit109.com >> I just looked quickly through the latest version of dcraw.c, and it >> looks pretty normal except for one thing; it seems to have four [quoted text clipped - 9 lines] >But if the values you're talking about are something else, I'm not sure >what to look for; the difference could be in the color matrices. It could also be that every other green has a different sensitivity. Do you know where I can download sample RAW files? Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > Do you know where I can download sample RAW files? Drop me an email so I'll have a valid address, and I'll get you a couple. I use DNG and delete the NEFs but I can take a couple shots and put the NEFs somewhere for you and send you a download link. SignatureJeremy | jeremy@exit109.com > >Is what you're looking at the same as the TIFF CFAPattern? If so, 3 > >would be cyan. However, according to the metadata in my DNG files, the [quoted text clipped - 5 lines] > > Do you know where I can download sample RAW files? Is it possible these are masked edge pixels and just labeled weird? Signaturehttp://www.pbase.com/bcbaird >> >Is what you're looking at the same as the TIFF CFAPattern? If so, 3 >> >would be cyan. However, according to the metadata in my DNG files, the >> >CFA layout is given as 0,1,1,2 (R/G/G/B), suggesting a standard layout. >> >But if the values you're talking about are something else, I'm not sure >> >what to look for; the difference could be in the color matrices. >> It could also be that every other green has a different sensitivity. >> Do you know where I can download sample RAW files? >Is it possible these are masked edge pixels and just labeled weird? I don't think so. The pre_mul[] values are for light-collecting pixels. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > >Is it possible these are masked edge pixels and just labeled weird? > > I don't think so. The pre_mul[] values are for light-collecting pixels. Strange. If the sensor had an extra (or different) color filter, I'm sure the marketing and PR guys would be trumpeting it. Signaturehttp://www.pbase.com/bcbaird >> >Is it possible these are masked edge pixels and just labeled weird? >> >> I don't think so. The pre_mul[] values are for light-collecting pixels. > >Strange. If the sensor had an extra (or different) color filter, I'm >sure the marketing and PR guys would be trumpeting it. I'd see what's going on with it, but dcraw.c is a bit thick for me. I can write C code, but I am not good at reading other people's, especially when it has grown the way dcraw.c has. What would be a lot easier is just having a D2X NEF of a gretag-Macbeth color checker. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > I'd see what's going on with it, but dcraw.c is a bit thick for me. I > can write C code, but I am not good at reading other people's, > especially when it has grown the way dcraw.c has. Which version of dcraw.c? The latest is 281 (an hour old), and the has a D2X color table entry. Previous versions show direct assignment to pre_mul[0] and pre_mul[2]. But yeah, it is well past the time to hack that program into smaller pieces. It would, however, complicate Coffin's distribution of the program. [snip] > I'd see what's going on with it, but dcraw.c is a bit thick for me. I > can write C code, but I am not good at reading other people's, > especially when it has grown the way dcraw.c has. What would be a lot > easier is just having a D2X NEF of a gretag-Macbeth color checker. http://www.moosepeterson.com/d2x/noise.html -- Barry Pearson http://www.barry.pearson.name/photography/ http://www.birdsandanimals.info/ >[snip] >> I'd see what's going on with it, but dcraw.c is a bit thick for me. I [quoted text clipped - 3 lines] > >http://www.moosepeterson.com/d2x/noise.html Thanks. This is the kind of stuff I like to see; too much speculation based on JPEGs going around. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >Strange. If the sensor had an extra (or different) color filter, I'm >sure the marketing and PR guys would be trumpeting it. They were for the last 6 months or so in every magazine ad I saw. >>If the sensor had an extra (or different) color filter, I'm >>sure the marketing and PR guys would be trumpeting it. > >They were for the last 6 months or so in every magazine ad I saw. Example please, McLeod? I'm genuinely interested. As I mentioned above, it would be no real surprise - Sony built the sensor, and they have already used an RGBE sensor on the F828. But no-one seems to be trumpeting it now on the D2X (inc. Nikon), and the review sites I've seen don't mention it. >>>If the sensor had an extra (or different) color filter, I'm >>>sure the marketing and PR guys would be trumpeting it. [quoted text clipped - 6 lines] >trumpeting it now on the D2X (inc. Nikon), and the review sites I've >seen don't mention it. I'll go back and check the advertising. I couldn't find any info online. I know information is downloaded from the sensor in 4 channels (2 for the green)...but I was sure they were saying they had an extra colour filter as well. > They were for the last 6 months or so in every magazine ad I saw. A different color filter? I know they've been really pumping the CMOS sensor, but I haven't seen anything about them using a different CFA. Signaturehttp://www.pbase.com/bcbaird 1: This is NOT a very interesting question 2: Who gives a f.ck anyway? > Nikon are going to great lengths to explain how they capture 12 megapixels > of data with their new "dual" sensor in the D2X. They are describing their [quoted text clipped - 15 lines] > are in a D2X Nikon to achieve 12 megapixel images? Careful now... It's a > loaded question! > Nikon are going to great lengths to explain how they capture 12 > megapixels of data with their new "dual" sensor in the D2X. They are [quoted text clipped - 16 lines] > are in a D2X Nikon to achieve 12 megapixel images? Careful now... It's a > loaded question! I know a few D2X shooters who couldn't give a rats arse as to how Nikon do whatever they do to get the pics. They simply use the cameras for (gosh!) taking pictures. SignatureSave Photography | Shoot some film today! Email: drop rods and insert surfaces >I know a few D2X shooters who couldn't give a rats arse as to how Nikon >do whatever they do to get the pics. They simply use the cameras for >(gosh!) taking pictures. Why are you participating in a discussion that doesn't interest you? Are you a masochist? A control freak? Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.