larry Posted October 23, 2006 Share #101 Posted October 23, 2006 Advertisement (gone after registration) I'd have to respectfully disagree with this statement. If the camera has any pretensions to being a pro-level tool, then its output has to be capable of withstanding the processing that's often required to make a magazine spread or even an online ad look as great as possible.8 bits just isn't enough as a starting-point for that type of post-processing. I'm afraid this really isn't correct. I work in magazine publishing and it's not that simple. Although it's always nice to get a RAW file, a good 8-bit JPEG will work just as well. RAW enables us to adjust the file more precisely for our printer's color space, but if you start with a good JPEG, you'll see absolutely no difference in print. That's because files are down-converted to 8-bit CMYK (from RGB) and most magazines are printed on a web press (lower quality than a sheet press). For high-quality prints, there's no argument that starting with a 16-bit RAW file is better. Larry Link to post Share on other sites More sharing options...
Advertisement Posted October 23, 2006 Posted October 23, 2006 Hi larry, Take a look here M8-why 10MB-vs-DMR 20MB . I'm sure you'll find what you were looking for!
plasticman Posted October 23, 2006 Share #102 Posted October 23, 2006 I'm afraid this really isn't correct. I work in magazine publishing and it's not that simple. Although it's always nice to get a RAW file, a good 8-bit JPEG will work just as well. RAW enables us to adjust the file more precisely for our printer's color space, but if you start with a good JPEG, you'll see absolutely no difference in print. That's because files are down-converted to 8-bit CMYK (from RGB) and most magazines are printed on a web press (lower quality than a sheet press). For high-quality prints, there's no argument that starting with a 16-bit RAW file is better. Larry Maybe i overstated my case there - what i should have written is that a 16-bit file is strongly >preferable< as a starting-point. (I work in advertising btw). The amount of manipulation that can be operated on a 16-bit original is clearly greater than if you start with an 8-bit file - and i'd contend that the M8 should definitely be outputting a file with that greater bit-depth. Every manipulation in Photoshop (or any other image-processing application) is essentially destructive: starting-out with 256 levels just isn't enough imho. Anyone REALLY think that's good enough for the M8? (i have mine pre-ordered incidentally). Link to post Share on other sites More sharing options...
fotografr Posted October 23, 2006 Share #103 Posted October 23, 2006 Thanks for your input, Larry. It's always good to hear from someone on the receiving end of the images. BTW, web saving also down-converts images to 8 bit. For my purposes (editorial photography for magazines) I think the best approach is to shoot RAW and save the RAW file so I can go back to it anytime and work 16 bit if I need a super high quality print for some reason, but to have my normal workflow in 8 bit. I always make it clear to photo editors that I'm happy to provide RAW files if they need them, but I've actually never had anyone take me up on it. The 8 bit files work just fine. I just hate to see anyone base their decision about whether to get the M8 on this factor because it's really a relatively insignificant issue. It would be nice to have the 16 bit option, but it's certainly not going to make or break my M8 purchase. Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #104 Posted October 23, 2006 I just hate to see anyone base their decision about whether to get the M8 on this factor because it's really a relatively insignificant issue. It may not bother you but it definitely matters in many other folks' books ... and 16-bit is what many folks believed the M8 should be. By the way, who would you "hate" to see people having their own opinions? Link to post Share on other sites More sharing options...
fotografr Posted October 23, 2006 Share #105 Posted October 23, 2006 It may not bother you but it definitely matters in many other folks' books ... and 16-bit is what many folks believed the M8 should be. By the way, who would you "hate" to see people having their own opinions? Yes, one thing that is crystal clear in this thread is that many people will be very upset if they don't get 16-bit files from the M8--enough that they will probably not make the purchase. I just won't be one of them. I'm not the least worried about the image quality that will come from this camera. I never in any way implied anything about "hating to see people having their own opinions." I have no idea where you got that from, but if you think I said something like that you are having translation problems. Link to post Share on other sites More sharing options...
johnll Posted October 23, 2006 Share #106 Posted October 23, 2006 In the M8 instruction manual they clearly state that RAW files are 16-bit. Many leading DSLRs (and up-market digicams that have RAW capability) are only 12-bit capture and this is expanded to 16 bits in RAW conversion. It is inconceivable that the M8 should capture only 8 bits. The EXIF data that shows this has presumably been updated along with the conversions (eg to JPG, "for web") that have been applied to the files on their way to your computer screens. Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #107 Posted October 23, 2006 Advertisement (gone after registration) In the M8 instruction manual they clearly state that RAW files are 16-bit. Page number? ... Link to post Share on other sites More sharing options...
EricC Posted October 23, 2006 Share #108 Posted October 23, 2006 Page number? ... It is stated here in the M8 Technical Data pdf http://www.leica-camera.us/assets/file/download.php?filename=file_1236.pdf Data formats DNGTM(Digital Negative Format not specific to any camera manufacturer), 2 different highly compressed JPEG levels. DNGTMfile information16 bit-color resolution, 10.2 Mbyte file size per picture. Eric Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #109 Posted October 23, 2006 It is stated here in the M8 Technical Data pdf http://www.leica-camera.us/assets/file/download.php?filename=file_1236.pdf Data formats DNGTM(Digital Negative Format not specific to any camera manufacturer), 2 different highly compressed JPEG levels. DNGTMfile information16 bit-color resolution, 10.2 Mbyte file size per picture. Hi, Eric ... if you've already read from post number 1 in the thread then you'll find that we've jumped through all the hooplas. Link to post Share on other sites More sharing options...
EricC Posted October 23, 2006 Share #110 Posted October 23, 2006 Hi, Eric ... if you've already read from post number 1 in the thread then you'll find that we've jumped through all the hooplas. Hi Simon, Yes, of course looking back this has already been posted. My apologies. Eric Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #111 Posted October 23, 2006 Hi Simon, Yes, of course looking back this has already been posted. My apologies. You don't need to apologize, Eric ... no one does ... in fact, we're just asking for a definite answer to a simple question - before we kiss our money goodbye. Link to post Share on other sites More sharing options...
EricC Posted October 23, 2006 Share #112 Posted October 23, 2006 You don't need to apologize, Eric ... no one does ... in fact, we're just asking for a definite answer to a simple question - before we kiss our money goodbye. I have just had a look through Leica World News 91505 02/2006, which the dealer i have pre-ordered my M8 from brought back from Photokina. It says on page 12 Quote " The professional RAW data convertorCapture One LE ensures that the data supplied by the CCD sensor and stored in the future-secure Adobe Digital Negative Format (DNG) are processed to yeild the best possible quality. All in all, the 16 bit-image data generate astonishing image resolution, an extended dynamics range and an amazingly wide range of tonal values." End Quote I know it is not the same as having a RAW file to look at but it is here in black and white print. Cheers Eric Link to post Share on other sites More sharing options...
Guest guy_mancuso Posted October 23, 2006 Share #113 Posted October 23, 2006 I hope everyone realizes the Leica guys on the factory floor are ROTFLTAO about this one. rolling on the floor laughing there ass off, just in case. Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #114 Posted October 23, 2006 Hi, Eric ... I've never doubted what Leica has come up with before I saw the question Andrej raised in his original thread. I'd like to see that his observation was incorrect however, before we draw such a conclusion, let's PROVE it. Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #115 Posted October 23, 2006 I hope everyone realizes the Leica guys on the factory floor are ROTFLTAO about this one. rolling on the floor laughing there ass off, just in case. So there seems to be a live broadcast from Solms just for you, Guy ... your passion over Leica surely impresses a lot of folks including me. In fact, your monster thread on FM has saved the company. Link to post Share on other sites More sharing options...
Guest guy_mancuso Posted October 23, 2006 Share #116 Posted October 23, 2006 I'm sitting here watching my customized M8 being built . Have a compass built in case i lose my way with Gps and wireless transmission of files right to this server. LOL Seriously though I understand the pasion both ways on this topic and i know folks want proof of it. I agree but i seriously doubt in the end they are not 16 bit files. They would get sued so fast for advertising it and not being there plus lose face which is even more important to them. Link to post Share on other sites More sharing options...
sdai Posted October 23, 2006 Share #117 Posted October 23, 2006 They would get sued so fast for advertising it and not being there plus lose face which is even more important to them. I highly doubt that anybody could start a lawsuit against them ( just in case this camera is not 16 bit) and I just can't find the 16-bit thing in any official literature regarding the M8, except the reference to "16-bit color resolution" (what the heck is that by the way ... ) in their data sheet. I would still be a happy camper with the M8 no matter what, just that I have to know for sure what I'm buying ... and to some degree, I must agree with some of the posters in this thread that 8-bit is surely good enough for most of the practical use of the camera - damn, for most of the photographs a typical Leica shooter is going to take ... even a Digital Rebel seems to be an overkill. Link to post Share on other sites More sharing options...
Andrej Kolev Posted October 24, 2006 Share #118 Posted October 24, 2006 I thought I've clearly showed that data is not 16 bit in any case. If you dont believe me use dcraw to look what is inside. "10.2 Mbyte file size per picture" - if it is stated in manual it means leica planned current type of coding for final version. The constant size of exactly 10.2 you will get only in one case - if will be used 8-bit coding of any type without any kind of compression lossy or lossless. If compression is used the size will depend of many parameters - iso=noise, detail - scene detail and detail level captured by lens and so on. In any case after linearization you will not get 16 bit data exactly as they stated, you will get maximum value about 2^14. OK you may say with some stretch that it is 16 bit values, but regarding only to a range of values, not to precision. I agree difference in many cases is unnoticeable, but it is a question of relation to consumer, some kind of fooling. Also I think professional or amateur camera differs from P&S by ability of choise. Who wants shoot 8 bit - who wants 16 (if think that it will deliever more quality). It is so simple to make option in camera menu. At least I hope even if final version will use current coding they will issue a firmware update. If I will have some time I'll beter make some futher research, maybe quick modeling of quantization error in case of typical RAW conversion pipeline. ---------------------------------------------------- Regards, Andrej Kolev kolevraw.com Link to post Share on other sites More sharing options...
scott kirkpatrick Posted October 24, 2006 Share #119 Posted October 24, 2006 Andrej, your last post before this made it perfectly clear: the stored data bytes are the square roots of the as-shot bytes, with some bit shifting to preserve significant data as much as possible. The restored data is 15 bits, from squareing the stored bytes and shifting back two places. That's a sensible thing to do, and with care and good raw file development software most of us will probably never see any artifacts. As a benefit we are getting twice as many pictures in the RAW file buffer, faster review speed on the LCD and maybe higher continuous shooting speed. The Leica documents do seem to be misleading. Sean Reid is aware of the situation and will ask Leica what's going on sometime fairly soon. If there's anything to be said about it, I expect he will do so. Sorry you are getting ignored while other preach about what Leica must do, but Leica's just a company with good hard-working engineers, the M8 is just a camera (but apparently a damned good one), and netizens will be netizens. If you want to model things, could you see if dithering by adding low order bits in the stretched upper tones can help prevent banding in the skies and highlights? regards, and good luck with your raw file development package. Does it have a sexy name yet? scott Link to post Share on other sites More sharing options...
Andrej Kolev Posted October 24, 2006 Share #120 Posted October 24, 2006 You're right Scott. I'm also respect Leica and Leica's engeneers, and long history of company, and a quality of M8 ouput is great. People will get the quality buying the M8, and small aspects of coding should not be a reason to change overall opinion about this camera. By the way it's intersting will other companies use nonlinear uncompressed data like sugested by Adobe as DNG option. I think it would be a good option for many cameras that now are writing uncompressed padded data and easy to implement, with a little loss of precision (but I suppose the ability to write linear data must be provided) . Regarding my RAW converter it's only at beggining stage. So I dont have a sexy name yet, noe even a good site design. At this stage I've finished development of major parts - good NR, updated interpolation alg. and support for almost all camers and now I'm starting to put them into program. As for interpolation - output from leica cameras without AA filters is a very good test - most coverters create artifical patterns or jaggies in colors channels. At this moment I divide all interpolation algorithms at two categories - ones with a little probability of artefact creation and a litlle sharpness and ones with a more sharpness and high probabilty of artefacts, especially in R and B channels. I suppose newer versions of my algorithm will give sharp image with good AA features and no artefacts at all. Also in my first post at DPreview I've noticed stranges in tag values describing features of M8 AA filter - it's denominator value is 0 so result far is from spec. range of 0-1.0. In Leica specs is stated that M8 doesnt has AA filter, but DNG tag can not confirm or disprove this statement. The DNG image with girl doesnt contain high-frequency areas so we can not according to it make a conclusions about AA. DMR shows great level of details without AA, but M8 in some way it is a camera of different type of shooting, and maybe leica engeneers decided to not scare this group of customers with aliasing. P.S. I've forgot to say, but Scott reminded me, if you want to get "16 bit-color resolution" and "All in all, the 16 bit-image data generate astonishing image resolution, an extended dynamics range and an amazingly wide range of tonal values" you need to remove '4' from denominator of linearization equation. ----------------------------------------------------- Regards, Andrej Kolev kolevraw.com Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.