bouic49 Posted October 29, 2010 Share #1 Posted October 29, 2010 Advertisement (gone after registration) Again as I am new to the M9. What are the merits/demerits to compressed or uncompressed DNG files? Bouic49 Link to post Share on other sites More sharing options...
Advertisement Posted October 29, 2010 Posted October 29, 2010 Hi bouic49, Take a look here To compress or not to compress. I'm sure you'll find what you were looking for!
Snapper UK Posted October 29, 2010 Share #2 Posted October 29, 2010 The major reason I compress is that it speeds up writing and reduces buffering and potentially missing shots due to the camera 'thinking'. Link to post Share on other sites More sharing options...
ho_co Posted October 30, 2010 Share #3 Posted October 30, 2010 Welcome to the forum! No disrespect intended, but the issue has been covered several times. Use the search function. Basically, the disadvantage of using compression with the M9 is that it costs some dynamic range. As I understand it, he M9's compression is different from that of the M8 in that the M9 writes a blackpoint into the file at different levels depending on ISO. In practice, it's doubtful that you'll ever see a difference; but if there's a disadvantage, it's greatest at base ISO, where you would expect the camera to offer its greatest dynamic range. Now you may sit tight, because Sandy or Michael or Carl or Geoff or a lot of other people who know what they're talking about will soon be along to correct me. That's why I say "welcome to the forum." There's a lot of knowledge here, and a lot of people willing to help. It's a good place to be. Link to post Share on other sites More sharing options...
rob_w Posted October 30, 2010 Share #4 Posted October 30, 2010 It's also possible to have Lightroom perform lossless compression of your DNGs when you import them to disk. Can be very useful if you need to make best use of storage space. Lightroom compression can be performed in addition to Leica's own in-camera compression. Link to post Share on other sites More sharing options...
jan_brittenson Posted October 30, 2010 Share #5 Posted October 30, 2010 The major reason I compress is that it speeds up writing and reduces buffering and potentially missing shots due to the camera 'thinking'. Really? I did some tests and didn't notice any difference whatsoever. Link to post Share on other sites More sharing options...
Wiggi Posted October 30, 2010 Share #6 Posted October 30, 2010 When you use compressed DNGs of the M9 you loose information. You have only 8bit color depth. Maybe you would not see it. But you do not have the full depth of color, when you are retouching the picture. The new compressing algorithm of the S2 should be lossless, Leica says. Link to post Share on other sites More sharing options...
ho_co Posted October 30, 2010 Share #7 Posted October 30, 2010 Advertisement (gone after registration) ... You have only 8bit color depth. ... No, not quite. The full so-called 16-bit space (actually 12 or 14 bits) is still recorded, but packed into a look-up table accessible to 8-bit pointers. You can check the algorithm, it's been published several times. You still have the full "16-bit" space, and when you convert from RAW, you still choose between 8-bit and 16-bit conversion. Effectively, you have 255 buckets (8 bits' worth), but the values contained still range from 0 to 65535 (16 bits' worth). Michael Hußmann wrote an LFI article explaining it. Although the original DNG specification had allowed for the look-up table, Leica was the first to implement it. Adobe hadn't included the unpacking algorithm until Leica did it, so it was a few days before their products supported it. It's "virtually lossless." To my knowledge, no one has ever reported being able to see a difference between the two. There's a demonstrable loss of information, but the 'lost' values are rounded down and saved. Once you understand how the M8's compression works (other threads or Michael's article), you can move on to check the threads on how the M9 has changed that slightly. Compressed files take up about half the space of uncompressed. Compressed and uncompressed files write to card at the same rate. Depending on the speed of your card reader, it may take longer to read the larger, uncompressed files to your computer. Link to post Share on other sites More sharing options...
ho_co Posted October 30, 2010 Share #8 Posted October 30, 2010 It's also possible to have Lightroom perform lossless compression of your DNGs when you import them to disk.... Correct. Lightroom and Camera Raw and DNG Converter all store the M8's compressed DNGs in about half the space they required in camera. I'm sure the same would happen with the M9's compressed files, but I don't know how much space you save with the M9's uncompressed files. In other words, you lose little information with the M9's compression but save on space. And then when you apply Adobe's lossless DNG compression, you'll save yet more space without losing any additional data. Link to post Share on other sites More sharing options...
hoppyman Posted October 31, 2010 Share #9 Posted October 31, 2010 Yep agree with everything Howard said, here. The file size answer is.... a lot! I always use "DNG uncompressed" in camera because I want to preserve everything possible. No, I have never seen an illustration of any difference either but have no problem with keeping the larger versions anyway, why compromise? About 400 frames on a 16GB card. An uncompressed DNG from the M9 is 34.75MB A typical Adobe losslessly compressed file after import here is 19 to 20MB from my samples here. Link to post Share on other sites More sharing options...
henning Posted October 31, 2010 Share #10 Posted October 31, 2010 The time to clear the buffer is noticeably shorter with compressed files. I've done a number of tests, and have not been able to see a difference in the files after I open them in Lightroom, but the speed advantage is definitely noticeable. I have run into the buffer limit quite often, and the faster it clears, the sooner I can get going again. On 'C', after 7 shots the camera can't buffer any more, and if I then remove my finger and wait for the data transfer light to stop blinking it takes 13 sec with compressed DNG and 25 sec with uncompressed DNG. This is with 16Gb class 6 Transcend cards. With class 10 cards it's slightly faster. The size of the file after it's shot doesn't matter to me. Cards are cheap, I have lots of them and hard drives are even cheaper. I'd use uncompressed if I could see a difference, but I can't so compressed it is for the advantage I can notice. Henning Link to post Share on other sites More sharing options...
hoppyman Posted October 31, 2010 Share #11 Posted October 31, 2010 I have not seen that. I very seldom shoot that way but I did some checking Providing I have a sufficiently fast shutter speed to not affect the frame rate, then (on continuous) I get 6 frames before the buffer is full and transfer slows more whether uncompressed or compressed is set. Very interestingly with the same options plus also saving a fine JPG I get 7 frames before the data transfer warning and then the long wait for the slow buffer to clear. Just looking at Henning's comments that appeared while I was replying maybe we are talking about different things, shot to shot rates with number of frames or the complete time to clear the buffer after you stop shooting new frames.. No question that the buffer is slow to clear. Much depends on your shooting style and requirements for sure. This is not the best tool for high frame rates by any means. The major reason I compress is that it speeds up writing and reduces buffering and potentially missing shots due to the camera 'thinking'. Link to post Share on other sites More sharing options...
MarkP Posted October 31, 2010 Share #12 Posted October 31, 2010 This issue has been puzzling me lately as I'm new to this. If no-one has ever seen a real difference between the final two outputs, then the file size on card and computer is THE issue. Therefore why use uncompressed, or is there a difference in the final output. :confused::confused: Link to post Share on other sites More sharing options...
mitchell Posted October 31, 2010 Share #13 Posted October 31, 2010 I don't have a M9, but I thought one of the big advantages of uncompressed files was in processing on the computer, having more "headroom" for adjustments, maintaining smooth transitions of tonality and color. Is this not so? Best, Mitchell Link to post Share on other sites More sharing options...
ho_co Posted October 31, 2010 Share #14 Posted October 31, 2010 Henning, I'm not arguing with your experience, and I don't have an M9, so I'm relying on my memory of the other threads and experiments I've read on the topic. Link to post Share on other sites More sharing options...
hoppyman Posted October 31, 2010 Share #15 Posted October 31, 2010 Mark you decide for yourself on the importance of the file size differences. In theory, there is a difference that may be more apparent with higher ISO files. For technical insight and a lot of other very interesting and useful information, you may like to spend time in Sandy's blog here ChromaSoft Search this forum for discussions But this one is particularly useful http://www.l-camera-forum.com/leica-forum/leica-m9-forum/120983-m9-compressed-dngs-blackpoint-noise-levels.html If you feel the need to understand more of the technical issues. In practice as mentioned no-one has shown any objective proof. Really no biggie and very likely many more important factors in just how much quality you can achieve in your files. Link to post Share on other sites More sharing options...
ho_co Posted October 31, 2010 Share #16 Posted October 31, 2010 Mark, Mitchell-- Nikon has a "virtually lossless" compression available on at least some of its cameras. Google "NEF compression." I've posted the link before, but can't find it now, to a very carefully done comparison of the compressed vs uncompressed NEFs. It's impossible to see any difference between them, but the fact that there is a difference is clearly demonstrated. I don't know the details of Nikon's "virtually lossless" compression (I think they've now given it up to save argument), but I think Leica's may be more sophisticated. A) If you can't see it (and you can't) then don't worry about it. If space is important to you, used compressed. C) If you want "everything the sensor saw" (giving you supposedly more room for manipulation), use uncompressed. Seriously, look up the M8 compression algorithm. It is utter simplicity, but enables saving a file in half the space with no noticeable loss of quality. It is simply brilliant IMHO. I started out in the camp of those incredulous that Leica could claim 16-bit data while packing it into an 8-bit box. I screamed bloody murder and accused a number of people on the forum of not knowing what they were talking about. I don't normally get that bent, and I was wrong. Only if you look it up and try it yourself will you see how it works. If you're worried about losing data, then use uncompressed. The Leica compression does lose data, but it keeps all values in the same ballpark; it doesn't toss them. The loss is not visible, but it's present. The argument about using all the data you've got is a good one. If you've got an 8-bit file in Photoshop and want to make a major adjustment, you will lose less data if you first convert it to 16-bit, do the adjustment, and convert back to 8-bit. (I've seen it demonstrated, but can't tell you how.) BUT THE M8 AND M9 DON'T EVER GO TO 8-BIT DATA. THEIR COMPRESSED FILES STILL COVER THE 16-BIT COLOR SPACE. This is a wonderful achievement by Leica. Stefan Daniel said IIRC in the Reichmann video that the M9 got uncompressed capability simply because "people wanted it." No one can see a difference, and Leica did it just to get us off their tail. Use whichever you want, but first check out how Leica's compression works. You can make a spreadsheet in ten minutes that will show you what happens, and how little adjustment is made. It is simple and effective. I promise you'll be impressed at how simple and effective. My opinion, anyway. Edit: I see Geoff trumped me while I was preparing this cool post. I say, go for it! Link to post Share on other sites More sharing options...
ho_co Posted October 31, 2010 Share #17 Posted October 31, 2010 ... Really no biggie and very likely many more important factors in just how much quality you can achieve in your files. Yeah, Geoff. But for those of us who have nothing else to do, admiring the brilliance behind the thing is still a pretty big deal. Link to post Share on other sites More sharing options...
ho_co Posted October 31, 2010 Share #18 Posted October 31, 2010 Oh, one more thing. Michael Hußmann's LFI article is the best source of information. There's more to it than just how data are manipulated. He covers also the fact that digital storage of photographic data by its nature offers fewer brightness steps as brightness increases. The Leica compression algorithm takes advantage of that fact and automatically tends to hold more shadow detail. Below is his diagram. 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/135371-to-compress-or-not-to-compress/?do=findComment&comment=1487395'>More sharing options...
ho_co Posted October 31, 2010 Share #19 Posted October 31, 2010 Explanations of what the M9 does will make more sense after you understand the M8's compression, which is discussed at: KammaGamma Link to post Share on other sites More sharing options...
henning Posted October 31, 2010 Share #20 Posted October 31, 2010 I have not seen that. I very seldom shoot that way but I did some checkingProviding I have a sufficiently fast shutter speed to not affect the frame rate, then (on continuous) I get 6 frames before the buffer is full and transfer slows more whether uncompressed or compressed is set. Very interestingly with the same options plus also saving a fine JPG I get 7 frames before the data transfer warning and then the long wait for the slow buffer to clear. Just looking at Henning's comments that appeared while I was replying maybe we are talking about different things, shot to shot rates with number of frames or the complete time to clear the buffer after you stop shooting new frames.. No question that the buffer is slow to clear. Much depends on your shooting style and requirements for sure. This is not the best tool for high frame rates by any means. Geoff, I don't shoot like that either, but I do run up against the buffer quite often when I'm shooting single shot. So I did some tests to see what would serve me better. When I use the class 10 Transcend 16Gb cards, I can get 7 shots in a row, and nearly 8 before the buffer is full. With class 6 cards, it's only 7 and a fair wait before the 8th. Card speed makes a difference. Btw, I just tested this again to get the numbers posted. In any case, the speed increase with compression is very helpful, and in my test has no downside. Henning Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.