Stealth3kpl Posted December 19, 2014 Share #1 Posted December 19, 2014 Advertisement (gone after registration) What, again? Film still seems to be hanging in there, even 35mm doing surprisingly well. Comparing the Image Quality of Film and Digital Pete 10 Quote Link to post Share on other sites More sharing options...
Advertisement Posted December 19, 2014 Posted December 19, 2014 Hi Stealth3kpl, Take a look here Film Vs Digital. I'm sure you'll find what you were looking for!
Doc Henry Posted December 19, 2014 Share #2 Posted December 19, 2014 (edited) Pete thanks for the link , very interesting. It's interesting first for color comparion secondly for the definition and general rendering when you look closely crops Their conclusion is also interesting : "Film still has a lot to offer, especially with the price of very high quality cameras so low. Using high resolution black and white film is well documented these days (although you have to process them yourself) and the latest version of slide and negative color film are stunning. Portra has been reformulated for scanning and has immense dynamic range and Fuji Provia is one of the highest resolving slide films ever made. I agree with these three photographers "Long life to film" Best Henry Edited December 19, 2014 by Doc Henry 4 Quote Link to post Share on other sites More sharing options...
philipus Posted December 19, 2014 Share #3 Posted December 19, 2014 Very interesting test, thank you for posting it Pete. I was particularly surprised by how the 5D2 made berries disappear. There's a K64 35mm image on the linked page which is also really quite impressive in in terms of resolution. 1 Quote Link to post Share on other sites More sharing options...
plasticman Posted December 19, 2014 Share #4 Posted December 19, 2014 Pete - they forgot to compare drone shots. 3 Quote Link to post Share on other sites More sharing options...
zlatkob Posted December 20, 2014 Share #5 Posted December 20, 2014 As the linked article states, a very high quality scan of a high quality 35mm slide film has a digital equivalent resolution of between 12MP and 24MP. The disappearing berries example compares 4x5" Velvia and full-frame digital (Canon 5D2). 4x5" sheet film has an area about 15X the size of full-frame digital. It's not surprising that it offers something more. 2 Quote Link to post Share on other sites More sharing options...
plasticman Posted December 20, 2014 Share #6 Posted December 20, 2014 As the linked article states, a very high quality scan of a high quality 35mm slide film has a digital equivalent resolution of between 12MP and 24MP. The disappearing berries example compares 4x5" Velvia and full-frame digital (Canon 5D2). 4x5" sheet film has an area about 15X the size of full-frame digital. It's not surprising that it offers something more. I thought the article very carefully documented the terms of each comparison. For instance, that the 4x5 berry image was downsampled to match the resolution of the 5Dii, and the purpose of that particular comparison was specifically to show the limits in color resolution of a Bayer sensor. The original 4x5 would obviously have offered 'something more' to a much higher degree. 3 Quote Link to post Share on other sites More sharing options...
hamey Posted December 20, 2014 Share #7 Posted December 20, 2014 Advertisement (gone after registration) I recently travelled to Japan with both my M7 and the M240, I used the Fuji Provia 100 slide films in the M7. All I can say ( IMHO) the images I got from my M7 are superior to the M240. Slide Film is so much nicer. LONG LIVE FILM. Ken. 11 Quote Link to post Share on other sites More sharing options...
250swb Posted December 20, 2014 Share #8 Posted December 20, 2014 I fully agree with everything in the article. I use 6x9 and 4x5 more than 35mm or FF digital, and sold my M240 body because it just didn't cut the mustard, not just in resolution (but I'm not obsessive about resolution), but because the images had none of the tonal depth or subtle colour of film. I suspect that if they had tested the Monochrom the FF digital result would have been different and way up there with the Mamiya 7 (but only in B&W of course), but even so, the Monochrom has a much narrower DR than a good film, and with none of the charm of film when highlights do blow or shadows block. Steve 3 Quote Link to post Share on other sites More sharing options...
jaapv Posted December 20, 2014 Share #9 Posted December 20, 2014 I'm just wondering: if the conclusion of the article linked is My conclusion? It’s a great time to be using film AND digital!,why should the title of this thread be : "Film VS Digital"? 1 Quote Link to post Share on other sites More sharing options...
philipus Posted December 20, 2014 Share #10 Posted December 20, 2014 The disappearing berries example compares 4x5" Velvia and full-frame digital (Canon 5D2). 4x5" sheet film has an area about 15X the size of full-frame digital. It's not surprising that it offers something more. The berries would be there in a 35mm image. I was merely surprised that the pixel arrangement on the sensor would result in lost berries. 1 Quote Link to post Share on other sites More sharing options...
zlatkob Posted December 20, 2014 Share #11 Posted December 20, 2014 I thought the article very carefully documented the terms of each comparison. For instance, that the 4x5 berry image was downsampled to match the resolution of the 5Dii, and the purpose of that particular comparison was specifically to show the limits in color resolution of a Bayer sensor. The original 4x5 would obviously have offered 'something more' to a much higher degree. Just saying it's no surprise. It proves that single-pixel red berries can get lost in a 5D2 image, but won't get lost in an image that starts out 15X bigger. The 5D2's color sensor has a threshold that's sometimes not met by a single-pixel object. If something like single-pixel berries matter to an image, then one should absolutely be shooting 4x5 sheet film. However, single pixels are near meaningless in an image one would typically print from 35mm. A single-pixel berry is on the level of a "hot" pixel -- which camera makers map out of an image without any noticeable loss in image quality. Photographers don't seem to worry about losing them. Apart from the loss of the single-pixel red berries, the 200% crops in the article look remarkably similar, so the difference would be trivial at a typical print size for 35mm, say up to 12x18 inches. 4x5 would have a big advantage for larger print sizes. But it should, having 15X the capture area. A scan of 4x5 must yield hundreds of megapixels -- compared to 21mp for the 5D2. Quote Link to post Share on other sites More sharing options...
plasticman Posted December 20, 2014 Share #12 Posted December 20, 2014 Apart from the loss of the single-pixel red berries, the 200% crops in the article look remarkably similar, so the difference would be trivial at a typical print size for 35mm, say up to 12x18 inches... Sorry to labor the point but it seems you might have missed that they downsampled the 4x5 image to match the resolution of the 5Dii for the purposes of the comparison. Quote Link to post Share on other sites More sharing options...
NB23 Posted December 20, 2014 Share #13 Posted December 20, 2014 Real sex vs. Cybersex, anyone? A fine wet print will always disintegrate an inkjet print on so many levels. 2 Quote Link to post Share on other sites More sharing options...
sblitz Posted December 20, 2014 Share #14 Posted December 20, 2014 I think one difference to note between film and digital is that there is a lot of steps from shutter snap to print or, for msny of us, looming at the film digitized. Along the way, many aspects of film can degrade, scratches, dust, poor scans, etc. nevertheless, I prefer film without question, but there many times when digital is better easier faster. Nice to have both Quote Link to post Share on other sites More sharing options...
zlatkob Posted December 20, 2014 Share #15 Posted December 20, 2014 (edited) Sorry to labor the point but it seems you might have missed that they downsampled the 4x5 image to match the resolution of the 5Dii for the purposes of the comparison. Missed what exactly? Yes, they downsampled it to match the resolution of the 5D2 (and then blew both up to 200%). So, downsampled to 21mp, each would make a 12x18" or 13x19" print at the same 21mp resolution (assuming 300 pixels per inch). I suspect that someone who goes through the time, trouble and expense of shooting 4x5 is likely doing so for the look of film and the look of 4x5, more so than to capture the single-pixel berries in 12x18" or 13x19" prints. If someone needs those single-pixel berries to show up in a very large print, they should obviously be shooting 4x5, not a 5D2. Scanned 4x5 offers hundreds of megapixels for very large prints. The 5D2 is just 21mp and was never intended to match 4x5 slide film. Look for a moment at how large that print is where the single-pixel berries matter. They are showing the 5D2 crop at 200%. If your display resolution is around 100ppi, then a 100% crop represents a 56" print from the 5D2 (max resolution 5616 x 3744px). A 200% crop represents a 112" print from the 5D2. So the 200% crop they're showing appears on screen in the size it would be in a print that is 9.33 feet wide. Their 200% crop proves that a 5D2 capture of single-pixel berries will not appear in a print that is 9.33 feet wide, but will appear if the print is made from 4x5. Obvious win for the 15X larger 4x5. But if someone is going to print 9.33 feet wide and needs every single-pixel berry to appear in their 9.33 foot print, why the heck would they be shooting a 5D2? And who makes 9.33-foot prints where single-pixel objects matter? The crop is basically the extreme of extremes — taking the 5D2 to where it was never intended to perform. If you print from the 5D2 at 12x18" or 13x19", you are getting 300 pixels per inch, or 90,000 pixels per square inch. Is the loss of single-pixel berries among those 90,000 pixels per square inch going to spoil a photo? Edited December 20, 2014 by zlatkob Quote Link to post Share on other sites More sharing options...
plasticman Posted December 20, 2014 Share #16 Posted December 20, 2014 Is the loss of single-pixel berries among those 90,000 pixels per square inch going to spoil a photo? In my opinion you've missed the point of the test: it wasn't intended to prove that shooting berry bushes (how often does anyone do that?) requires a large format film camera, it was a general test of color resolution. The same color averaging is happening when you shoot a portrait, or a cat. The test wasn't about a missed berry in 90,000 pixels, it was about what happens to color detail and nuances. The berries were just a graphic representation. Anyway, I sense this isn't a productive discussion. 3 Quote Link to post Share on other sites More sharing options...
zlatkob Posted December 20, 2014 Share #17 Posted December 20, 2014 The same color averaging is happening when you shoot a portrait, or a cat. And my point is, does it matter? When you shoot a portrait or a cat or anything, does that color averaging matter? If it doesn't matter for a missed single-pixel berry in 90,000 pixels, why would it matter for a portrait or a cat or anything else? Quote Link to post Share on other sites More sharing options...
Steve Ash Posted December 20, 2014 Share #18 Posted December 20, 2014 If it does not matter at all, why would anyone get an 5d II instead of a simple P&S? Quote Link to post Share on other sites More sharing options...
zlatkob Posted December 20, 2014 Share #19 Posted December 20, 2014 If it does not matter at all, why would anyone get an 5d II instead of a simple P&S? The "it" we're talking about is the representation of single-pixel objects in prints that are some 9.3 feet (almost 3 meters) in size. If that matters to someone, then it matters. The 200% crops prove it. But there are so many other factors that matter more and that go into the selection of a format, from cost to portability to color to lenses to noise/grain, etc., etc. The 200% crops focus on something that is pretty trivial among all of those factors. It is extreme pixel-peeping, designed to show where the 5D2 fails, which is not hard to do in such a large print. Every format/system can be shown to fail at some point. When Oscar Barnack thought about a format that would replace 13x18cm glass plates, he decided on 24x36mm negatives, with a goal of enlarging to make 13x18cm prints, or preferably 18x24cm prints. That's just under 10 inches. A century later, someone compares 35mm digital format to 4x5" sheet film for a 9-foot print and, not surprisingly, finds that it falls short. Quote Link to post Share on other sites More sharing options...
philipus Posted December 20, 2014 Share #20 Posted December 20, 2014 If someone needs those single-pixel berries to show up in a very large print, they should obviously be shooting 4x5, not a 5D2. The test should have been done between the 5D2 and 35mm film, of course. And in the latter, I believe the berries would have shown up. I find it interesting (and this is not at all aimed at zlatkob but at the comments discussion in the article) that in the occasional test where film performs "better" than digital, those more in favour of digital are not interested in the pixel-peeping which is so oftern legio in the digital domain. I like berries though so find it sad that they disappeared in their digital incarnation. Philip 4 Quote Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.