tuanvo1982 Posted February 1, 2012 Share #1 Posted February 1, 2012 Advertisement (gone after registration) I read review they said the effective resolution of Nikon V is 3900dpi so should I scan at 3900 or 4000dpi. Thanks Link to post Share on other sites More sharing options...
Advertisement Posted February 1, 2012 Posted February 1, 2012 Hi tuanvo1982, Take a look here Nikon coolscan V (should I scan @3900dpi or 4000dpi). I'm sure you'll find what you were looking for!
andybarton Posted February 1, 2012 Share #2 Posted February 1, 2012 I scan everything at 4000 dpi. Never had a problem. Link to post Share on other sites More sharing options...
pico Posted February 2, 2012 Share #3 Posted February 2, 2012 I scan everything at 4000 dpi. Never had a problem. I'd not expect the scanner software to jump up and say, "Hey, 4000 is just a marketing thing. I'll do what I want, and you can figure out at your leisure what I am really doing." It does not do 4000ppi. Ever. There are no meaningful pixels captured at that resolution. . Link to post Share on other sites More sharing options...
Michael Hiles Posted February 2, 2012 Share #4 Posted February 2, 2012 It does not do 4000ppi. Ever. There are no meaningful pixels captured at that resolution. I am curious - who do you know that? I'm not arguing - I don't know, but I'm interested in the basis for the statement. Link to post Share on other sites More sharing options...
leicamann Posted February 2, 2012 Share #5 Posted February 2, 2012 I use Ed Hammericks " Vuescan"...I think we are at version 9.079.. I set preview at 400dpi and scan at 4000dpi...... the general rule is get as many pixels as you possible can Of course that means large files, thats where it helps to have a beefy cpu, 64 bit OS, lots of RAM and if you can afford it an SSD hard drive with the latest version of Photoshop. I scan RAW, TIFF files up to 2gigs in 16 bit resolution.... Cheers, JRM Link to post Share on other sites More sharing options...
adan Posted February 2, 2012 Share #6 Posted February 2, 2012 Classic internet thread development here - nothing but opinions unsupported by details and facts. Real scanner resolution depends on at least three factors - the width of the scanning CCD rows, how many pixels the scanner steps through over an inch (or mm), and how many lines the scanner lens can deliver to the scanning CCDs. The Nikon V, 5000ED, and 9000ED all move the CCDs 4000 steps per inch. They do not have the same quality of lens, however (the V lens is not ED, and various reviews say the 9000ED lens resolves more than the 5000ED lens). Scanning a piece of film out of focus may generate a file with 4000 pixels per inch - but not a lot of resolution. A weaker-quality lens can have the same effect. More data than information. My Epson 3200 claims 3200 ppi - it is obvious from results that while the scanner takes 3200 readings per inch, the lens in that antique can't deliver more than about 1800 lines per inch. Above that, one is just getting bigger and bigger images of blur. However, it doesn't hurt to oversample. Another point, which may be a source of further confusion. 35mm images are 24mm wide, and 24mm is only 0.945 inches. So even with a perfect lens and perfect scanner precision, a 35mm image scanned at 4000 pixels per INCH will produce a file only 3780 pixels tall/wide in the shorter dimension. (0.945 x 4000). Link to post Share on other sites More sharing options...
andybarton Posted February 2, 2012 Share #7 Posted February 2, 2012 Advertisement (gone after registration) Frankly, I don't care. I scan at 4000 pixels and I get nice scans. That's all that matters to me. Link to post Share on other sites More sharing options...
tuanvo1982 Posted February 2, 2012 Author Share #8 Posted February 2, 2012 I am curious - who do you know that? I'm not arguing - I don't know, but I'm interested in the basis for the statement. it was from here Nikon Coolscan 5 V ED LS-50: Test report, review, field report, experiences, resolution, image quality: film scanner, slides negatives As I performed a resolution test with a USAF-1951 test chart I could not see any difference between the Nikon 5000 and the Nikon V. The Nikon V displays element 6.2, which complies with a resolution of 3650dpi, as clearly as the Nikon 5000. When looking at element 6.3, which complies with a resolution of 4100dpi, you can see that the horizontal lines can be easily distinguished while the vertical lines can not be clearly perceived anymore, just like with the Nikon 5000. Thus I assign an effective resolution of 3900dpi to the Nikon Coolscan V ED, which means 2,5% less than specified 4000dpi. This is a creditable result since many other film scanners fall below their specified resolution with several umpteen percent. I was surprised that the smallest Nikon is as reliable as his bigger brother concerning resolution. Link to post Share on other sites More sharing options...
tgray Posted February 2, 2012 Share #9 Posted February 2, 2012 Classic internet thread development here - nothing but opinions unsupported by details and facts. ... The Nikon V, 5000ED, and 9000ED all move the CCDs 4000 steps per inch. They do not have the same quality of lens, however (the V lens is not ED, and various reviews say the 9000ED lens resolves more than the 5000ED lens). COOLSCAN V ED from Nikon Sounds like the V is ED as well. It's in the name, and Nikon touts "an incorporated Nikon ED-glass lens". Link to post Share on other sites More sharing options...
adan Posted February 2, 2012 Share #10 Posted February 2, 2012 Yeah - oops - sorry! Same lens. I guess the differences with the 5000ED are speed and other features. Link to post Share on other sites More sharing options...
pico Posted February 2, 2012 Share #11 Posted February 2, 2012 I am curious - who do you know that? I'm not arguing - I don't know, but I'm interested in the basis for the statement. My skepticism began in practice - scanning transparent film, color & B&W at several settings. Larger files often produced no significant improvement of image "quality". Quality depends upon many factors other than whatever the manufacturer claims is "resolution", such as noise, color interpretation, lens quality, on-board processing methods. (To be accurate, 'noise' doesn't necessarily belong in this list since noise is _anything_ of the image that detracts from the information captured.) I am skeptical of what manufacturers mean by 'resolution' - is it sampling frequency? I'll bet not. Does it include the above? I'll bet not. Some lower resolution scans produce images of greater sharpness (modulation). I know it sounds impossible, but it is true. (Sharpening is most destructive of fine detail when using 'unsharp mask', somewhat better when using a custom high-pass filter and adding images.) Unfortunately, I no longer have free access to the ISO printouts, but it would help to refer to ISO 16067-2:2004 (for transparencies, film) ISO 16067-1 (for reflective material.) BTW - scanning a printed military resolution chart is almost meaningless to those looking for quality of scanning transparencies. I won't get further into it. I no longer have access to the good hardware (retired) and at home I have only and Epson 3200 with a dirty glass. I am happy with how it works for scanning print material, but for film it's inadequate. I might try to take it apart. Dunno. Link to post Share on other sites More sharing options...
stunsworth Posted February 2, 2012 Share #12 Posted February 2, 2012 Frankly, I don't care. I scan at 4000 pixels and I get nice scans. That's all that matters to me. Same here. Link to post Share on other sites More sharing options...
Michael Hiles Posted February 2, 2012 Share #13 Posted February 2, 2012 Frankly, I don't care... I get nice scans. That's all that matters to me. My position too. The physics is mildly interesting, but personally, my main interest is in creating a satisfying photograph. Success never hinges on resolution. I am interested in capturing the full tone range in my negative - that is where almost all of the character and emotion reside. Link to post Share on other sites More sharing options...
tgray Posted February 2, 2012 Share #14 Posted February 2, 2012 The take home message for me (I have a V) is that it might not make a full 4000 dpi, but I surely can't tell the difference between a scan of pictorial material done at 4000 and 3900 dpi. So just use 4000 and go with it. The file sizes aren't particularly different and neither are the scan times. On the other hand, there are flatbeds (and other scanners) that advertise things like 6400 dpi when they might only produce 2000-3200 dpi. In that case, scanning at 2-3 times the dpi will drastically increase your file sizes and scan times for no real benefit, so you are probably better off scanning closer to that actual resolution. Link to post Share on other sites More sharing options...
tuanvo1982 Posted February 2, 2012 Author Share #15 Posted February 2, 2012 Honestly, if the scanner is only good at 3900dpi that means 4000dpi is like using photoshop to resize pictures to 4000dpi from 3900dpi. I am thinking of digital pixels and optical pixels Link to post Share on other sites More sharing options...
tgray Posted February 2, 2012 Share #16 Posted February 2, 2012 There's also a difference between what the scanner's optical system can resolve, and how many pixels are on its CCD. I'm pretty sure the V and 5000 have a 3964 pixel chip, which defines the max resolution on one axis, not accounting for optics. Noise coming from the electronics of the scanner is going to be occurring on the pixel scale, not at at the optical resolution of the full system. So there's an argument to be made for scanning at the full resolution of the electronics if you are worried about noise (think deep shadows in slides). Again, in this case, I'm sure most people couldn't distinguish between a pictorial image at a real 4000 dpi and one at a real 3900 dpi resized to 4000 in Photoshop. It's just not enough of a difference for me to worry about. But, if this really worries you, by all means, scan at 3900 dpi. For me, the V makes close enough to it's stated resolution that I don't worry about it. Link to post Share on other sites More sharing options...
tuanvo1982 Posted February 2, 2012 Author Share #17 Posted February 2, 2012 I got it now. Thank everyone for very positive discussion. Link to post Share on other sites More sharing options...
berndr Posted February 3, 2012 Share #18 Posted February 3, 2012 Hello, I skipped the ultimate high resolution scanning with my coolscan 4000. I'm using vuescan and lightroom postprocessing. Scanning with 3600dpi and saving the files as DNG you get a filesize of 50-60MB. Scanning with 4000dpi and saving the files as DNG you get a filesize of 90-100MB. Scanning with 4000dpi and saving the files as TIFF you get a filesize of 100-120MB. In my opinion you don't loose to much resolution when scanning 3600dpi compared with 4000dpi. The grain of the film is still visible. When comparing the two resolutions directly (pixel-peeping) I see no real advantages. Therefore I keep my files smaller and the processing is faster too. Regards, Bernd. Link to post Share on other sites More sharing options...
stunsworth Posted February 3, 2012 Share #19 Posted February 3, 2012 Almost 100% of what I scanned was B&W negatives. I scanned those as 16 bit grey-scales, so the file sizes were approx one third of those. Remember you can also save as (losslessly) compressed Tiff, so that reduces the file size more if you need the space. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.