Re: filmscanners: OT: Native intelligence
At 11:51 PM 7/25/01 -0700, Art wrote: >Getting back to scanners, why is it there is so much discussion of >"banding, banding, banding"... is it that manufacturers think we "want" >banding in our scans? Of maybe it has to do with problems is design >(gee, could that be engineers who made errors?... no, couldn't be)... > >Or maybe, there is lack of precision in the components? Changes in >dimensions due to temperature?, or changes of electronic component >characteristics? Or, yes, some might be software programming defects as >well. Actually, it's been a couple of days since there was any mention at all of banding. I'm looking, at this instant, at a datasheet for a Toshiba CCD (TCD-2901D) freely available on the internet. There is a spec for non-uniformity across the array: TYP: 15%, MAX: 20%. So, in a way, it's amazing these CCD scanners work at all. What makes them work is white-point and black-point compensation. The base measurement has to be done as often as possible - typically just before each scan. The corrections are applied on a pixel-by-pixel basis, for every pixel in the resulting image. Of course, the calibration presumes that the light source is also constant over the course of a scan. If that assumption fails, all bets are off. This is why scanners with cold- cathode tubes often have annoyingly long lamp-warmup times. And of course, there are 2nd-and 3rd-order effects, some of which have been mentioned earlier in this thread. Eg., line-frequency noise, poor grounding (to explain the periodic banding in the 8000 ED) and maybe a host of mechanical issues as well. The yellow/brown streaking (not quite banding) that I saw occasionally on my SprintScan and Microtek scanners may also be due to poor sensitivity in the blue channel. Again, the datasheets tell the story -- the blue channel has about 1/2 the sensitivity of the red and green channels. >My point, very simply is that your assumption that mechanical engineers, >by nature of their diplomas automatically make they more capable of >understanding or implementing design is only partially true, and if they >were all so good at it, we'd like in a world where things held up a lot >better than they do. The difference between Austin and you, or me and you, is that we *are* engineers. So, rather than talk generalities, we look for root causes. I've cited one case (just above) where one can learn a good deal just by looking at data sheets for typical CCD arrays. rafe b.
Re: filmscanners: Re: autolevels was re: filmscanners: Vuescan blue anom aly
At 04:43 PM 7/26/01 +1000, Rob Geraghty wrote: >OK, let me rephrase the question slightly - isn't the intention of the black >and white point to define where the minimum and maximum brightness points >are? If so, why is a point of sun reflection in a photograph not a good >point to use for the white point? Because it's not representative of the >majority of the image? I think the presumption is that these catchlights are blown out, and therefore not representative of any real measurements. rafe b.
Re: filmscanners: Re: filmscanners: autolevels was re: filmscanners: Vues can blue anomaly
At 01:21 PM 7/26/01 +1000, Rob Geraghty wrote: >Rafe wrote: >>With white point, it's important *not* to use >>a specular highlight -- eg., a reflection of >>the sun off of a shiny surface. > >But if that's the only area that could be described as white, what *should* >you use? :-7 Find the next-whitest thing, and make appropriate allowances. When you set white/black points with the droppers, you can make the RGB value anything you want to. This is not a huge problem in practice. I can almost always find a decent black point and white point in every image. This is actually a notion that predates digital imaging by many decades. rafe b.
Re: filmscanners: Wet-mounting slides?
At 05:59 PM 7/25/01, Jeff Goggin wrote: >>AIUI, it's common practice with drum scanners, >>but I've never heard it being done on a CCD >>scanner, unless one were using a glass film >>carrier. (Which might be useful if one were >>scanning a 6x9 cm negative, perhaps.) > >I have a Minolta Scan-Multi and it uses a(n anti-newton) glass carrier for >medium-format film. I've been tinkering with a modification to the holder >that makes it glassless and while this works -- and works quite well -- >it's a PITA to scan more than a few images at a throw because it takes so >much time to load and align the film properly. I thought maybe if I use >mounting fluid with plain optical glass in the carrier, I might achieve >better results still and faster loading to boot... Well, if you do go the route of wet mounting, keep us posted on your adventures. I've been tempted to locate some glass carriers for my 8000, but apprehensive about the side effects -- Newton rings, or wet-mounting to avoid them. The 8000's glassless carrier for MF film is pretty decent, I think, but not foolproof. I'd be curious to hear what others have to say on the subject of scanning MF film, and keeping the film flat. What's the deal with this "Kami" fluid? Does it evaporate completely? What does the film look like afterwards? rafe b.
RE: filmscanners: autolevels was re: filmscanners: Vuescan blue anomaly
At 01:37 AM 7/26/01 +0100, Jawed wrote: >Of course you soon discover that curves are where it's at - but Levels does >things that Curves is a RPITA to do, e.g. set gamma! There's nearly as much >value in learning how to grapple with Curves as there is in choosing PS in >preference to something less functional (well, I say this, but I haven't >used any other graphics editing software anything like as seriously as I've >used PS). Jawed, I'm not sure how you mean that last paragraph, but IMHO, the Curves tool is the single most powerful tool for color correction. Period, end of story. I wouldn't use an image editor or scanner driver that didn't have one. In a way, the Levels tools is "Curves Lite." You can set black point, white point, gray point and gamma. But Curves can do SO much more. Any Photoshop book worth its salt will show you how to use the Curves tool, but the absolute Master at that, bar none, is Dan Margulis. rafe b.
Re: filmscanners: autolevels was re: filmscanners: Vuescan blue anomaly
At 09:34 AM 7/26/01 +1000, Rob wrote: >Rafe wrote: >>* You can set black point to any RGB value you like >>* Ditto for white point >>* You can *place* the black point where it's most >> appropriate within the image. >>* Ditto for white point. > >I'm only a beginner with things like the levels tool. I do actually find >the autolevels useful, although I find it tends to make the colour balance >green on a lot of negs scanned with Vuescan. The problem I've experienced >with using the droppers is that in a 2700dpi scan of an underexposed 400ASA >or 800ASA neg, the colour balance shifts really wildly with tiny shifts >of the mouse, especially when trying to set the white point. Is there a >way to set an area for the dropper to use rather than a pixel? Double-click on the dropper tool. Under "Eyedropper Options", you can select "Point Sample", 3x3 Pixel Average, or 5x5 pixel average. It's important to find an appropriate physical black point for each image. Fortunately, it's not that hard to locate with either the Curves or Levels tool. In Levels, for example: Click the "Preview" checkbox. Now grab the right slider (the highlight slider) and shove it hard to the left. This will reveal where the darkest pixels are in the image. Conversely, to find the white point, grab the left slider (the shadows slider) and shove it hard to the right. With white point, it's important *not* to use a specular highlight -- eg., a reflection of the sun off of a shiny surface. When you use Curves or Levels in this way to set black point or white point, don't hesitate to use "Undo" if the result isn't satisfactory (Alt-Cancel resets the tool without dismissing it.) rafe b.
Re: filmscanners: Wet-mounting slides?
At 03:28 PM 7/25/01, Jeff Goggin wrote: >Has anybody tried this with a film scanner? Does it yield better scans? >Is it worth the added expense and hassle? Is Kami mounting fluid really >the hot setup these days? Inquiring minds need to know! AIUI, it's common practice with drum scanners, but I've never heard it being done on a CCD scanner, unless one were using a glass film carrier. (Which might be useful if one were scanning a 6x9 cm negative, perhaps.) rafe b.
re: filmscanners: Vuescan blue anomaly
At 12:41 PM 7/25/01 +, Lynne wrote: >Rafe wrote: > >>"Auto Levels" is a bit of a sledgehammer approach to >>color correction. Not very subtle, and quite often >>wrong. > >Yes, but it's quick. When you're working on images that differ greatly in >subject, film, time of day, and exposure (and I always am), it saves a lot >of time to get color casts, levels, etc. out of the way right away so you >can do the *real* corrections. :-) I disagree with the last part of that paragraph. By the time "Auto Levels" is done with your image, it may have mangled it beyond repair. Go ahead and use it if you're in a hurry -- most of the time, it does more or less the right thing. A far better approach, IMHO, is to set the black point/white point with the droppers in either the Curves or Levels tool. There, you have much more control -- * You can set black point to any RGB value you like * Ditto for white point * You can *place* the black point where it's most appropriate within the image. * Ditto for white point. The latter approach only takes a bit longer than "Auto Levels" but allows much greater control, with little chance of losing critical image data. "Auto Levels," IMHO, gives far too much freedom to the machine -- the freedom to screw up my photo. No thanks... rafe b.
re: filmscanners: Vuescan blue anomaly
At 05:48 PM 7/24/01 -0700, Alan Womack wrote: >I cannot say I do use PS "Auto Levels" often. Will go off and try it quickly to see. On the last couple of photos that I adjusted the white point up .05 for the blue component, auto levels in Photoshop doesn't work out too well. Makes the colors even more off, took the blue out though! "Auto Levels" is a bit of a sledgehammer approach to color correction. Not very subtle, and quite often wrong. However, it's often instructive to just give it a go and see how it renders your image, followed by a quick Undo (Control-Z.) rafe b.
Re: filmscanners: Nikon Coolscan IV ED vs 4000 ED
At 07:49 PM 7/23/01 -0700, Tim Halloran wrote: >Tomasz, > >Don't buy the Nikon 8000ED unless you need the medium format capability. I agree, though if you *do* need MF scanning capability, there aren't that many choices at the moment, in the $3K price range. Minolta's upcoming entry may be interesting, but I have no idea when it will become available. rafe b.
RE: filmscanners: Vuescan question
At 12:51 PM 7/24/01 +1000, Julian wrote: >Last time I tried Vuescan's IR dust removal I found it didn't work as well >for me as ICE, but this may have improved since then, or at least I should >say it definitely has improved going by what I have read here. > >The bottom line for me is that I have both, and I actually use >Nikonscan. There are plenty of others for whom the opposite will apply. I >will say that for most people there is nothing wrong with Nikonscan, and it >is one of the most powerful OEM scanning softwares around. I suggest the >obvious - try Nikonscan (which you have) and try Vuescan >(try-before-you-buy version) and compare. Then tell us what you discover. I'm with you, Julian. I own Vuescan, and have tried it on my 1640 (and on my SS 35, unless I'm imagining that part) and found it not compatible with my way of working. NikonScan 3.1 is one of the better vendor- supplied drivers I've worked with. rafe b.
Re: filmscanners: Nikon Coolscan 8000
At 11:23 PM 7/21/01 -0700, Tom Christiansen wrote: >Hi, > >I just joined the filmscanner mailing list so maybe my question has already >been answered. If so, could someone please drop me a summary. > > >Has anyone tried the Nikon Coolscan 8000 35mm/MF scanner? May I request a >review? Check the archives of this list, at: http://phi.res.cse.dmu.ac.uk/Filmscan/ Look for posts during June and July of 2001, by myself, Lawrence Smith et. al. There are also a number of online reviews. The latest and most glowing is from Michael Reichmann, at http://luminous-landscape.com/nikon-8000.htm rafe b.
RE: filmscanners: Scanning Mechanisms
At 09:29 AM 7/21/01 -0400, you wrote: >> Clearly the film has to be in SOME kind of carrier, >> whatever the scanner brand. Austin's Leaf uses >> aluminum carriers (Beseler) but all the scanners >> I've owned have plastic-molded film holders of >> varying complexity. Frankly, I feel a bit more >> comfortable with my negatives up against plastic >> than against metal. > >I've thought about that...but you know, those carriers have been the staple >of high end darkroom work for over 30 years! I'm hard pressed to believe >this is actually an issue. I agree, if the edges are all properly de-burred and polished. But sometimes they're not . As I recall, most enlargers (Beseler, Omega, etc.) at least in the old days, used flat aluminum film holders. rafe b.
filmscanners: Scanning Mechanisms
At 02:43 AM 7/21/01 -0700, Art wrote: >Moving the film via a carrier, which is likely molded plastic, with >plastic gearing, and also having it need to "mesh' with the motorized >transport, and being that the carrier is prone to dust and dirt >attraction and "the elements", makes it much harder to maintain >integrity of precision movement. Clearly the film has to be in SOME kind of carrier, whatever the scanner brand. Austin's Leaf uses aluminum carriers (Beseler) but all the scanners I've owned have plastic-molded film holders of varying complexity. Frankly, I feel a bit more comfortable with my negatives up against plastic than against metal. There are hybrid schemes that work very well, such as the older SprintScans. For all my griping about those old SprintScans, they do have a very elegant and well-executed scanning mechanism and optical bench. The older SprintScans used a flimsy plastic film holder, but quite adequate for the purpose. The film holder clicks firmly into place onto a moveable carrier, on precision linear slides (stainless steel,) within the scanner. The film moves, during scanning, and with great precision -- even though it's in a flimsy plastic carrier. The disadvantage of this scheme is that it does not support batch-scanning at all. The scanner has no way to advance to the next image -- that has to be done by moving the plastic film holder by hand. rafe b.
RE: filmscanners: 1640 SU Re-Install Question
At 09:54 PM 7/20/01 -0600, Frank Nichols wrote: >Try going to the Device Manager and removing the device. Then reboot and >hopefully the Wizard will show it face asking to install the new hardware, >then show it the path to the new drivers. Thanks, Frank. That was exactly it. rafe b.
filmscanners: 1640 SU Question: Cancel That
Solution was found seconds after asking the question. Seems you have to unnstall the scanner itself (from Device Manager) so that PnP can "rediscover" it and then install TWAIN along with the scanner. Apparently, TWAIN can't be installed after the scanner, but only "with" the scanner. PITA. rafe b.
filmscanners: 1640 SU Re-Install Question
I can't for the life of me get my 1640 SU TWAIN driver re-installed on my machine (Win 98 SE.) It was happily working a while back, but was deinstalled when I got the 8000. I had a need for it this evening and tried to reinstall it, with no luck. Strange thing is, the scanner itself is showing up nicely in Device manager, and all seems well with it -- no warnings, exclamation marks, or other signs of foul play. A download of the TWAIN driver from Epson (v. 5.00A) yields a self-extracting .EXE file. When exploded, this file yields a single .INF file and two .CAB files. I can unzip the .CAB files with WinZip, but not sure if I should. So now what? I tried installing from the scanner's distribution CD, and no luck there either. There's a directory on the CD with the same 3 files as in the download (a .INF and two .CAB files.) Taking the simple "dummy" route gets me no closer. It says (alternately) "Installing TWAIN driver." "Reboot computer." or "TWAIN Driver deleted." "Reboot computer" -- but in any case, I can't see the TWAIN driver listed in PhotoShop (Select TWAIN source) for the life of me. Any ideas? Epson documentation (FAQ on web site) suggests that Plug and Play should have a role here, but it doesn't seem to. I've not seen any windows "Wizards" kick in to ask for that .INF file. rafe b.
filmscanners: Scanner Service, Banding
At 02:11 PM 7/20/01 +1000, Julian Robinson wrote: >I have reported on this list about the poor focus of my LS2000. I sent it >back for warranty repair and today after 4 weeks I got it back - - - >without trying to encourage Art any more in his campaign, what I got back >is enough to drive me into a rage. I can believe every word of this story, and have heard others like it. I could tell rude stories about my experiences with Brand X and Brand Y film scanners. But not today . I've reconsidered sending in my 8000 at this time, unless or until the banding gets a good deal more bothersome or obvious. In my experience, one might get a unit back from service in worse shape than it went in. Getting the thing fixed *properly* (with a net gain in functionality) can turn into a long, protracted battle. To quote an old saw: "If it works, don't fix it." For the most part, my 8000 is doing the job I expect of it, and quite well, in fact. The difference between Lawrence's experience and mine, on the same scanner, appears to be Lawrence's use of multi-sampling, which I generally avoid. I spent yesterday evening trying hard to find images that would induce serious, obvious banding akin to the images that Lawrence has posted. With 1x sampling, I just couldn't get there. The banding I do see -- and only occasionally -- is quite subtle; I generally have to point it out to people before they see it. This is not a good scenario for a clean, successful repair. I may try again tonight, but failing that, will put "banding" on the back burner for a while. Fixing intermittent problems is always a PITA. Hard enough for competent engineers, and probably not a job for the monkeys at Ye Olde Scanner Service Depot. rafe b.
Re: filmscanners: My replacement 8000 is banding like the first one :-(
At 12:52 AM 7/20/01 -0700, Art wrote: >I don't know how heavy the ED 8000 is, but these days most electronics >have minimal heft to them and aren't very solid. Hi Art. Your "sources" might have told you that it's 19.8 lbs, which information is freely available on the internet, at http://www.nikonusa.com/specs/9246.pdf You need to hire a better research team. rafe b.
RE: filmscanners: My replacement 8000 is banding like the first one :-(
At 12:03 PM 7/20/01 +0100, Jawed wrote: >Dare I say it, but I suspect a scanner moving the film is "less accurate" than >a scanner that moves the scan "head". I disagree, and I'm sure Austin will chime in here too . All film scanners I've worked with move the film -- except for flatbeds with TPUs. The lamp and CCD stay put. This applies to: * Microtek 35t+ * Polaroid SprintScan Plus * Minolta Scan Speed * Nikon 8000 ED * LeafScan 45 All of the above scanners move the media. CCD and lamp are stationary. In fact, except for flatbeds posing as film scanners, I can't think of any film scanners that *don't* work that way. rafe b.
Re: filmscanners: image samples of digital artifacts
At 10:48 AM 7/20/01 +, Lynn wrote: >Rafe wrote: > >>>The sky in the "Prarie" photo looks smooth as silk >>>on my PC, with 24 bit video. With the screen set >>>to "256 colors" I get topo maps in the sky. > >and Bob wrote: >>Thanks Rafe. Mine looked smooth as silk too. I couldn't figure out >what >>I was suppose to be seeing and wasn't. Now I get it. > >OK, I'm not exactly sure what's going on here, that one display set to >factory specs (mine) shows posterization in an Internet JPEG, and two others >(Rafe's and Bob's) do not. > >Should Internet picture postings come with the caveat, "Warning, This >Picture Must Be Viewed At 48-Bits!"? That doesn't sound altogether >realistic, to me. :-) I'm viewing it at 24 bits and it's fine. The one video option that's not acceptable is "256 colors". This is also refered to as "indexed color." Now it's also possible that the video driver or software in Dan's notebook is *using* an indexed-color mechansim while appearing to operate as "16-bit TrueColor." This is where you may need to dig deeper into the tech details of your video hardware. And of course the vendor/manufacturer of that notebook may not be altogether up-front with the necessary details. I think there's a good deal of evidence to indicate that notebook computers, with LCD screens, are poorly suited to viewing and/or editing color graphics. This is asking for trouble. In your case, Lynn, I'm more puzzled. Can you describe the hardware and software options on the system you're using that yields posterization on the "Tibet" jpg? rafe b.
RE: filmscanners: Link to Nikon 8000 banding example...
At 09:55 PM 7/19/01 -0400, you wrote: > >> Did you say, "Leaf 35" ? Not 45? You've got the "little guy" too? > >Rafe, > >I did. Two reasons. One was because the electronics are identical to the >45, so I can use the power supply, CCD board, processor/SCSI board etc. if I >have any problems with my 45, and mostly because I wanted to use it to add >features/re-write the application/plug-in, and not put my 45 out of >commission. I also got a Mac 8500/366 (I think that's the model) to drive >it. Interesting. So how long would it take to scan a 35 mm negative at 5080 dpi? It might be interesting to pit it against the Nikon or the SprintScan. As I mentioned on the list, I may hang on to the Nikon for a bit, unless/until it screws up in a more obvious way. The banding I do see (rarely) is probably too subtle to make an impression on the gorillas at Nikon service rafe
Re: filmscanners: RE: filmscanners: image samples of digital artifacts
At 11:33 AM 7/20/01 +1000, Rob wrote:
>Rafe wrote:
>>I'm willing to bet that Dan Honemann has his video
>>set to 256 colors ("indexed" color.)
>
>Some video drivers in Windows (particularly the generic Windows ones as
>opposed to OEM) only display 256 colours despite being set to 16bit or 24bit.
> It was one reason I had to throw out a video card when I went from Win
>3.11 to Win95.
Well, this may be what Dan Honemann is up against
on his notebook computer. I told him to ditch it.
rafe b.
Re: filmscanners: Link to Nikon 8000 Review
[rafe b:] >isn't Michael >> the same guy who says that the Canon D30 makes >> better images than 35 mm Provia? [Ian] >Nope, he was quoted out of context. You might want to read what he did >write, there being a subtle but important difference. Out of context? Hmm. I read it, again. Yes, it's the same Michael Reichmann, same review I read a few weeks back. And, if I'm not mistaken, he sure does think *mighty* highly of the D30, as compared to Provia/Imacon/EOS-1V. It doesn't matter. I'm hanging in with film for a bit. rafe b.
RE: filmscanners: image samples of digital artifacts
At 10:08 PM 7/19/01 +, Lynn Allen wrote:
>Hi, Dan--
>
>That looks like "Posterization" to me (at least, tha's whut ah calls it! :-)
>--cf definitions (-:|:-) ). I'd say it's probably a result (in this case,
>anyway) of pushing the sizing and JPEG compression too far. A good reference
>is Larry Berman's Compression Comparisons (BermanGraphics--You can look it
>up--I can't access the URL without losing my link on this service).
I'm willing to bet that Dan Honemann has his video
set to 256 colors ("indexed" color.)
JPG doesn't produce "topo maps" Topo maps are a result
of extreme posterization (loss of intermediate tones.)
Indexed color is, by definition, a severely posterized
working space. *Entirely* unsuitable for any graphic
arts work.
To see posterization in Photoshop, go to
Image->Adjust->Posterize, and select a small integer,
say 10 or so. Some of the effects are quite nice,
in fact, but hardly "photographic."
Amazingly, if the integer is over 50-100 on a well-
adjusted image, you won't see the posterization at
all. Which is one reason that I think all this
talk about needing 48-bit color is... well, missing
the point somehow. 16 million colors seems to do
the trick for me.
256-color (indexed color) associates 256 triplets
of RGB values, with the integers 0..255. Those
256 triplets are called a "pallette." The video
card can switch between pallettes quickly, and may
be able to store several pallettes in its memory.
But it can only *use* one pallette at a time.
This is how color video was done, typically, about
10 years ago, before "True Color" became the norm.
JPG doesn't cause "topo map" or posterization effects.
The typical "signature" of JPG is little blocks (8x8
pixels) that are clearly discernable in the image.
rafe b.
RE: filmscanners: Link to Nikon 8000 banding example...
At 06:08 PM 7/19/01 -0400, you wrote: >> On Thu, 19 Jul 2001, Austin Franklin wrote: >Rafe, > >Exactly, and that's my point. If what was suggested is an issue, these guys >made a very basic design flaw...which I am hard pressed to believe they did, >so I question this being a problem. > >I'd like to get together when I get back, and see this first hand, if you >don't mind. Plus I'd like to bring a few negatives scanned on the >"unmentionable" scanner and see how your Nikon does with it. While your >Nikon is gone, if you want to borrow my Leaf 35, you're welcome to. I'll be >developing film for at least a week after this trip...so I'll hardly miss >it. Did you say, "Leaf 35" ? Not 45? You've got the "little guy" too? I have a working SprintScan Plus, so there won't be a problem with scanning 35 mm. The URL for Lawrence's banding pic is: http://www.lwsphoto.com/banding.htm I've never seen anything on my 8000 quite as pronounced. To be quite honest, I'm reconsidering that rendezvous with Nikon service, unless/until I can get some serious, repeatable banding to show up, and preferably banding that isn't "defeated" by the Super-Fine Scan trick. So far, no dice. At the moment, no banding whatsoever, with or without Super-Fine Scan. It may not be wise, I think, to send the scanner in until the problem is obvious and repeatable. Lawrence -- is there a type of image, in your experience, that's most likely to make the banding appear? Slides vs. negatives, for example? At present, I'm scanning slides with deep blue skies. I'm wondering if your problems are exacerbated by the multi- sampling you use. I never use it (multi-sampling, that is.) rafe b.
RE: filmscanners: image samples of digital artifacts
At 05:44 PM 7/19/01 -0400, you wrote: >Lynn, Rafe, Rob and others: > >One thing I've always been curious about is what causes the topographical >map type of lines you see in the blue sky portion of this image: > >http://www.chebucto.ns.ca/~taiji/gallery/t21.htm > >??? > >I see this sort of artifact a lot in jpegs on the web. Is this what is >called "jaggies?" Do they show up in prints? Hold everything! Do you mean, "Prairie, Northern Tibet?" If you're seeing "topo map" effects in the sky, it's almost certainly because you have your video set to "256 colors." There's no way you want to attempt ANY image editing or capture with your screen set that way. This is something you'd change (on a PC) using Control Panel->Display->Settings. What you want is True Color, most likely 24 bits. Using 24 bits with a high resolution requires a video card with a decent amount of video RAM. So you may find that some of the higher resolution settings are grayed out when you select 24 bit color. The sky in the "Prarie" photo looks smooth as silk on my PC, with 24 bit video. With the screen set to "256 colors" I get topo maps in the sky. Get yourself an up to date video card, with at least 8 or 16 Mbytes of video RAM. Matrox is a decent pick for graphic arts and 2-D images. Jaggies are an altogether different matter; they're a consequence of scanning and/or printing at too low a resolution. For example, if you were to try to grab this little image off the web, and print it as 8x10" on your Epson, you'd get "jaggies." There are ways to smooth out jaggies, but they invariably involve softening the image. rafe b.
RE: filmscanners: Vuescan gripes
At 05:45 PM 7/19/01 -0400, Austin wrote: [someone else:} >> I can even live without a >> histogram. [Austin:] >I'm shocked that 1) Viewscan doesn't have a histogram, and 2) that you can >live without it! Ayup. I still wonder why Vuescan is so revered by so many. Earlier versions didn't even have a preview window. All I need or ask from my scanner driver is a decent preview window, a working curves tool and "densitometer," and a *real* exposure control. Ah well, no need to worry about it, Austin. I don't think there's a version for the Leaf, anyway. rafe b.
RE: filmscanners: Link to Nikon 8000 banding example...
At 08:42 PM 7/19/01 +0100, Jawed Ashraf wrote: >For Vuescan, Nikon Scan and Photoshop you'll find a PC does at least as good >a job for rather less money than a Mac. Careful, Jawed. While I might just agree with you, your post is quite likely to upset a few folks. Folks get attached to their particular computers (PCs or Macs) and may have invested all sorts of time learning the ropes on that platform. I only a know a few folks who've switched platforms willingly or succesfully. Most are afraid to try. Macs have a pretty devoted following among graphic arts professionals. Just stating a fact here, not making a judgment. rafe b.
filmscanners: On A More Positive Note
I've posted a few small scans from my 8000 ED at: http://www.channel1.com/users/rafeb/scanner_test4.htm (Photos of shed, and snow-covered boats.) These might explain why some of us are pretty excited about this machine, in spite of all the negative talk 'round here. This was a totally uncorrected scan, at 1x scanning, no ICE, no nothin'. I let the scanner auto-expose the negative, and did no further image adjustments in Photoshop. As raw a scan as you can get. There are several other scans (from different scanners) on this page, so please be patient while it all loads. There are links to additional sample scans, from several other film scanners, at the bottom of the page. (Eg. Epson 1640 SU, for those considering the super-duper CompUSA sale price this week.) rafe b.
filmscanners: Scratch the Gear Teeth Theory
A quick measurement of those "teeth" on the 8000 ED film holders shows 8 teeth per inch (0.125" pitch.) OTOH, the banding that I've seen has a period (width) of about 30-35 pixels, which is well under 0.01" at 4000 dpi. Scratch that theory. rafe b.
RE: filmscanners: Link to Nikon 8000 banding example...
At 01:20 AM 7/19/01 +0100, Jawed wrote: >I agree - I was just about to write as much. > >I don't really know how big a 645 neg is, but the thought of a 4000 dpi scan >across two or three inches (guess) of film makes the mind boggle. Hmm, are >you prints 36 inches square? Crikey I'm not Lawrence, but I'll weigh in on this. If there had been an $1500 scanner that delivered an honest 2500 dpi on MF film, I'd have bought it in an instant. I'm not convinced I need 4000 dpi. Scans of 645 negatives at 4000 dpi yield 160-170 MByte images (24-bit color.) For Lawrence, double those sizes since he's using 48-bit color. >Lawrence, have you verified that you *need* to do multi-scanning? Surely >the DMax of the 8000 is way beyond any negative you might be scanning. And >have you evaluated a scan with no multi-scanning to see if it has banding? I can attest that the banding issue occurs even at 1x scanning. Though I also wonder whether Lawrence really "needs" to to 16x scans. On the larger issue -- I disgree strongly with the poster who suggested that a $3000 scanner couldn't (or shouldn't) be expected to do better. Oh, heck, I know that Imacon scans, drum scans, and Eversmart scans may well be better, but for $3K, I expect an absence of banding. rafe b.
Re: filmscanners: Nikon Service
At 09:37 PM 7/17/01 -0700, Art wrote: >>From my read on this, I don't give a rat's ass about your observations on this topic, Art. I can browse the internet as well as the next Tom, Dick or Harry, and don't need your help to form my opinions on such matters. Have you owned a Nikon scanner? Have you ever used one? Have you ever scanned one of your images in a Nikon scanner? Have you personally dealt with Nikon technical support in any manner, ever? >their service is at least equally as bad, while >costing more, as do their scanners. Further, they tend to maintain an >arrogant attitude about consumer complaints when the repairs are not up >to standard. Again: how does Nikon compare to other brands, offering similar products, on any of these matters? THAT is the question. Not Art Entlich's unfounded opinions, based on his tea-leaf interpretation of internet posts. >If you think my comment is inaccurate, feel free to spend your time >proving otherwise, and report back to us. Bullsh*t, Art. You made one of your many unfounded, accusatory, broad-brush statements about Brand X. It's clear that you haven't any facts to back up your accusations, vis-a-vis Nikon's record, as compared to any other brand. It's not my job to refute your unfounded statements. You made the statement; YOU provide the facts to back them up. Can't cite facts? The give it up. Find something else to talk about. rafe b.
Re: filmscanners: 1640SU @CompUSA $150
At 10:06 PM 7/17/01 +0200, Tomasz Zakrzewski wrote: >This applies to above mentioned Epson >scanners as well as new recently discussed Canon flatbed with 2400dpi. It is >enough to observe that Canon doesn't mention dmax and this is enough for me >to conclude that they don't tell you the whole thruth. >Spec-inflation. Nothing more. Yes, but given the dearth of reasonably priced MF scanners, the Epson 1640 really is a pretty remarkable value. It's a pity there isn't a scanner positioned halfway between the 1640 (currently $250) and the Nikon/Polaroid 4000 dpi models (currently $2800 or so.) rafe b.
RE: filmscanners: 1640SU @CompUSA $150
At 02:10 PM 7/17/01 -0600, [EMAIL PROTECTED] wrote: >At 02:59 PM 7/17/2001 -0400, you wrote: >>Try Buy.com $279 >> >>Harlee Little >> >>How good is the 1640SU? This price is approaching my threshold for an MF >>scanner. >> >>Stan > >Yep, there it is. And not only that, but a little popup window informed me >that I could get another $25 off for any purchase over $250. That would >make the final cost of the 1640SU with TPU $254. Seems like a winner for MF >scanning. Now for some questions: > >1. Has anybody tested this? How good is it? Yes. See: http://www.channel1.com/users/rafeb/scanner_test2.htm It's not 1600 dpi, I'm fairly sure. But it's the best you're going to find for $250 -- of that I'm fairly certain. >and 2. Does VueScan support it? Yes. rafe b.
Re: filmscanners: 1640SU @CompUSA $150
At 07:25 AM 7/17/01 -0600, Stan wrote: [rafe b:] >>If I'm not mistaken, the ad stated specifically, >>"With Transparency Adaptor." > >Yes, it did. Unfortunately, the placard on the item at the store states >specifically that this was a misprint. The 1640SU is being sold without the >transparency adaptor for 149.77 after rebate. I suspected there was a misprint somewhere. But even so, not a bad deal. If the TPU is $100, then the total is $250, compared to the $400 I spent in February 2001 for this same outfit. rafe b.
Re: filmscanners: Which Buggy Software?
At 03:57 PM 7/16/01 -0700, Winsor Crosby wrote: >Judging from the responses it would appear that the Nikon software >for the PC has problems with many Windows set ups, but the software >package for the Mac has no problems? You've apparently missed several reports from users of Nikon Scan on Windoze, reporting no problems. Check out the many reviews on photographyreview.com -- you'll see reports, both pro and con, from both camps. IMHO, where NS has problems, it's in terms of conflicts with other applications and drivers. Other than that, it's not bad at all. rafe b.
Re: filmscanners: Nikon Service
At 01:38 AM 7/17/01 -0700, Art Entlich wrote: >I also think that Nikon might be responding to the dismal service record >that has been well documented on the internet. Let's hope that's the >case. Art, can you document that Nikon's service record has been any worse than that of other companies offerering similar products? rafe b.
RE: filmscanners: Nikon MF LED light source...
At 12:18 AM 7/17/01 -0400, Austin Franklin wrote: >It's 1270 BTW. I'd put the Leaf up against the 1640 any day, as far as >quality of scan goes. On 4x5 media? Remember, that's where I posed the comparison. Yes, I suspect the Leaf would win, but not by a huge amount, unless dynamic range were the determining factor. rafe b.
Re: filmscanners: 1640SU @CompUSA $150
At 07:46 PM 7/16/01 -0700, Winsor Crosby wrote: >If I am not mistaken that is the one that MacWorld in the US did not >like because its tested resolution was no better than a good 1200 and >there was a red bloom along edges. I already own one of these, and have reported on it extensively. It's not a *great* scanner, but at $150, with TPU, it's a steal. rafe b.
Re: filmscanners: Nikon MF LED light source...
At 06:42 PM 7/16/01 -, Lynn Allen wrote: >Does anyone on the List know a good source for these? I for one would pay a >few dollars (US, and cash ;-) ) for one that detailed the HP 5000-6800 >flatbed scanners. I may not be an engineer, but (at least so far) my hands >and my brain still work. Which is more than I can say for my warranty and my >HP scanner. :-) The problem is that even if you had complete schematics, and could understand them, what would you do with that information? Most modern electronic gear costs huge amounts of money to design, but very little to build. The "main" board of a printer or scanner probably is a surface mount board with one or two proprietary ASICs on it, and a handful of passive components. So what are you going to do with that "service" manual? Things like potentiometers and setscrews have largely disappeared. Pots have been replaced by software calibration routines (with results stored in EEPROM or Flash memory) and setscrews replaced by optical sensors. Fact is, most of this stuff is designed with one goal: to be mass-produced cheaply. Servicing is very low on the priority list. The assumption is that it's cheaper to replace than to fix. Here's a fact: typical Lexmark print engines are sold to OEMs for $15-$20. Ditto for flatbed scanner mechanisms. There's no point fixing such a beast; it's just not cost-effective. The warning on the bottom is largely true: "No user- servicable parts inside." rafe b.
Re: Unsharp mask was Re: filmscanners: Getting started question
At 01:11 PM 7/16/01 -0400, Dave King wrote: >I disagree with him (Margulis) on one point however, and I consider >myself a color balance freak. Why? In an "average" color photograph, >global color contrast is maximized at one point only -- the most >"accurate" color balance possible for that scene. I just don't see >how one can get there working by the numbers only (unless one also >wants to make prints by the iterative "hard" proofing process), but I >do see how one can get there on a properly color calibrated system. >Or at least much closer. I would guess it's 80% vs 95%. There's no >substitute for *looking* at actual color when judging this (that I'm >presently aware of). The most accomplished fine art color >photographers also making digital prints would seem to agree judging >by their approaches. Early on in Professional Photoshop (v.4 -- the one I read, way back) Dan explains how he had a color-blind friend doing color corrections, using the basic principles/goals that he outlines. This friend made a few errors, but in fact most of his corrections yielded beautiful results, which do appear in the book. Dan insists that you could use a monochrome monitor to do color corrections. Now, I admit I haven't tried that. But it is quite a provocative claim, and follows logically from Dan's numerical approach. I don't remember Dan using the word "accuracy" anywhere in that book. Ie., color accuracy, per se, isn't held up as a major goal. Speaking for myself: my goal is to produce pleasing, believable photographs, of subjects I've chosen. Matching colors to Pantone swatches is nowhere on my list of priorities. In this regard, I reserve for my own color work the freedom that BW photographers enjoy, where nobody argues about the "accuracy" of the rendition. It's inherently subjective. So, maybe it's not for everybody. If you have clients with specific demands for color accuracy, you may need to go with the more mainstream, ICC-sanctioned methods. rafe b.
Re: filmscanners: Polaroid 120 carrier doesn't line up
At 07:14 PM 7/16/01 +0100, Ian Lyons wrote: > New carriers won't cure this problem, but a firmware update will fix it >real easy. So don't let Polaroid go bust. Ironically, the "best" solution to this problem that I have seen is in the Epson 1640 TWAIN 5 driver. Since the TPU (transparency unit) really has no idea what sort of media you're presenting, it simply scans the entire 4x5 area of the TPU in the preview window. Then all you do is select the frame you want with the crop window. What could be simpler? And there's absolutely no reason that the Nikon 8000 and/or Polaroid LS-120 couldn't offer the same approach. I don't see why they tried to take it upon themselves to "guess" the image locations, and then fail miserably at their horrid guesses. It's one of those situations that leaves you wondering, "What were they thinking??" I would rank this among the major annoyances with the Nikon Scan software on the 8000. rafe b.
Re: filmscanners: 1640SU @CompUSA $150
At 03:58 PM 7/16/01 -0600, Stan wrote: > >>This weekend's circular from CompUSA shows >>the 1640SU -- with transparency adapter -- >>going for $149.77, after rebate. Maybe it's >>a misprint; I haven't called them to verify >>this price. >> >>If it's not, and you have any need at all >>to scan MF media, this is one heck of a deal. >>I paid $400 for mine about six months ago. >> >>rafe b. > >CompUSA's website shows the 1640SU, which does not, AFAIK, include the >transparency unit, at 149.77 after rebate. It also shows the 1640SU Photo, >which does include the transparency unit, at 369.97. It also says it is >out-of-stock (backordered). It says that my local store has the 1640SU, so >I'll swing by there and check it out to see if it does include the TU after >all. If I'm not mistaken, the ad stated specifically, "With Transparency Adaptor." rafe b.
Re: filmscanners: Nikon MF LED light source...
At 06:42 PM 7/16/01 -, Lynn Allen wrote: >OK, I've seen many posts similar to this in the last few months (even made a >few, myself). If it's a "given" that service and repair are such terrible >problems (and believe me, they are), why can't/don't mfgrs make service >manuals more available? Hell, they have the specs and the drawings--how much >more could it cost to make service manuals? (not much, I can tell you--I've >done it). I was able to get a parts manual quite easily for my Mamiya 645E. $20 and it's a done deal. Only trouble is, when I called up to order specific parts, not one of the critical parts was in stock. rafe b.
Re: filmscanners: Nikon MF LED light source...
At 01:09 AM 7/16/01 -0400, Todd wrote: >You know what I hate most about the Leaf? It's that each stage of the >process before you get to the scan is a separate operation, with too many >dialog boxes. Todd -- I'm not taking sides in your debate with Austin, though enjoying the dialog, nonetheless. The point you make above (and the steps you elaborate in the following paragraphs) are fairly typical for other film scanner drivers also. Which is one of the reasons I generally disregard scanner "speed" comparisons. 1. Time to "set up" the scan often exceeds the time taken to actually perform the scan, at least for the way I work. 2. Time spent fiddling with the image in Photoshop afterwards completely dwarfs the scan time + scan setup time. It's like asking which brand of car will get you from 34th St. to the Bronx in the least time. I've also questioned Austin before (and agree with your skepticism) that exposure times can be varied willy-nilly with no effect on scan quality. This is quite contrary to my experience with film scanners and photography in general. rafe b.
Re: filmscanners: Polaroid 120 carrier doesn't line up
Jim Sillars wrote: >When I put a strip of 5 Pentax 645 negatives in the >carrier, the only one that can be scanned is the first >one. My spacing between negatives is apparently not >what Polaroid decided the spacing should be. When >I specify negative 5 for a preview, I get half of >4 and half of five with a lovely big bar in the middle. Ian Lyons noted this as a problem with 645 negatives on the LS-120. I can also confirm that it is a problem with 645 negatives on the 8000 ED. There is, however, a workaround on the 8000, and no need to cut up the strips (but the carriers on the 8000 only hold four 645 negatives, even on a good day.) On the 8000, the workaround is 100% effective, though a bit time-consuming. You dial in an "offset" in the driver GUI, and re-do either the thumbnail scan and/or preview scan until the image is centered. rafe b.
Re: filmscanners: Which Buggy Software?
At 07:29 PM 7/15/01 +0100, Tony Sleep wrote: >On Sat, 14 Jul 2001 07:14:41 -0400 rafeb ([EMAIL PROTECTED]) wrote: > >> And as it turns out, I am a big Dan Margulis >> fan... hence my rotten attitude about ICC >> color management, etc. I think, once you >> start working in the "Margulis" mode, you're >> probably spoiled forever from using these fancy >> profiling and monitor-calibration tools. > >Just as a matter of interest, how does he accomodate interchange of RGB >files with other systems, eg reprographic houses? Lend them your monitor? >;) Nope. Dan's approach is to go by the numbers (RGB values, or L*a*b values, or CMYK values) rather than the appearance of the image on the screen. So in a way, Dan's approach is quite absolute and mathematical, if followed rigorously. I've sent PS 4 files for printing on Lightjet and have never been disappointed by the output. No profiles involved, nor did the lab inquire about profiles, or make recommendations. AFAIK, the lab did not modify the images I sent. Call me lucky, maybe. rafe b.
RE: Unsharp mask was Re: filmscanners: Getting started question
At 01:10 PM 7/15/01 -0400, you wrote: > >> That's not being a Luddite, that's being a cheap bastard. > >I think they are not mutually exclusive ;-) > >> As with your Leafscan, I've compared the output of this >> "lense" to my newer and more expensive zooms, and found >> the latter lacking by comparison. I am content to give >> up auto-focus for the sake of better images. > >I did not know that zoom was an auto-focus! Speaks even more highly for the >amazing results you obtained from it. No, I believe you misunderstood, or I was unclear. That "Access" zoom is manual focus. But it is a much better lens, I believe, than either of my newer, auto-focus Nikkor zooms. To be honest, neither of the Nikkor zooms can be called "professional grade." >I do not, and have never, used autofocus. Though, I have started to use the >automatic shutter speed modes on my Hasselblad...it's good for 1/12th of a >stop...now whether I'll get better results, is another question...but I do >like having the meter in the body. Autofocus on the 8008 is a mixed bag, and I generally don't bother with it. Auto-exposure is another matter. I don't use it on my Nikon FE, but I do often use it on the 8008, which gives the desired results about 90% of the time -- at least with C41 films. Matrix metering really works well, but requires the newer Nikkors. rafe b.
RE: Unsharp mask was Re: filmscanners: Getting started question
At 11:11 AM 7/15/01 -0400, Austin wrote: >On the other hand, speaking of Luddites, what about that ZOOM lense...what >was the brand name of that? That's not being a Luddite, that's being a cheap bastard. As with your Leafscan, I've compared the output of this "lense" to my newer and more expensive zooms, and found the latter lacking by comparison. I am content to give up auto-focus for the sake of better images. rafe b.
Re: Unsharp mask was Re: filmscanners: Getting started question
At 10:18 AM 7/15/01 -0500, you wrote: >Please don't let his arrogance turn you off - he knows what he's talking >about to the nth degree. His specialty is color correction, and I would >venture to suggest that the vast majority of graphics amateurs and >professionals have read his book and use what they have learned from him. Well put, Maris. FWIW, I've met some CM "experts" that were far more arrogant, and far less knowledgeable than Dan M. Though I must admit to suprise at Dan's use of the word "moron" that Austin quoted. I guess we're all subject to foot-in-mouth disease on occasion. rafe b.
filmscanners: 1640SU @CompUSA $150
This weekend's circular from CompUSA shows the 1640SU -- with transparency adapter -- going for $149.77, after rebate. Maybe it's a misprint; I haven't called them to verify this price. If it's not, and you have any need at all to scan MF media, this is one heck of a deal. I paid $400 for mine about six months ago. rafe b.
Re: Unsharp mask was Re: filmscanners: Getting started question
At 02:37 AM 7/15/01 -0700, Art wrote: >Also, you should know that USM usually looks more intense on your screen >(for images to be printed) than it does during the actual printing >process, due to the nature of the dithering process. So it can look >a bit exaggerated on the image on the screen (at 100% view) and still >look fine on the print. You'll need to experiment for finding how far >you can push this. Margulis begins one of his articles by posing the question, "How much USM to use?" And his answer is -- it's like asking for a raise. IOW: as much as you can get away with. I must admit, some of Margulis' USM techniques go way over my head. In the article I saw, he elaborates it into a multi-step process involving several layers, color spaces, and layer-merging steps. When it comes to sharpening, there are many ways to skin the cat. Quite of few of these have been discussed on the Leben Epson list as well. rafe b.
Re: Unsharp mask was Re: filmscanners: Getting started question
At 07:44 AM 7/15/01 -0400, Todd wrote: >Austin, > >You are doing yourself a great injustice to dismiss the work of Margulis >based upon his style. He is an iconoclast who bases his approach on what >works in the real world, as opposed to the theoretical, and is hell bent on >dismantling many of our conventional wisdoms, and the pundits who support >them. > >I think you might actually like him. ;-) Knowing what I know of Austin, I agree. I'd proudly include Austin among my favorite high- tech Luddites. Anyone that swears by and uses a ten-year old film scanner is worthy of membership. >But, one thing you should know, his emphasis is on color work destined for >press. However, if you are interested in the architecture of Photoshop, in >my humble estimation, he's the Dean of the university. .. but not necessarily the Color Management Department. Funny how geniuses (like Margulis) often have a point at which they stop believing. For Margulis, it's color management. For Einstein, it was quantum mechanics. rafe b.
Re: filmscanners: Nikon MF LED light source...
At 07:47 AM 7/15/01 -0400, Todd wrote: >on 7/14/01 3:28 PM, Austin Franklin wrote: > >> How fast can it scan a 6x6 B&W? >>> >>> On a 700 MHz Athlon PC: >>> >>> 2 minutes, 10 seconds with "Super Fine Scan" OFF. >>> 5 minutes, 15 seconds with "Super Fine Scan" ON. >> >> FYI, the Leafscan is well under 4 minutes. >> > >At 4000 DPI? Worth mentioning also that the 8000 scans RGB with these same timings. rafe b.
RE: Unsharp mask was Re: filmscanners: Getting started question
At 12:15 AM 7/15/01 -0400, Austin wrote: > >> One article is online at http://www.ledet.com/margulis/Sharpen.pdf > >I haven't read enough to know if this guy Margulis knows what he's talking >about or not, but to quote from one of his articles: > >"Anyone who thinks that if a fine screen is good, than a finer one must be >better is a moron." > >Right or wrong, I really have no interest in reading anything from someone >who is so disrespectful of his readers and feels he needs to call them >names, no matter how much of a genius he may be. Austin, don't be so harsh on the guy, or so quick to judge. Maybe read a chapter of his book(s). (I think there's a chapter online somewhere.) He *is* deliberately provocative, that's true. I see a bit of myself in Margulis -- sort of a high-tech Luddite. One of Margulis' main shticks is that monitors lie, and are not to be trusted. And of course, he really goes to town on the subject of monitor calibration. rafe b.
Re: filmscanners: Inkjet Printer List?
At 06:35 PM 7/14/01 -0600, you wrote: >Can anyone point me to a good/active mailing list for discussions of ink jet >printers (specifically Epson) used to print my scans? That seems off topic >for this list... www.leben.com has the instructions you need to join the Epson list. See ya there... rafe b.
Re: Unsharp mask was Re: filmscanners: Getting started question
At 08:28 AM 7/15/01 +1000, Rob wrote: >Gad, unsharp mask over 100%? I've been using a radius of 2.0 and only 60%. >Is there something I'm seriously missing about USM? Using the USM tool is a black art. Dan Margulis has written some great articles about using this tool most effectively. I believe the the text from these articles also appears in the newer versions of "Professional Photoshop." You might try backing off on the radius and cranking up the amount. Eg., radius 1.0, amt=100%, for a 2700 dpi image. Lower resolutions require lower numbers. Higer resolutions allow higher numbers. rafe b.
Re: filmscanners: fogged film
At 08:38 AM 7/15/01 +1000, Rob Geraghty wrote: >Fair enough! I just took some photos of coral on the Great Barrier Reef >and underwater photography is a whole new challenge. Some of the >coral near the surface looked slightly bleached to me. If they're anything like the corals off the coast of Florida... that might be because they're dead or dying. rafe b.
filmscanners: Canon D2400F
A discussion of this scanner has been going on on several forums. I hear mostly lukewarm or negative reports. The following URL was recently posted on rec.photo.equipment.medium-format: http://www.tbns.net/kevinspages/canoscan/ with sample scans. The owner of this scanner is not too pleased. Like I say -- reminds me of my Epson 1640SU, only with even more spec-inflation. rafe b.
Re: filmscanners: Semi OT: 16-bits [was Which Buggy Software?]
At 12:24 PM 7/14/01 -0500, you wrote: >I've read Professional Photoshop 5 cover-to-cover, and just bought >Professional Photoshop 6 since I also bought PS6, and I use his methods >though I think he does go overboard respecting color management's >shortcomings - he has a new chapter in "6" concerning it that I am eager to >read. > >Meanwhile, I am attaching his public challenge, as posted on his Color >Theory group http://groups.yahoo.com/group/colortheory , which asserts that >no photographic image can be corrected better in 16-bit color than it can in >8-bit color. Over the years I've taken a lot of heat for defending Margulis. When I read "Professional Photoshop" I had no idea who these "calibrationists" were that Dan gets so riled about. I've since met a few of them, on the various lists that I partake in. I won't mention names. I agree that Margulis (like any other "expert") should be taken with a few grains of salt. But I'm way grateful for the lessons I've learned from Dan, and I believe these lessons have saved me loads of grief over the years. The grief I do experience is with making "color management" work, and arguing with the CM gurus. rafe b.
RE: filmscanners: Nikon MF LED light source...
At 09:19 AM 7/14/01 -0400, Austin wrote: >How fast can it scan a 6x6 B&W? On a 700 MHz Athlon PC: 2 minutes, 10 seconds with "Super Fine Scan" OFF. 5 minutes, 15 seconds with "Super Fine Scan" ON. If using NS in TWAIN mode, that's about the extent of it. However, if using NS in Stand-Alone mode, you get to deal with its horrifically slow file- save times. 1 minute, 30 seconds for this 76.6 MB image. FWIW, Photoshop can save this same image in about 3 seconds. I'll send a couple of 400 x 400 snippets to your private email address, JPG compressed. rafe b.
Re: filmscanners: Which Buggy Software?
At 02:30 PM 7/14/01 +0100, Tony Sleep wrote: >On Fri, 13 Jul 2001 18:19:28 -0700 Pat Perez ([EMAIL PROTECTED]) wrote: > >> the >> most infamous being 3rd party manufacturer video drivers > >Yup! Absolutely (one reason why I conservatively stick to Matrox is that - >eventually anyhow - their drivers usually get to be well behaved). > >It may be useful if people suffering Nikonscan crashes could identify >their graphics cards and drivers. Maybe you're onto something here, Tony. I use a Matrox G200 AGP board. But strangely enough, its installation was rather messy (driver-wise) and it still crashes during one of the more obscure Wintune tests (though not on any apps that I normally run.) rafe b.
Re: filmscanners: Test Imacon, Nikon.Polaroid
At 01:17 AM 7/14/01 -0400, Dave King wrote: >If there are no mirrors in either, what would explain better sharpness >in the Imacon (assuming flat film in the Polaroid and Nikon)? A good question, Dave, and I don't have an answer.. just an observation. In the last Traveling Portfolio that I participated in, there were some prints -- printed on an Epson 3000, no less -- that very much stood out in terms of sharpness and tonality. I asked the author of those prints about the specifics, and was told that they were from large negatives (6x7, I think) and scanned on an Imacon. That may have been one of the factors that led me to pursue larger film sizes, and all the extra aggravation and expense. Since then, I'm happy to say -- a few of my 645 shots, scanned on the 8000, come pretty close to reproducing that sharpness and tonality. Certainly a big step up from 35 mm stuff, scanned on the SprintScan. Whether it's up to "Imacon" quality, I can't say. Idle speculation, mostly, considering the Imacon's cost. Now if I could only get my subjects and compositions to match those that came so easily on my old Nikon FE, I'd be even more pleased. rafe b.
RE: filmscanners: The hunt for a scanner for contact-sheets: Microtek 8700
At 10:07 AM 7/14/01 +0100, Tony Sleep wrote: >Thanks for the info. That sounds expensive though. There's always eBay rafe b.
Re: filmscanners: Which Buggy Software?
At , Maris wrote: >>You're in Dan Margulis's camp, then? He maintains (and I have no opinion >>one way or the other) that 16-bit color are not necessary. I was not aware that Dan Margulis said that, but it doesn't surprise me, and in fact I'm pleased to have such prestigious "backup." And as it turns out, I am a big Dan Margulis fan... hence my rotten attitude about ICC color management, etc. I think, once you start working in the "Margulis" mode, you're probably spoiled forever from using these fancy profiling and monitor-calibration tools. I really think anyone seriously involved in "digital darkroom" owes it to themselves to at least hear out Dan, or read some of his stuff. (He's the author of the "Professional Photoshop" books.) Ask yourself -- how did the pros manage to get nice looking colors before the ICC came along to "fix" everything? rafe b.
Re: filmscanners: Nikon MF LED light source...
At 12:18 AM 7/14/01 -0400, Austin wrote: >Nikon gurus... I believe we discussed that the new Nikon MF scanner uses >LEDs as the light source, but does it use only one CCD row, and switch each >color on/off and scan each row three times? > >Wouldn't that make it quite a bit slower than using a tri-color CCD? It uses a 3-row *monochrome* CCD sensor. That's in the specs on the Nikon website, I believe. In normal operation it uses all 3 rows, but I believe that's for speed's sake only. There's a checkbox called "Super Fine Scan" in the scanner driver. That causes the scan to use just one row of the CCD. It also turns out that this feature is *key* to eliminating the banding that Lawrence and I have seen (at least on Lawrence's first 8000.) Of course, it runs at 1/3 its normal speed when you do that. The manual is a bit cryptic about why or when you might use this "feature." Using all 3 CCD rows, the Nikon is plenty fast. Even on one row, it's no slouch. A heck of lot faster than your Leafscan, that's for sure, either way . So... I'm clearly annoyed that this euphemistically- labeled feature has become a necessity (for certain scans and situations) but am living with it. As long as I *never* see banding again, I will live with the slower scans. rafe b.
RE: filmscanners: Scanwit: Seeing through mount?
At 09:36 PM 7/13/01 -0600, Frank Nichols wrote: >I think the bottom line is that my next scanner will definitely have a >manual exposure mode! I wouldn't consider owning a slide scanner that didn't. I even returned one (Minolta Scan Speed) when I discovered that it had no way to control the exposure manually. Aside from that, it was a pretty decent machine. rafe b.
Re: filmscanners: Getting started question
At 07:15 PM 7/13/01 -0700, Pat Perez wrote: >I'd suggest using slide film for learning. It is less expensive to process, >and you can see the actual result, not having to guess what is on the film, >as you would with negatives. > >I wish I had the discipline to shoot that much for practice's sake. I could >certainly use it. For scanning, there are some real advantages to working with negative film. Negatives are far less likely to hit the dynamic range limits of the scanner. For shooting, negative film will give you far more exposure lattitude than slides. There are many "natural light" scenes that can be captured quite nicely on negative film, but not on a slide, due to the lack of exposure lattitude on the latter. Granted, the primary problem with negatives is not being able to "see" the image on the negative itself. I find this to be a small price to pay, given all the other advantages. Fuji Reala, Kodak Supra (100) and Kodak Royal Gold (100) are beautiful films for natural- light landscape photography. Does anyone know if Reala is available in 120 format? I haven't found it yet. For 120, my favorite so far is Fuji 160 NPC. rafe b.
Re: filmscanners: Which Buggy Software?
At 10:13 AM 7/14/01 +1000, Rob wrote: >Nikonscan 3.1 seems to work fine on my system, no crashes. Having said that >I don't use it because the scans come out posterised in comparison to >Vuescan output because Nikonscan only works with 8 bits of data from the >LS30. This shouldn't be an issue with newer Nikon scanners. I've been making scans in 24-bit color for years, on about 3 or 4 different film scanners. I don't use the 48-bit color mode, ever, even on the 8000. No posterization. I suspect there's another reason for the posterization you're seeing. rafe b.
Re: filmscanners: Grain, Noise, et al
At 10:26 AM 7/13/01 -0400, Norm Unsworth wrote: >>From a practical, rather than a causative approach, how have folks dealt >with this issue, both in terms of minimizing apparent grain from scans and >in improving (ie: reducing) the appearance of 'grain' in Photoshop? I deal with it by shooting slow films -- ISO 100. Beyond that, I just live with it . Obviously, I don't do sports photography or photo- journalism, and I use a tripod quite a bit. rafe b.
Re: filmscanners: Which Buggy Software?
At 11:59 AM 7/13/01 -0700, Winsor Crosby wrote: >There seems to be some difference in the experience of people using >Nikon scanner software. Some people say it is fine. Others complain >bitterly about its "bugginess" with out much more in the way of >additional information. Since Nikon provides two software packages, >one for the Mac and one for Windows, it might be useful to know the >operating systems of those people who offer their experiences one way >of the other. FWIW, my reports (mostly positive; no crashes) are based on a Win 98 SE platform (Athlon 700, ABit KA-7 motherboard, VIA chipset.) NikonScan 3.1, on an 8000 ED. rafe b.
RE: filmscanners: Test Imacon, Nikon.Polaroid
At 03:57 PM 7/13/01 -0400, Austin wrote: > >> The primary advantage of the Imacon design is the unfolded light path >> correct? The mirrors can't be helping with the less expensive >> scanners. Only absolute disadvantage to the straight path approach is >> physical size of the scanner(?), and of course, in the case of the >> Imacon, cost. > >Same thing with the Leafscan, it also has a straight light path, no mirrors. >Also, one feature of the Imacon is the magnetic curved film holders. I am >not sure if it actually is better or not, but it is a feature. Are we certain that the 8000 ED and/or the LS-120 use mirrors? Where does this information come from? This was commonly "reported" to be the case on some other scanners. I can tell you that it is categorically not true for the Microtek 35t+ and the SprintScan Plus -- no mirrors in either of those; the both use the identical optical/mechanical bench. rafe b.
Re: filmscanners: Buying on eBay
At 12:40 AM 7/12/01 -0700, Art wrote: >Please read the paragraph below, which was in my earlier post. Yes, >insurance is a no-brainer, bit have you ever tried to collect on >shipping damage? No, I've had no need. I've bought nearly $2000 worth of photo gear and accessories on eBay over the last couple of years -- cameras, lenses, etc. -- and not had any problems with careless packaging or items damaged in shipment. To reiterate: I believe you over-state the risks involved. rafe b.
Re: filmscanners: Nikon 8000ED
At 12:50 AM 7/12/01 -0700, Art wrote: >Visible light does not need to be an LED source for dICE to work. Acer >2740 uses a hybrid situation, with an IR LED, but cold cathode light >source. Canon FS 4000 has FARE, which is also an infrared defect repair >system and is not, to my knowledge using an LED visible lighting >system. Finally, the Minolta Elite doesn't use an LED visible lighting >source, and it also has dICE. I think the cheapest way to provide good >IR is an LED array, but it doesn't seem to preclude use of cold cathode >for the rest of the lighting. You raise some interesting points here, Art. Clearly there are hybrid solutions that get around the design compromise that I cited. But curiously, our man from Polaroid is in ICE-denial, saying that his (perceived) clientele doesn't value ICE. Nor did I, until I had a chance to work with it. As I recall, David was in similar denial when some of us informed him (way, way back) that we'd really like a TWAIN driver for our Polaroid scanners. rafe b.
Re: filmscanners: SS120 & Nikon 8000 ... how do they work?
At 03:14 AM 7/12/01 -0700, Art wrote: >It seems to me for some reason that most of the newer medium format >scanners manufacturers decided to forego the zoom lens approach >that Minolta has and continues to use with their Multi models, >and just basically use the same optics for all the film formats. > >That eliminates the need for fancy (and maybe problematic) >mechanics that change the film position or zoom lenses. >Other than the Minolta Multi, the other MF scanners seem to work >similarly to a flatbed, in the sense that regardless of the size of the >original, the resolution of the CCD remains fixed (in this case at 4000 >dpi.) I was mistaken. The lens in the 8000 ED is not a zoom lens. To quote: "Imaging Optics: Scanner Nikkor ED lens (14 elements in 6 groups including 6 ED glass elements)" No mention of zoom here. This would explain why its 35 mm res is no better than the res on 120 film. However, the upcoming/newly-announced Minolta medium-format scanner continues in Minolta's tradition of differing resolutions for differing media. See: http://www.imaging-resource.com/NEWS/993665852.html Minolta also continues the fine tradition of spec inflation here, claiming a 16 bit A/D and "4.8D dynamic range." Res is spec'd at 4800 (in the URL above). I have heard from other posts that this is the spec for 35 mm, and that for 120 film, the res is around 3000 dpi. This is also "consistent" with specs on the Minolta Multi -- the advertised res is for 35 mm, and the actual res for 120 film formats is some fraction of that number. rafe b.
Re: filmscanners: SS120 & Nikon 8000 ... how do they work?
At 11:16 PM 7/10/01, Jeff Goggin wrote: >Perhaps someone can clarify something for me ... how do the new Nikon and >Polaroid scanners achieve their claimed 4000dpi resolution for multiple >formats? Unlike most film scanners that accomodate multiple formats, the >claimed resolution of these scanners doesn't decrease when they switch from >35mm to medium-format, which suggests (to me, somebody who knows very >little about scanner mechanics) that they're doing something other than >using a different lens for each format. Has anyone here figured out >exactly what they've done and how? Good question, Jeff. >From Nikon's specs, we know that they're using a 10,000 x 3 element CCD. Across a 2.25" media, this comes out to roughly 4000 dpi. We also know from the specs there's some fancy 14-element zoom lens in the box. What's unclear then is why they can't or don't provide better than 4000 dpi on 35 mm media, or why a "zoom" lens, at all. rafe b.
Re: filmscanners: SS120 first impressions and a few questions.
At 06:51 AM 7/11/01 -0400, Cary Enoch Reinstein wrote: [Rafe B:] >>Your complaint regarding NikonScan 3.1 being "buggy" >>is surprising to me. I had some initial problems >>getting NS installed, but it has not been remotely >>buggy since then. The installation issues turned >>out to be due to device conflicts. >> >>Which leads me to suspect that a good number of >>reported bugs in NS are in fact due to device >>conflicts. This info is of no use to you at the >>moment, but others reading this may care. [Cary:] >Since I installed NikonScan 3.1 I haven't had any problems with its TWAIN >module. When used stand-alone it may have memory problems. The integrated >curves and histograms dialog is the best I've ever seen in any graphics >application. The version at Nikon's European site is more complete than the >one at their US site. It includes a utility named regsweeper.exe which >should be used before installing ver. 3.1. The version of NS 3.1 supplied on the Nikon CD has these utilities on it, though they're well-hidden. The Nikon Level-1 tech knew where to find 'em. Re: stand-alone NS versus TWAIN. I don't have problems with either one in terms of stability, but there's a tradeoff with regard to useability and efficiency. Specifically: 1. Using TWAIN, scan data is sent to Photoshop (the TWAIN host) quite rapidly. But you have to close the TWAIN applet to do anything in Photoshop. Then, when re-entering the TWAIN applet later, to do the next scan, you have to go through the whole process of obtaining thumbnails, re-locating the images, etc. Alas, with 170 MB files, you really want to save them to disk and close them before going for the next scan. 2. Using NS stand-alone, you avoid the above issue. However, NS stand-alone is *very* slow at saving scans to disk (several times slower than Photoshop.) I agree that NS 3.1 is (in most regards) quite a decent scanning app, on the issues that most directly affect image quality. The preview image is as large as I care to make it. The densitometer readings are accurate, and the curves tool works smoothly and predictably. Its major weakness, IMO, is its method of "locating" individual images on a 645 film strip. Working around this problem generally costs 2-3 minutes per mis-located image. >I didn't use Nikon's throwaway IEEE1394 board because I already had an >Adaptec 1394 card installed for DV capture. The Adaptec card is much more >robust and I'm guessing that it won't cause any conflicts. I suspect that >some reported conflicts are due to the Nikon card. The Adaptec doesn't need >any drivers or installation routine because Windows 2000 recognizes it >automatically, probably the same for Win9x also. Hmmm. Interesting. In the course of resolving my initial problems with NS, Nikon mailed me a second Firewire board, which didn't change or help matters at all. But it was the same exact board (vendor and model) as the first. The vendor is Ratoc, and I did have a word with them via email. It seems the model they bundle with the Nikon scanner is a special OEM version; there's no info at all about this model on the Ratoc website, and Ratoc defers all support to Nikon. Ratoc's "advice" at the time was fairly lame, and not much help. (Reinstall Windows, reinstall VIA drivers, etc.) I was tempted to pick up a Firewire card at Circuit City, but the models I saw were $75, and were definitely not Adaptec. rafe b.
Re: filmscanners: Nikon 8000ED
At 10:31 PM 7/10/01 -0700, Art wrote: >It sounds to me that this is a heads up to companies like Polaroid and >others that it might well me worthwhile to consider production of at >least one scanner line with D-ICE or equivalent type products with an >infrared channel. "ICE" was a non-issue for me in choosing the 8000 ED. I assumed that ICE was some sort of dumb gimmick and that image quality would suffer from using it. I'm happy to say that I was quite wrong about this. In fact, when I consider the countless hours I've spent in the last few years spotting and retouching scans, I kinda kick myself... However... there IS a price to pay, and it gets back to the LEDs vs. cold-cathode lighting issue, I think. If I'm not mistaken, ICE requires IR-LED illumination. The Nikons have this, the Polaroids don't. But it also seems that shallow depth-of-focus may be a side- effect of LED illumination, at least according to one of theories floating around. To wit: the LEDs are less bright than cold-cathode, hence wider apertures (and lower depth-of-focus) in the internal optics. It would be interesting to put this to the test, somehow, perhaps with deliberately bent or curved media. Shallow depth of focus *is* an issue on the 8000, when scanning 645. I have to be extremely careful loading the filmstrips in their holders to ensure that they're quite flat. It's often a hit-and-miss thing. Not so much an issue on 35 mm filmstrips. rafe b.
filmscanners: Buying on eBay
At 07:39 PM 7/10/01 -0700, Art wrote: >As those who bother reading my comments know, I am an advocate of buying >used equipment and non-bleeding edge. It saves money, it sometimes even >protects you from the "first buyer screw" which means you end up paying >the most for a product which isn't perfected, and end up playing beta >tester without any of the benefits, and sometimes that means being very >poorly treated by the manufacturer. > >Having said that, I use ebay a fair amount (I do very much understand it >is difficult enough to do so if you live in Canada, and nearly >impossible if you are anywhere else, since it a very U.S.-centric >service). However, based upon my experience, I would be very cautious >buying something like a film scanner (other than a new one with full >manufacturer's warranty) from ebay merchants. >rafeb wrote: >> For those interested in buying their very first film >> scanner, or trading up from an existing model, I humbly >> suggest eBay as a source of very good deals, particularly >> at the moment -- where new market entries from several >> major vendors are causing a lot of turnover in equipment. There's no harm in being careful, that's true, but no need to overplay the dangers either. When I upgraded from my first film scanner (a Microtek) I sold that unit to a fellow in Vancouver BC. Long story short... the unit was carefully packaged, and arrived quite intact, surviving the long trip from Boston to British Columbia and through Canadian customs. I would expect that expensive gear would be insured for shipment. Wouldn't that be the sensible thing? Anyone considering buying a Leafscan should be extra- careful, though; it is a very large, very heavy unit, and will cost a bundle to ship. As to the potential savings... I see (just for example) a SprintScan 35+, "New In Box" (NIB) for $425, with less than one day to go. No reserve, and no bids yet. The seller seems quite well qualified. Hmm. there's even an Imacon up there at the moment, no idea what the reserve might be. For those eyeing the LS-120 or 8000 ED, eBay may not be viable, as scanners with this price/performace simply don't exist on the used market. OTOH, if you can find someone selling an Imacon for a song, or a Leaf 45 for, say, $1500 or less - go for it! rafe b.
Re: filmscanners: Polaroid Sprintscan 120
At 10:41 PM 7/10/01 -0400, Dave King wrote: >Enjoy. This and the new Nikon are the first generation of CCD film >scanners that are capable of results that are essentially "good >enough" for any conceivable critical use with film up to medium format >size. I'm not sure I agree there, Dave. The Leafscan 45 and the Imacons (both CCD) have been around for a while. The two new models (from Polaroid and Nikon) are poised, IMHO, somewhere between these two very worthy (but dated) standards. On 35 mm, the Leaf 35 and 45 can probably still beat either the Polaroid or Nikon. Er, that is, if you have an hour or so to wait (on the Leaf.) What *is* quite significant is the price that these new models are being offered at -- roughly 1/3 to 1/4 of the Leaf's original price, or Imacon's current retail price. rafe b.
RE: filmscanners: SS120 first impressions and a few questions.
At 05:37 PM 7/10/01 -0700, Slavitt, Howard wrote: >I too wonder whether the Polaroid may have problems with dust getting insde. >I strongly suspect that with either the SS120 or Nikon 8000ED scan quality >may decrease over time as more dust gets inside the machine. You should >definitely put a dust cover over the machine whenever it's not in use. Surprisingly, in 2 years time, this hasn't been an issue for me on my SprintScan Plus. If you look at the construction (of that model, anyway) it turns out the optics are embedded in a long narrow tube, in between the film carrier and the CCD sensor. I guess dust has a hard time getting onto the front side of the lens, and no way, really, to get at the back side of the lens. FWIW, the 8000 ED has two separate doors covering the film carrier slot. There's that thin hinged door which flops up when the film carrier's inserted. It also turns out (geez - I had to read the manual to find this out ) that that "decorative" black panel is also a secondary cover for the film carrier slot. Most photos of the 8000 show this panel pushed down to reveal the film carrier slot. When the scanner's not in use, this panel slides up to cover the slot. Jeez, it's hard writing this without sounding like Masters and Johnson . rafe b.
Re: filmscanners: SS120 first impressions and a few questions.
Paul, thanks for (yet another) LS-120 review. A couple of comments if I may. Your complaint regarding NikonScan 3.1 being "buggy" is surprising to me. I had some initial problems getting NS installed, but it has not been remotely buggy since then. The installation issues turned out to be due to device conflicts. Which leads me to suspect that a good number of reported bugs in NS are in fact due to device conflicts. This info is of no use to you at the moment, but others reading this may care. Surprising that you say the LS-120 is loud. My 8000 ED is by far the noisiest film scanner I've owned. Hard to imagine the LS-120 being louder. Do you wear ear protection ? . Re: Photoshop Books: * Photoshop Artistry, by Haynes and Crumpler. * Real World Photoshop (5 or 6) by Blatner and Fraser * Professional Photoshop by Dan Margulis rafe b.
Re: filmscanners: Primefilm 1800i
For those interested in buying their very first film scanner, or trading up from an existing model, I humbly suggest eBay as a source of very good deals, particularly at the moment -- where new market entries from several major vendors are causing a lot of turnover in equipment. If you're unsure of which model to buy, use the web to get a feel for the various models -- there are many review sites, comments on usenet forums, and so on. Bear in mind that there will be a lot of hype, both pro and con, on almost any model. Take the more extreme opinions with a large grain of salt. To search usenet archives, go to www.deja.com. Usenet groups that discuss scanners include (for example) comp.periphs.scanners, rec.photo.equipment.35mm, and rec.photo.equipment.medium-format. Sites that offer reviews of various film scanners (either by columnists, or by the public) can be found via the usual web-search engines. One of the better ones I've stumbled into recently is www.photographyreview.com I've personally had only good luck buying photo gear on eBay. Standard caveats apply, of course. Choose your sellers carefully, read the fine print, and ask appropriate questions before you bid. rafe b.
Re: filmscanners: Nikon 8000ED
At 02:04 AM 7/9/01 -0700, Art Entlich wrote: >I'm musing whether Nikon has a factory in the "deep south" of the US. >I'm noting a very strong allegiance to the company coming from those >environs... More witty observations from Brother Art, who doesn't even live in the USA. Allegiance based on geography is what Kurt Vonnegut calls a "granfalloon." Art, it might interest you that I live in a suburb of Boston, MA, which is where Polaroid's factories are. Yes, that same Polaroid which, just a few weeks ago, announced layoffs of 25% of their workforce. Last I checked, it was a long way from here to North Carolina. rafe b.
Re: filmscanners: Nikon LS IV/Nikoscan 3.0
At 02:46 AM 7/9/01 -0700, Art Entlich wrote: >My comment, based upon fact, was that Nikon's scanner software has had a >history of many problems, and that these same problems extended to >several Nikon peripherals their scanners work with. Oh, poop, Art. Go to the PhotographyReview web site and look at how people complain about scanner software -- both Nikon and Polaroid owners. I'd be hard pressed to tell which ones are more upset with their scanner software. In their defense, Polaroid has had the good sense to bundle a decent 3rd-party package with their latest scanners. And in their defense, Nikon seems to have finally done a good thing with NS 3.1, which by most accounts is a solid step forward. >I am continually amazed how defensive some Nikon owners have been and >continue to be about criticism of Nikon products. I think this is >called "compensation" in my text books. Sort of like people who buy hot >sports cars and think women will assume certain of their body parts are >bigger than they really are, as a result. ;-) Spare us the pop-psychology, please Art. The point was that you don't even *own* a Nikon scanner, by your own admission, but are quick to take sides in an argument involving Nikon scanner sotware. Why not just take a breather on this topic? Give it a rest. I wasn't defending Nikon software. I was criticizing you for offering an opinion on a controversial topic, where your opinion has no basis in direct experience. >What any of this has to do with Polaroid is beyond me. By singling out Nikon as a company writing lousy scanner software, others might presume that Canon, Minolta, Polaroid, Acer, HP, Leaf, Kodak or Brand Z must be doing a better job in that department. rafe b.
filmscanners: Scanner Reviews Link
For those trying to decide between Nikon's and Polaroid's latest medium-format scanners, you might want to have a look at http://www.photographyreview.com/reviews/film_scanners/ There are at present four reviews of the 8000 (mine among them) and three reviews of the LS-120. I won't spoil the fun and tell you which one gets the higher marks. I've searched around for other user reviews, but haven't found any. Apparently there aren't that many copies of either model out in the field, quite yet. So, aside from Ian, where are the LS-120 users on this list? rafe b.
Re: filmscanners: Nikon LS IV/Nikoscan 3.0
At 06:11 AM 7/7/01 -0700, Art Entlich wrote: >You know, some people have had problems with Nikon software... a LOT of >problems. And quite a few have complained about Polaroid's scanner software as well. >Not having their scanner or need for their software... So why add fuel to the fire, Art? rafe b.
Re: filmscanners: Film Scanner Question Again
At 01:56 AM 7/8/01 EDT, Roger Miller wrote: Roger, there were a couple of points in your recent post to Rick Decker that I'd like to comment on. My experience with the 1640 SU is that there is absolutely no advantage to setting 3200 dpi resolution (as compared to 1600.) There are a number of scanners out there with "assymetrical" resolutions, and it's usually a shell game. Ditto for printer resolutions. The number that matters is the lower one. The higher number is simply for ad copy. "Looks good on the side of the box" -- as the marketing guys say. The other is the matter of resizing/resampling the image in Photoshop. You (and Rick) should understand the difference. In Photoshop's Image->Image Size dialog, there's a check-box labeled "Resample Image." If you CHECK this box, PS will either "create" or throw away pixels according to the resolution, height, and width that you ask for, and the resolution, height and width of the existing image. If you UN-CHECK this box, PS will neither create nor destroy pixels; it merely changes and internal tag, somewhere in the image file, that determines the physical size of the printed image. If you scanned a 35 mm frame on the 1640SU, you get a file that's 1600 x 2400 pixels (let's use round numbers here.) If you set target size at 100% in the scanner driver (I'm working from memory here) it will arrive in Photoshop sized at 1" x 1.5". If you print it that way, you'll get a 1" x 1.5" print. So you want to resize or resample. Which to choose? Fortunately in Photoshop, it doesn't matter much -- Photoshop does a good job resampling. But just bear in mind -- with "Resampling" an entirely new image is created, pixel by pixel. With "Resize" the original pixels in the image remain untouched. (So "Resize" happens almost instantaneously, whereas "Resample" takes some time, maybe 15-30 seconds on this image, on a reasonably fast machine.) A "Resize" of this 1600 x 2400 image might yield, for example: -- an image 2" x 3" at 800 dpi -- an image 4" x 6" at 400 dpi -- an image 8" x 12" at 200 dpi and so on. "Resize" is probably more of a purist's approach. There's no possibility of degrading the image in any way. "Resample" will either create new pixels (by interpolation) or throw them away (by averaging and decimation.) With "Resample" an entirely new image is created for you. Finally... bear in mind that the scanner's rated dpi has almost nothing to do with sharpness. I can prove to you easily that the 1640's so-called "1600 dpi" yields an image much less sharp than a Polaroid SprintScan Plus working at 1350 dpi -- half its rated resolution. rafe b.
filmscanners: Scanner Comparisons
My two cents. The scanner comparisons just posted by Lawrence Smith seem to have sent the armchair quarterbacks into a tizzy. Not unlike a set of scans (Leafscan vs. Nikon 8000) that were discussed to death on this list just a couple of weeks ago. To me, these comparisons say more about how similar the scanners are than how different. And I can easily believe that all three of these scanners -- Leaf 45, SprintScan 120, and 8000 ED -- are pretty much in the same league. I've worked with two of these three. It makes little sense to analyze to death subtle tonal differences, especially when viewing JPGs from a web page. Any of these can be easily caused by minor changes in scanner-driver settings. The default behavior of the two scanners (in terms of tonality) could easily account for any and all such changes. Bear in mind that the quality of your scans has a lot to do with *you* -- assuming that the driver software has at least a minimal, common, and necessary set of controls. I recently was sent a pair of JPGs comparing a SprintScan 4000 scan and one done on a Scitex Everfast (a $30K machine, or at the very least, way out of my price range.) The fellow who sent me the scans was rather determined that the SS 4000 scan was markedly inferior in both resolution and dynamic range. With ten minutes of mask-making and curve-drawing in Photoshop, I was able to restore much of the detail to the blown-out highlights in the SS 4000 scan. I guess a few years of experience (working with less-than-perfect imaging gear) has taught me how to avoid some of the common pitfalls. Oddly enough (and pleasantly enough) -- with each new generation of scanner I buy, the gear improves to the point where some of the cool tricks I've learned become moot. Guess what I'm saying is: don't fret the tools too much. These discussions tend to degenerate into Ford-Chevy, Canon-Nikon arguments. Unfortunately, some things that really matter aren't known at this time: how long will each of these machines last? How will the service be, when you need it? How will the units behave on really trashy input -- curved media, scratched media, dense images, thin images, too contrasty, etc. And there are other aspects (of both scanners) that can be known now, but aren't revealed by sample scans on a web page -- the quality and usability of the software, the quality of the mechanics, the documentation, etc. Many moons ago as part of a "Traveling Portfolio" I was truly honored when a pro photographer opined that my prints appeared to have been drum scanned -- when in fact they'd been scanned on Microtek (!!) and/or Polaroid desktop film scanners. Why did I swap scanner brands this time around? Let's just say I went with the Devil I Don't Know. rafe b.
RE: filmscanners: Nikon 8000ED
At 03:53 PM 7/6/01 -0600, Frank Nichols wrote: > I took at look at the detail comparison crops (of the label) in >Photoshop at 1200%. The SS120 has much smoother transitions in colors >(softer?) while the Nikon 8000ED has sharper variations between pixels. I >converted both to greyscale and measured the levels at various points and >the SS120 seems to have slightly higher contrast. Which would explain why >it appears to me that the SS120 shows slightly more grain. Noise levels in >both appear to be quite low - I am green with envy (I expect the jpeg >compression affected both about the same...) /fn-Original Usually, increased contrast yields at least the perception of increased sharpness. If the SS-120 has higher contrast yet has lower apparent sharpness (on the close-up images) that suggests (to me) that the 8000 has much better optical resolution. Not that I'm an unbiased observer . rafe b.
Re: filmscanners: Nikon LS IV/Nikoscan 3.0
At 10:29 AM 7/6/01 +0200, you wrote: >Hi! > >I've recently purchased a Nikon LS IV scanner. It was delivered with >Nikonscan 3.0. While the scanner is ok, I don't like the software, since it >crashes all the time and I have to restart my computer and I hate that. I >heard the new version 3.1 is better. How can I get it? The salesman told me >that new version could be downloaded from internet, but he couldn't give more >details. If so, where and how could I get this program? >My system is a Pentium III, 1 GHz, 512 Mb SDRAM, Matrox G450 32 Mb. > >Thanks a lot in advance, >Claudiu Try: http://www.nikontechusa.com or http://www.nikon-euro.com/ I've had mostly good luck with NikonScan 3.1, though there were problems with the installation. Specifically -- there were conflicts with some existing peripherals that caused NikonScan to run very slowly. But no crashes, ever -- then or now. Be on the lookout for conflicts with existing SCSI and/or USB devices. Also, potential conflicts with Wacom pen pads. I don't know if the LS IV is Firewire or not, but if it is, there are a couple of additional gotchas: you need to be running Win98 2nd Edition or later, and you need to apply an OS patch to make Firewire behave, if it is Win98 SE. The patch was supplied on my NikonScan CD, or you can download it from Microsoft's website -- www.microsoft.com/windows98/downloads and search for "242975USA8.EXE" or maybe "IEEE 1394" Like I said -- it was a messy installation, but now that the dust has settled, I like this program a lot. rafe b.
RE: filmscanners: Canoscan 2400 UF
At 07:22 AM 7/6/01 +0100, Derek Clarke wrote: >Er how do you know? Have you tried it? > >It's interesting, because this is a flatbed that's snuck onto the market >with a built in proper transparency lid, 2400x4800dpi optical resolution >and 48 bits colour depth. This is an incredible spec for a flatbed, >especially at the price. > >About the only Achilles heel is lack of speed. Making it USB-only costs. I'm not Jack, but I'll offer this response -- 1. There have been negative reviews of the 2400 UF recently on usenet (eg. comp.periphs.scanners.) Do a deja search. 2. 2400 x 4800 optical, for $450, is a pipe dream, IMHO. 3. Specs have been "upped" from even the Epson 1600,1640,1680 family (rated 1600 x 3200) and it's known that the Epsons don't come close to their rated specs either. 4. You say "2400 x 4800 optical resolution." That's not at all accurate. More accurately: "2400 x 4800 non-interpolated, as far as we know." There is no scanner vendor that I know of that says anything at all about their optical resolution -- in the sense of providing meaningful metrics about the optics inside the box. The specs they do give only describe the maximum theoretical resolution imposed by the CCD sensor and the mechanics of the scanning mechanism. The spec gamesmanship among scanner manufacturers is outrageous. Not unlike the gamesmanship in audio gear. (How many different ways are there to describe the power output of an audio amplifier? Let me count the ways...) The manufacturers always find a way to inflate the specs. rafe b.
Re: filmscanners: GEM
At 04:20 PM 6/19/01 -0400, Norman Quinn wrote: > "GEM and ROC do not require hardware, but have to be "tuned" to the > So, in theory, any scanner could have GEM > but it requires that Applied >Science Fiction be hired to make the profiles, and that the scanner >company pay a licensing fee to them." > >What scanners come with GEM and ROC. Is Nikon the only scanner with ICE? I believe the Canon 4000 (or was it the newly announced Minolta MF scanner?) that also had ROC and GEM. IMO, these two aren't nearly as useful as ICE. They're OK if you want to fix a *really* bad slide or negative in a hurry. But they don't really accomplish anything that couldn't be done by hand, by a skilled operator. If used blindly, they can create ugly artifacts. You could even argue that ICE is that way also (ie., scans can be retouched "by hand.") But the time required to do that, on a really mangled image, would be prohibitive. What impresses me most is that there seems to be little or no penalty (in terms of image sharpness) for using ICE. rafe b.
Re: filmscanners: Nikon 8000ED
At 07:44 PM 7/5/01 +0100, Ian Lyons wrote: >PS: I think the silence has just been broken, or maybe as a VERY satisfied >SS120 user I just needed to crow and let you Nikon users know that the grass >IS greener on the other side of the fence and judging by some of ex Nikon >8000 users on the list; the ICE has already began to melt Ian, I'm not going to respond to your post point-by-point, though it's tempting, and could be fun. Another poster asked, "Why the silence from the 8000 users." This struck me as odd, given that Lawrence and I have been not at all silent, and yet -- in the month or so that I've been back on this list -- I hadn't heard a peep from any LS-120 users. Mr. Hemingway mentions that reviews are posted somewhere on the web, but that's not the point; I was curious to hear a "warts and all" discussion of the LS-120, here on this list, from an "ordinary" user such as myself. Finally -- an objective, professional review would probably come off more believable and palatable if one abstained from expressions like "this machine wipes the floor with Brand X" or, "I'm real sorry for you Brand X users." rafe b.
Re: filmscanners: Film grain vs 2700 DPI scan resolution
At 12:07 AM 7/2/01 +0100, Tony Sleep wrote: >On Sat, 30 Jun 2001 19:26:33 -0600 Frank Nichols ([EMAIL PROTECTED]) >wrote: > >> 1. With 200 film, is the grain "large" enough for the 2700 DPI to >> record it? >> If so could some one describe it (or email me a couple scan clips >> showing >> examples?) > >No, IME you are unlikely to see genuine grain off ISO200 Fuji using >2700ppi. You are likely to get some aliasing which looks like grain. A way >to check this would be to have a reasonably large C41 print made (say >12x8"). I'll bet. Well, I don't know if it's grain or an alias of grain, and not sure I'd recognize the difference. To me, it just looks like grain. Not just on the print, but on screen as well. To be honest, I never did color darkroom work on C-prints, just a bit of Cibachrome printing. But I sure have seen my share of 35 mm, Tri-X, pushed to obscene speeds and printed at 8x10" and 11x14", on an Omega B22 condenser enlarger. Whatever this stuff is that I'm seeing, it clearly correlates to film ISO rating and quality... just like grain. So I guess to my feeble mind, simple "grain" is an adequate model for what I'm seeing. Not having the wherewithal to get 8000 dpi scans of my images, this simple (though flawed) model will suffice until new observations warrant its replacement. rafe b.
Re: filmscanners: Problems with LS40 and Nikon Scan 3
At 09:58 PM 7/1/01 +0100, Jawed wrote: [re: NikonScan peculiarities:] >2. how can I persuade NS not to leave a gap of 20ish where the shadow data >should be filling in? It seems to me that if I don't set the black point on >incoming scans, the images have a good overall tonality - but somewhat milky >in the shadow range. I see this over here also (on the 8000, with NS 3.1) and it is a bit odd, yes, but not that evil. Sometimes I'll manually intervene in the Curves tool to move the black point to a few points below the darkest pixel. Sometimes I just leave it that way and fix it up later in Photoshop. Whatever it's doing, NS seems to do the right thing *most* of the time on my color negatives. I haven't really learned how to outsmart it yet, nor do I feel any urgent need to "correct" its choices about color settins. rafe b.
Re: filmscanners: exposing C41 for scanning ( was gibberish
At 06:46 PM 6/30/01 +0100, you wrote: >On Sat, 30 Jun 2001 07:26:58 -0400 rafeb ([EMAIL PROTECTED]) wrote: > >> I'm hoping (without evidence) that you're mistaken about >> the swiveling LEDs. A 645 negative is approximately >> 7,000 scan lines along the length of the strip (4.5 cm, >> at 4000 dpi) and that would mean 21,000 mechanical >> motions of the LED array (or 28,000 if you add the IR >> channel.) >> >> I much prefer to believe they're switching the illuminant >> colors electronically. > >Rafe, please don't take what I said as gospel - I'm relying on my raddled >memory of a description I saw of the LS1000 mechanism many years ago, and >may well have it *completely* wrong. > >What I *think* I recall is that the LED array is both switched >electronically and moved mechanically at each scan line position, so the >successive R, G & B exposures are made with R, G & B LED's and monochrome >strip sensor 'looking at' precisely the same strip of film image 1 pixel >high. IOW the LED array is positioned differently for each channel >exposure, the relevant LED's turned on for the exposure duration, and then >the cycle repeated for the next scan line, and so on. Ah, Tony, but I do take your word as gospel... Your word is surely as good as anyone else's around here. I guess I was thinking that the LEDs are small enough (and hopefully diffused somewhow) so that mechanical re-positioning is not required. I am sorely tempted, but I dare not open up my scanner to investigate further. I will admit that the noises emanating from this scanner suggest mechanical goings-on that I'd rather not speculate on. IOW, a coarse clicking/rattling/grating noise that seems much too coarse to correspond to steps at 4000 dpi. I've never heard another scanner (either film or flatbed) make noises like that. rafe b.
Re: filmscanners: Film grain vs 2700 DPI scan resolution
At 07:26 PM 6/30/01 -0600, Frank Nichols wrote: >1. With 200 film, is the grain "large" enough for the 2700 DPI to record it? >If so could some one describe it (or email me a couple scan clips showing >examples?) To answer your question. Yes. Frank, the biggest single improvement in my photo "technique" these last couple of years was giving up on generic ISO 200 negative films. I may have stumbled on to that discovery, but the difference in the scans was huge. And I made this discovery (and saw the vast improvement in the scans) while using a 1950 dpi film scanner (Microtek 35t+). Try a roll of Fuji Reala or Kodak Royal Gold and see if you don't agree. Both are ISO 100. As I write this, I'm scanning one of my favorite (old) images to see if the Nikon can improve on earlier scans. What a pity -- the image is on Kodak Gold 400, and the grain is just awful. rafe b.
RE: filmscanners: exposing C41 for scanning ( was gibberish header)
At 01:04 PM 6/30/01 -, Lynn Allen wrote: >Rafe wrote: > >>Shoulda listend to my wife. She said to give up >>on film, get a digital camera. > >Hope Rafe has a good, sturdy kitchen table! ;-) >--LRA Huh? Sorry, that one went right over my head. rafe b.
