Scanning old slides

Gerald,

What I really need is to be able to show you a sample, but I don’t have a web site to post to.

In maybe 30 minutes or so you can get your photos available here:

<http://www.pbase.com/help/uploading>

The first 30 days is free and you aren’t obligated to spend anything.

— Burton —

Gerald…

Not sure as i don’t use Kodachrome… are you using digital ICE? If so, that may be the problem. I think, i may be wrong though, that ICE cannot be used on Kodachromes.

Also, suggest you try the comp.periphs.scanners and
alt.comp.periphs.scanners newsgroups. Quite a few helpful people there as well.

Cheers…

JJ

Gerald,

If it wasn’t that I had scanned some slides done by an expert photographer, I would have thought there was something wrong with my scanner. However the slides from the pro were dynamic in there appearance. Is it possible that what I see on a projection screen isn’t really as sharp as I think, or am I overlooking something?

If you still have some of the professional’s slides, how do they look when projected compared to your old slides?

I have used a low power microscope to examine slides and inkjet prints. Something like that would let you compare the pro’s slides versus your old ones.

A professional photographer can afford extra sharp expensive lenses, use a steady tripod, and a quality camera to produce some really high quality slides, far better than typical amateur handheld snapshots. However, considering that your slides look pretty good when projected should count for something. A projector can show image defects in big detail. Are you taking care to have the emulsion side of the slides on the same side each time? I assume these slides are not mounted in glass mounts.

I haven’t printed any of the images in larger sizes yet to see the results, partly because I need a newer printer than my trusty Epson Color 800.

You can get a decent preview of what your 13" x 19" prints might look like, detailwise, by cropping a piece of your image off and printing that piece on your Epson Stylus 800 printer. Or you could preview the whole 13×19 print by doing a tiled print of it on your Epson 800. The Epson 800’s colors and details won’t be as good as what you would get from the Epson Stylus Photo 2200, but you still should be able to preview how much detail you would get on an Epson 2200.

— Burton —

Gerald,

Is the emulsion on the right side? Try to reverse the slide. If the autofocus of the scanner uses the film surface, and the emulsion is on the other side, it may get blurred.

I had this with some contact-copied material. Scanning them reversed and re-mirroring them in PS worked fine.

Rob

Gerald,

Did the ‘expert photographer’ use Kodachrome. Make A comparative test scan with Ektachrome or Fujichrome.

Kodachrome is a diffrent type of film and I have often felt that it does not scan well.

Jeff

The Coolscan 4000 has had focus problems related to its design. I bought one just over 2 years ago not long after they were released and ended up going through 6 of them before I got one that worked properly. I couldn’t get proper focus across a slide related to the tiny depth of field of the lens. The first one I got was by far the worst but each one I had was different. I ended up bringing them home two at a time and testing them until I found one that was satisfactory. There was a lot of variability between the different scanners.

Firstly, check if any of the slide is in focus. If not, you probably have a duff scanner. If you are getting sharp focus in some areas, check out focus across the slide by rotating the slide 180 degrees and comparing the results. This is best done by putting the two scans on different layers in PS and flicking on and off visibility.

Poor focus across a slide would be greatly exaggerated if you are using old mounts as the slides themselves may be warped. You may need to remount in solid plastic mounts. If using autofocus, setting the number about 3 higher than indicated often gives sharper results across the slide if the slide is pretty flat.

PS. ICE doesn’t work with Kodachrome.

Thanks so much for the responses.

Jay Jhabrix said,

Not sure as i don’t use Kodachrome… are you using digital ICE? If so, that may be the problem. I think, i may be wrong though, that ICE cannot be used on Kodachromes.

I know that it is reported that certain brands are a problem, but I can’t remember right now which ones. However, at Burton’s suggestion, I set up a temporary page so that everyone could look at what I’m talking about.

Check out: <http://www.pbase.com/gshantz/slide_scanning>

You’ll notice there is a little less softening of the edges without Digital Ice, but oh what a job if I didn’t use it! Honest, I really DID try to get the film clean. And it LOOKED clean under low magnification. I always use loops for things like that. My 60 year old eyes don’t quite do what they did a few decades ago.

Burton Ogden said,

If you still have some of the professional’s slides, how do they look when projected compared to your old slides?

Good question. I drove into the office and borrowed the slides AND digital projector from my boss. I spent half a day setting up both projectors side by side and did some extensive comparisons. What I learned made me a little ill at the loss of quality from the whole digitizing process. The clarity and depth of field and even color matching are a LONG way from ideal. (How I wish I could afford a good color management setup.) While the doctor’s slides were overall better in sharpness, the Kodachrome slides I am working with, were still much sharper from the slide projector than anything I have scanned. Even the scan without Digital Ice. I felt that my monitor (19" ViewSonic G90fb) looked quite a bit better than the projected image. (It’s a Canon LV X2 – 1200 dpi res.)

Are you taking care to have the emulsion side of the slides on the same side each time? I assume these slides are not mounted in glass mounts.

Yes, always. No, just paper mounts and the film is quite warped. I have wondered just how the focus works on these scanners, but if they pick a certain point, like the middle, wouldn’t the edges be either more in or out of focus? I can’t detect this, so does it average? I guess I should save all this for another forum, but these are things I wonder about. I posted here with the thinking that I was getting as good a scan as possible and wanting to be able to salvage the image with PS. So far, I’m not doing too good.

Rob Keijzer said,

Is the emulsion on the right side? Try to reverse the slide. If the autofocus of the scanner uses the film surface, and the emulsion is on the other side, it may get blurred.

I had this with some contact-copied material. Scanning them reversed and re-mirroring them in PS worked fine.

Rob, I tried this just to see, and I couldn’t tell any difference. The focal point was decidedly different, but the image looked about the same after flipping it.

Jeff Darken said,

Did the ‘expert photographer’ use Kodachrome. Make A comparative test scan with Ektachrome or Fujichrome.

Kodachrome is a diffrent type of film and I have often felt that it does not scan well.

I’m thinking you may be right on that. The good looking slide is a copy and it says Fujichrome on the mounting.

Mick Murphy said,

The Coolscan 4000 has had focus problems related to its design. I bought one just over 2 years ago not long after they were released and ended up going through 6 of them before I got one that worked properly.

This doesn’t sound particularly encouraging, but I did a LOT of research on the unit before purchasing approx. 6 months ago, and I got a tremendous amount of good press and almost no bad press. As you can see from the one slide I linked to above, I feel it CAN do a grand job with the right film.

Firstly, check if any of the slide is in focus. If not, you probably have a duff scanner. If you are getting sharp focus in some areas, check out focus across the slide by rotating the slide 180 degrees and comparing the results.

Mick, I’ve scrutinized the images very carefully for this, and I can’t tell ANY difference in focus on any part of it. However, as previously mentioned, look at the scan of the girl on my link. I know it is a rather unfair comparison to my group photo in some ways, but you can see the detail and sharpness so well, I feel confident the scanner is doing it’s job.

Poor focus across a slide would be greatly exaggerated if you are using old mounts as the slides themselves may be warped. You may need to remount in solid plastic mounts.

My gut instincts are that this may be the major factor. However, I know nothing about mounting slides. Is this difficult? It sounds very time consuming – of course not nearly as much as fixing the images afterwards. Do I need special tools to do this?

If using autofocus, setting the number about 3 higher than indicated often gives sharper results across the slide if the slide is pretty flat.

Although the film wasn’t flat, I tried doing a manual focus, and I couldn’t do any better than the autofocus.

PS. ICE doesn’t work with Kodachrome.

Please note my comment at the top of this post.

I will pursue the scanning aspects of this topic on one of the other groups, but suppose I can’t do any better with the actual scan? What can you folks recommend for improving the images? I have tried several techniques to mask and USM, but it seems like trying to make a silk purse out of a pigs ear.

Thanks again.

Gerald: it looks like a dirty lens in the scanner, have you tried cleaning it?, maybe condensation on the lens.
Jorge

Gerald,

I have wondered just how the focus works on these scanners, but if they pick a certain point, like the middle, wouldn’t the edges be either more in or out of focus? I can’t detect this, so does it average?

I have always been distressed by how "non-flat" my slides are. Only the mount is flat. However, the hype that came with my slide projector said that the projection lens was specially designed to have a curved focal plane that coincided with the curved shape of slides. And it seemed to, because it projected pretty sharp, curved edge to curved edge.

I don’t think it is possible that the optics of the Nikon Super Coolscan 4000 ED also have a curved focal plane to correspond with the average slide shape, simply because, as I understand it, the lens is a tiny thing that focuses on just a small spot at a time. And besides, if a curved focal plane were involved, when you flipped that slide that would have reversed the curvature of the slide and played havoc with any curved focal plane.

Do you think it is possible that your Coolscan is continuously re-focusing on the scanning point? I guess that isn’t beyond the realm of possibility. Or maybe the scanner lens is really very slow and has a lot of depth of field.

You certainly made your point about ICE working with your Kodachrome. I never cease to be amazed by how much crap survives my attempts to clean my prints before scanning them. And I also have to reclean my scanner glass each time. Would ICE also remove dust from your scanner’s glass plate? Or is this scanner a design that doesn’t even have a glass plate? Did you also try Focus Magic as an alternative to Photoshop’s Unsharp Mask on this slide?

— Burton —

Gerald

This is a quote from Nikon Scan Manual (Acrobat Version on CD) P52

Digital ICE does not perform equally well with all types of film.

If you are using Kodachrome film, you may find that Digital ICE does not have the desired effect, or that there is an overall degradation in image quality when Digital ICE is in effect. We recommend you turn Digital ICE off when using black-and-white
(color monochrome film excepted) or Kodachrome film. >

I’ve never managed to succesfully remove dust specs using Kodachrome and ICE so I don’t know what you did but it certainly is remarkable looking at your images.

There was a lot of stuff on the net a few years ago about focus problems when the scanner and its big brother the LS8000 came out but perhaps Nikon have rectified this or improved their quality control. The LS8000 was much worse because it takes medium format which is even harder to keep flat.

Using a Minolta Dimage 5400 scanner on 34 year-old Kodachrome 25 slides taken with a Nikon F and Micro-Nikkor lens at f/32 or f/45, the images are sharp as a tack. Printed on Epson 2200 at 13" by 19" show no signs of grain anywhere. I was blown away by the image quality. Of course, the images are huge tif files. Each is 91 MB. I scan at 360 dpi, 8 bit color and 6836 x 4666 pixels (varies slightly depending upon scanner crop). I usually use Digital Ice and there is no noticeable image degradation or loss of sharpness. 95% of the time I use the autofocus and autoexposure (they’re well-exposed slides).

Must be the scanner. It’s not the Kodachrome. BTW, the new(!) Kodachrome 64 is not as sharp as the old Kodachrome 25 or 10 (am I dating myself?). Good, but not as sharp.

A.B.,

Using a Minolta Dimage 5400 scanner on 34 year-old Kodachrome 25 slides taken with a Nikon F and Micro-Nikkor lens at f/32 or f/45, the images are sharp as a tack.

Glad to hear that. The Minolta Dimage Scan Elite 5400 is at the top of my wish list. Next I will need a sharper SLR lens.

— Burton —

Boy, I looked at the specs and price of the Minolta Dimage 5400 scanner. Combined with your experience abclapp, and the price, it looks like a winner in it’s class. Only about 2/3 of what I paid!

Burton Ogden said, Would ICE also remove dust from your scanner’s glass plate? Or is this scanner a design that doesn’t even have a glass plate?

There is no glass plate on the CoolScan 4000. From what I’ve read, I believe ICE uses a separate UV light to detect specs and scratches. I noticed when I told it to recalibrate, that it made 3 passes with 3 different colors of light. The light passes down directly through the film onto whatever is below – mirror or CCD.

Did you also try Focus Magic as an alternative to Photoshop’s Unsharp Mask on this slide?

Yes. Because even Focus Magic can do unpleasant things to graininess and texture, I even used it on an edge mask. Dissappointing results on this slide, but I have used it on other images with pleasing results.

2 questions I had, are still begging. One was, is there any PS magic I can do to improve what I have, And does anyone know the intricasies of slide mounting?

Gerald,

Try the news groups i suggested in my earlier post. comps.periphs.scanners has more traffic than alt..comp.periphs.scanners.

Nikon scanners are definitely considered very good…. but they do have that hassle of small depth of field. your slides need to be flat…

here’s a link you may find useful:

http://members.austarmetro.com.au/~julian/photography/ls2000 -focus.htm

it’s about the ls 2000 but applies as well to the 4000. i’ve worked out an excel file that gives me the mean dof… email me dierctly and i can send it to you.

The Minolta Dimage Scan Elite 5400 is at the top of my wish list. Next I will need a sharper SLR lens.

I have the 5400, it’s a great scanner, if a little slow on my old system. Next I will need a faster computer 🙂

Gerald,

Looking at your straight, unretouched scan, the film grain (and dust!) appears sharp. Ergo: it is the original photographic image, not the scan, that is unsharp.

Fred.

wrote:

Is it possible that what I see on a projection screen isn’t really as sharp as I think, or am I overlooking something?

It is in fact quite possible that this is the case. In general, projector lenses aren’t particularly sharp, especially the inexpensive ones that come with your run-of-the-mill Kodak Carousel. After some measure of frustration I finally bought a Schneider lens for my projector, which was a significant improvement over the Ektanar lens it came with (which I believe has plastic lens elements), but even it is not as sharp as I would like. A couple years ago I spoke with a Zeiss rep about the Schneider lens he was using for a presentation, and he said it was the best he could find but that it was really not as good as he would have liked (although in fairness there was a good measure of Zeiss chauvinism in his
remarks). We tend to view projected slides from a distance and are impressed by the size of the images, and often don’t notice the lack of real sharpness in the projected image. The bottom line is that you can’t really judge the quality of an image using a slide projector–you need to look at it on a light box with a quality loupe (the Schneider 4x loupe is probably the least expensive quality loupe currently available). In a pinch you can use one of your SLR lenses as a loupe, but that’s not really a long-term solution.

I have a lot of Kodachrome 64 slides from the ’70s, and while they are fine for projection and probably for optical enlargement as well, the few I have tried to scan (with my somewhat aging Minolta Scan Dual) have not come out particularly well, showing some of the characteristics that you and others have mentioned here. In particular they were quite dark and did not clean up very well, even using Edwal Anti-Stat film cleaner and compressed "air." I believe this is due to the unique nature of the Kodachrome emulsion. You may recall that a couple years ago Kodak introduced an line of pro negative films (Supra, now discontinued) which were claimed to be optimized for scanning, and while scanning negs poses
different problems than does scanning transparencies (and some scanners seem to do better with one than with the other), this would suggest that the emulsion has a lot to do with the outcome of scanning. This may have something to do with the dyes, but also I believe (but am open to correction) that Kodachrome is a more dense emulsion than modern E6 films, which might account for why it is difficult to scan. And then there are other complications such as using the right color profile for the film and all that.

Finally, while it’s near-impossible to judge the quality of a photo or scan from a jpeg, it does seem that the Kodachrome image you posted isn’t very sharp to begin with. It’s true that the dirt on the slide isn’t particularly sharp either, although that could be a function of the conversion to jpeg; but to my eye at least the grain of the film doesn’t seem to be much if any fuzzier than the dirt, which suggests that the image is not sharp. This would not be surprising, given that the ASA rating of the film was either 25 or 64 and the photo appears to have been taken indoors without a flash, hence with a slow shutter speed. And of course we don’t know how sharp the camera lens was (I think it’s safe to say
that even inexpensive SLR lenses of today from the major manufacturers are sharper than most of what was available in the ’50s and ’60s, with the possible exception of Leica and Zeiss).

I have read other posts of people who have scanned a great many slides and film for a living, and am wondering if any of you would be kind enough to give me some perspective. I think my biggest struggle is with what I see in a slide viewer or on a projector screen, with what I see on my monitor. Possibly my biggest enemy is my own perfectionist ways. In any case, I look forward to your input.

I know you’ve said you don’t want to spend a lot of time on each slide, but this may be unrealistic. From my perspective, 20-30 minutes per image is not much time at all–I figure if I can get a reasonably good result with a single image in an afternoon I’ve done well (I make no claims to being an expert Photoshopper). I’ve also found that I prefer not to do any adjustments in the scanner softwarewith possible exception of gamma, preferring to use PS for all that, as it seems to do a better job. So I make raw "master" scans at the scanner’s maximum resolution at 16 bits per color (the scanner only does 10 bits, but VueScan only offers 8 and 16, and the scanner’s own software only offers 8) and archive them
as .tif files, saving any edited versions as separate files. YMMV. 🙂

One other thing to keep in mind is that the image you see on your PC monitor is only as sharp as the monitor and video card can produce, so if either of these are sub-par, you won’t get a good read on just how sharp an image is. I have no experience with the model of monitor you are using, but a few years ago I ended up returning three ViewSonic monitors because they were not sharp and paying the premium price for a Nanao (Eizo) monitor, which I am still using with complete satisfaction. Coupled with a Matrox video card, my display is about as sharp as it gets–even those new-fangled LCD monitors on display at CompUSA don’t look so good when you switch from the fancy graphic to a text page (possibly because
they’re being driven by crappy video cards). It’s also been my experience (FWIW) that a top-quality shadow-mask CRT is better for both text and image editing than even the best Trinitron-type displays, because the former (the good ones) offer a sharper image (even though the Trinitrons tend to have more vibrant colors, making them popular with gamers). So depending on what you’re using now, you might want to consider an upgrade of your video card at the least. I’ve not seen any of the current crop of gaming video cards, but I have to wonder about their "2D" image quality, which is never mentioned in reviews (just their "3D" speed); my last experience with an ATI card was not a good one–the image quality
was dreadful–and nVidia has a long history of substandard 2D image quality as well (although I gather this has improved in recent years). By all accounts, on-board video (if that’s what you have) is inadequate for critical image-editing work or serious writing on the PC, with the possible exception of the built-in video on Athlon motherboards with an nVidia chipset. At the moment the best bang for the buck in video cards for image editing and text is probably the Matrox G550–not exactly cheap, but not absurdly priced like the gaming cards du jour. Regardless, you may want to try printing out a couple of those Kodachrome scans to see how they look on paper; if you do, try pushing the Unsharp Mask past the
point where the image looks good on screen, as often an image will print with less apparent sharpness than your screen image. (I know this seems a contradictory statement, but if you’re not getting a sufficiently sharp image on your monitor, this is one way to find out).

Hope this is of some use.

Craig