Which scanner should we use for slides?
June 29, 2011 12:13 PM   Subscribe

My museum is about to set off on a big slide-scanning project. After finding out how much it would cost to outsource the scanning, we've decided to just buy a higher-end scanner and throw an intern at it. I'd like some guidance as to which scanner to buy.

The Epson - Perfection V750-M PRO looks pretty good to me, but is expensive as hell. The Canon CanoScan 9000F Color Image Scanner isn't as good on the specs, but is a ton cheaper.

Has anybody used either scanner, or have any other suggestions or thoughts? One salient issue I can see is that the Epson can do 8 slides at a time while the Canon can do 4; that matters, but I'm not sure if it matters ~$500 worth.
posted by COBRA! to Technology (25 answers total) 6 users marked this as a favorite
If for no other reason, look at the time it will take.

8 slides vs 4 slides means, roughly, the Epson should take around half-as-much time to do the job.

Throw some weasel words about loading 4 vs 8, and you should still see an improvement using 8 at once vs 4 (say 1.8 instead of 2).

How many slides, what's the intern's hourly wage ? Will you save more than $500 viz fewer hours spent doing the task ?
posted by k5.user at 12:20 PM on June 29, 2011

regardless of how many it could do at once, you are going to need one heck of a computer that can handle processing 8 slides at once...

consider a slide/film scanner.. .designed for that task...
posted by fozzie33 at 12:32 PM on June 29, 2011 [1 favorite]

Do not use a flat bed scanner even with a film adapter. The keys to film scanning is dynamic range, optical resolution, and ICE. ( or something similar to help clean up scanned images). The $99.00 slide scanners and most flatbeds do not have the dynamic range or the optical resolution. So in a nutshell you either pay the piper to do the scans on a proper slide scanner or you buy a proper slide scanner. The Nikon Cool Scan series is expensive, but does what it is supposed to do.
posted by Gungho at 12:39 PM on June 29, 2011

My old man always said "If you're gonna do something, do it right" and "Use the right tool for the job." Do like fozzie33 said and get a scanner designed for the task.

Yes, I know your budget is probably tight, but you're presumably scanning for The Future; how long will these scans be around?

If you don't have a use for the more expensive scanner after this project, sell it. It's worth the investment.
posted by entropicamericana at 12:43 PM on June 29, 2011

Find out how long it will take to scan a batch. You may have to find people who use those machines as I can't find that information on the pages you linked.

We looked into doing a slide digitization project at the (art) library I worked at but decided against it because the slid escanner we had took about 2 minutes per slide, I think, and we had about 25,000 slides. That's almost six months of 35 hours a week of nothing but scanning.

Other things to consider when looking at the time - the intern will have to keep track of what the slides are, in some kind of catalogue or file system. The computer will have to be able to handle that many images - and you want high quality, so that's big files. Make sure the intern is responsible - not a slacker - and knows how to deal with slides i.e. don't touch the image, don't drop them, etc.

(We decided against scanning for those reasons and others. Instead we purchased archival storage for the slides.)
posted by kyla at 12:47 PM on June 29, 2011

$800 for a "big slide-scanning project" doesn't seem that expensive to me.

As the above said, if you're going to do it, do it right.

I've seen too many of well-intentioned projects like this end up with unusable end results due to scrimping or lack of knowledge on the technical side.
posted by davey_darling at 12:47 PM on June 29, 2011

I run what used to be a slide library at a large university. We have digitized well over 100k slides in the past ten years. At this point, when I have anything over a few dozen slides, I farm the scanning out to a vendor. It really is much easier and worth it, despite what you may see as excessive prices. A good imaging lab can use the appropriate tools and deliver within specified times and to your particular needs. When I use interns or student labor, it's a total crapshoot as to timing, quality and return on my time, even when I've spent hours training them on scanning, file management, etc. It's not about what scanner you use, that's just one part of the process.

Memail me for some recommendations for labs we use - there are places that specifically cater to cultural institutions and understand the tradeoffs between different parts of the process.

FWIW in the archive I use Nikon coolscan 5000 scanners, but I think they are no longer available.
My vendor delivers my images to me on hard drives, color-corrected, with attached lightroom catalogs, production metadata already embedded in the files, at the sizes and formats I need. I let him worry about the choice of scanner/workflow/etc.
posted by gyusan at 12:48 PM on June 29, 2011

My experience with slide scanners is over a decade old, so may be outdated, but my advice would be, based on that experience, to buy two slide scanners. Why? As we discovered, when setting out upon a digital imaging project in the slide collection I curated at the time, the sorts of scanners that are affordable for the general public, as opposed to the $10K+ ones, are not designed for this sort of large-scale scanning project.

You get two scanners, so that your intern will be able to continue the project when you send the first scanner off to be repaired. We ended up sending the scanners off fairly routinely every couple of months or so, luckily within the warranty period. Once it went out of warranty, it was cheaper to buy a replacement scanner instead of sending the broken ones out to be fixed. When I left that position, we had three student workers and four scanners, one of which was out getting repairs. As soon as it came back, we gave it to the person using the next scanner to break.

Also, do not get a flatbed. Get the sort of scanner that allows you to stick a large stack of slides into a mechanism that will feed them into the scanner. You will prevent a multitude of mistakes, skipped slides, duplicate slides, etc. We had slide scanners that scanned four slides at a time, and we had students putting the slides in with the wrong orientation (so that the top and bottom of the slide was cut off, and which wasn't discovered until after he scanned several hundred slides), losing track of where they were in the collection so skipping slides or duplicating them, etc.

And finally, if I were facing a huge program like that now, I'd contract the scanning out. It's well worth the money not to have the constant. hassle.
posted by telophase at 12:52 PM on June 29, 2011

Thanks for the replies so far...

To elaborate a little on a couple of things: in this case, we're talking about 2000 slides. Local vendors have said about $2k to do the job. Part of the reason we've thought of picking up a flatbed is that we could also use one around the office... so, while I get that a slide-specific scanner is ideally better than a flatbed, I'm curious what the margin between the two is. Is it possible to get decent-but-imperfect results from a flatbed, or are you just destined for crap, even with a good flatbed?

The other thing is that this isn't actually my project, and I'm not 100% sure what quality level they need for the digitization (these are mostly installation shots of groups of objects in galleries, and all of the objects have already had high-quality photography done).
posted by COBRA! at 12:54 PM on June 29, 2011

The V750, at least, is a great flatbed scanner but completely wrong for the task at hand.

I've used the Nikon Coolscan 5000, as mentioned by Gungho. With good software like Silverfast, it's eminently possible to produce fantastic, archival quality scans. If that's the intended purpose of your project, you're going to need to invest into an expensive scanner.

I'd also look at what it's going to cost to pay people for the time taken to scan the slides. We found that, at best, we were getting approximately 1 slide every 2 minutes. This is comprised of both the scanning time, and hand-calibrating the few necessary settings for each individual slide. In a realistic working environment, it goes down to 1 every 4-10 minutes.

Kodak did do an excellent batch scanner. I can't remember what's called right now, but with a good operator it's a bit faster than the Coolscan, and capable of some excellent results.

We ended up shipping out a few thousand of the slides to an excellent company. Unfortunately, it's in the UK, so not a useful recommendation. About 99% of the returned files were perfect, and the cost was certainly no more than doing it in-house.

On Preview, I see that you've received some excellent advice from others, but I thought I'd add some additional datapoints.
posted by Magnakai at 12:55 PM on June 29, 2011

Local vendors have said about $2k to do the job.
If they produce quality results, that sounds like a good price to me.

If anyone's interested, the Kodak scanner that I couldn't remember is called the Kodak HR500.
posted by Magnakai at 1:04 PM on June 29, 2011

Destined for crap. Think this through in another context -- "hey, we can buy and hire an intern and do it ourselves!"

Bad idea.

posted by rr at 1:30 PM on June 29, 2011

Do it right, which includes capturing all available metadata about the images (title, artist, date, description, etc...). Once scanned, the original slides are probably going to be lost or left to rot. It's a lot more worthwhile to do the job right the first time around than to scan them all over again later, plus the cost of restoration to deal with the old decaying slides, assuming it's even possible to re-do it later.
posted by zachlipton at 2:01 PM on June 29, 2011

It's implicit in the project that the intern'll be going back and working out the metadata after they're digitized. That's been the defined Stage 2 for a while.
posted by COBRA! at 2:11 PM on June 29, 2011

I have to echo some of the other sentiments expressed here. I've seen plenty of in house slide scanning projects go to naught. Dumping a ton of uncategorized images into a big folder with no decipherable metadata included, poorly scanned images, etc.

2000 slides isn't a ton to scan in house but you end up needing to make a project plan, a categorization scheme, train a monkey (intern) and then watch them to make sure they aren't doing a shit job that someone else will have to finish or even worse redo from scrap.

Sometimes it's worth it just to come up with a set of expectations and give the job to a set of professionals that will give you a quality product.
posted by vuron at 2:30 PM on June 29, 2011

One of my in-laws found a bunch of color glass and Kodachrome slides from the 30s and 40s stored in their attic. I bought a CanoScan8800F on sale at Amazon for a reasonable price and offered to scan them.

Pros: I am happy with the quality from the CanoScan. It was super easy to figure out how to use. I can scan the slides at a high resolution. For the glass slides, we did not want to attempt shipping them somewhere for scanning. I use the flatbed scanner a lot. I can also scan negatives with my CanoScan.

Cons: It only does 4 slides at a time. It is a skull-numbingly brutally slow workflow. All the slides I had needed to be carefully dusted off with a soft paint brush. I couldn't really do anything else on the computer while the slides were scanning. The scanner takes awhile to scan when set to high res. So there is a lot of time just sitting and waiting. Then all the saving and adding data took some time. Some of my slide images needed color correction or dust specs removed from the final image.

Even working on this project in the comfort of my home, with a beer in my hand and a movie playing while I worked, I about went crazy from the monotony. Out of about a thousand slides, I ended up getting 200 or so fully scanned and processed before giving up. Paying someone to scan all of these for $1 a slide, is a really good deal. Then you can have your intern add the slide info into your database after they have been scanned.
posted by pluckysparrow at 3:40 PM on June 29, 2011

I've seen this kind of work quoted at $7/image. $1/image is good value.
posted by flabdablet at 4:33 PM on June 29, 2011

Again, thanks everybody. I was brought into this purely in the "we're buying a scanner, what scanner should we buy?" mode-- when asked earlier, I suggested outsourcing. I'll use this thread to reiterate suggesting outsourcing; but I also really appreciate any scanner-specific talk.
posted by COBRA! at 6:36 PM on June 29, 2011

any money you spend on a scanner and training an intern will be wasted. seriously. scanning is an art to itself. You really need to know what you are doing to make results that look decent.
posted by sully75 at 7:46 PM on June 29, 2011

Get a Nikon CoolScanner used on ebay.
Ebay it when you're done.

Take the money you save and take the office out for drinks.

Have everyone wear black and white because someone on the internet told you to do it.
They'll love it.
posted by Murray M at 7:57 PM on June 29, 2011

Could you rent? It might be cheaper to rent a high end scanner for the duration of the project rather than buy a lower grade one for keeps.
posted by freya_lamb at 2:22 AM on June 30, 2011

Murray M: "Get a Nikon CoolScanner used on ebay.
Ebay it when you're done.

If you're determined to do it in-house, then this is a good idea. They hold their value very well, since they're excellent and no longer made.
posted by Magnakai at 3:49 AM on June 30, 2011

Scan Cafe does a very good job for .29 per slide. I believe they ship the slides offshore. That being said the batch of about 80 slides of mine they have done took about two months. As I write this I have my images, but am waiting for my originals to be shipped back.
posted by Gungho at 5:20 AM on June 30, 2011

follow up. Scan Cafe finally returned my originals in the exact same box I shipped them in!. They also apologized for the unusually long time it took. They said they were trying out a new shipping process.
All in all well worth the .29 per slide. The scans were high res and the contrast and exposures were spot on.
posted by Gungho at 5:24 AM on July 5, 2011

« Older anti-radiation maternity clothing   |   One Pot, One Love? Newer »
This thread is closed to new comments.