Drum scanning technology: Why so expensive?

[Terence Patrick]

To be honest, I don't really understand drum scanning. But I'm curious as to why the technology hasn't come down in price or become the a consumer technology the way flatbeds have. Are the machines that make drum scanning possible so complicated that only professional labs are able to afford/use them? Seems like everything in photography has come down in price, except perhaps MF digital backs, but drum scanning technology feels like it's stuck in time with nothing new over the horizon. From the few used drum scanners I see pop up every now and then, it requires the use of a Mac G4 with SCSI, and the scanner is still quite expensive. Perhaps it's my ignorance on the subject, but it is rather difficult to find new information (from my searches).

[robc]

they are not designed for consumer use. That means they are low volume sales which means the cost per unit is high. Simple economics. If they were like hard drives which work to equally fine tolerances but are sold in hundreds of millions then they would be cheap. The technology is nothing special by todays standards. Why aren't they consumer level products? Because they require too much user input and are slow by comparison with flatbed doing a quick copy of a sheet of A4.

Are they expensive? Not for a lab doing many scans a day. They will pay for themselves quickly if you are doing scans commercially. At least they did before the advent of "Good enough" scans from flatbeds. And lastly, all the cheapskate fine art photographers who are not worried about ultimate quality just won't spend the money on one. I blame the open source software community which has created the belief that everything should be free.

[Ed Richards]

> And lastly, all the cheapskate fine art photographers who are not worried about ultimate quality just won't spend the money on one.

I hope this is tongue in cheek - even just scanning 200 sheets of 4x5 a year on a drum scanner, at $100 a sheet, which is low end for good services, is $20,000.

[robc]

exactly! if scanning 200 sheets brings in $20,000 dollars of revenue then that makes a drum scanner look pretty cheap to someone doing scans commercially. And if $100 is cheap for scanning services then some people think its value for money or they wouldn't pay it. Its only those who won't pay it that think its expensive. I wonder who they are?

[David Spivak-Focus Magazine]

> And lastly, all the cheapskate fine art photographers who are not worried about ultimate quality just won't spend the money on one.

I hope this is tongue in cheek - even just scanning 200 sheets of 4x5 a year on a drum scanner, at $100 a sheet, which is low end for good services, is $20,000.

Drum scans only cost me $15 a scan because my printer does them on such vast quantaties per year for hundreds of other magazines. I would be happy to offer any photographer who needs drum scanning done for whatever purposes the same value with a 72 hour turnaround time. I don't know the price for a scan done above 16 x 20, but I am sure I can find out. Anyone who is interested feel free to PM me here or e-mail me offline at david@focusmag.info.

[Bruce Watson]

To be honest, I don't really understand drum scanning. But I'm curious as to why the technology hasn't come down in price or become the a consumer technology the way flatbeds have. Are the machines that make drum scanning possible so complicated that only professional labs are able to afford/use them? Seems like everything in photography has come down in price, except perhaps MF digital backs, but drum scanning technology feels like it's stuck in time with nothing new over the horizon. From the few used drum scanners I see pop up every now and then, it requires the use of a Mac G4 with SCSI, and the scanner is still quite expensive. Perhaps it's my ignorance on the subject, but it is rather difficult to find new information (from my searches).

The technology did come down in price. And rather remarkably at that. Around 1990 a drum scanner was the size of a large office desk and weighed as much as 5000 kg. You had to reinforce the floor and place it with a crane. Literally. They could cost up to $250K.

By the mid 1990s there were a number of "desk top" and "desk side" scanners available. They often weighed less than 100kg and cost less than $80K. Some as low as $40K IIRC. You can thank the semiconductor industry for much of this gain.

In the middle 1990s however, digital capture finally got "good enough." Commercial photographers working for magazines and advertising dropped film and went digital practically overnight. The reason wasn't quality, it was time. They eliminated the entire process-and-scan cycle. They could see at the photoshoot what they had, get client approval, and move on. And time is money in these industries.

This of course devastated the pre-press market. They lost 90% of their business overnight. None of the makers of drum scanners could give the things away. There was massive over supply in the market -- why would anyone buy a new drum scanner? When the market collapses, where's your incentive for R&D and manufacturing new products?

The drum scanner basically ended around 1996. Everyone who was in the market at that time is gone with the notable exception of Screen. For the picky among you, Howtek isn't in the drum scanner business anymore - they sold the scanner biz to Aztek who keeps the name going. ICG in England went under, but managed to resurrect themselves a couple of years ago after being out of business for an appreciable chunk of a decade.

There are right now three drum scanner makers on the planet that I know of. Aztek in USA, ICG in England, and Screen in Japan. The vast majority of what they do is refurb. and service of existing scanners. The number of new drum scanners sold world wide could be less than 100 a year.

Those companies that wanted to stay in the scanner business dropped drum scanning and concentrated on flatbeds. And no wonder. Flatbeds are much cheaper technologies.

[Henry Ambrose]

Ten years ago nearly every photo printed in ink was scanned from film. Now nearly all are the product of direct digital capture. The pre-press industry barely needs drum scanners - certainly not in the numbers they did just a short while ago. As old machines die, they likely will not be replaced. There won't be any cheap drum scanners other than old ones that no one else wants. There may not be any new ones designed and made at any price for that matter.

Go to a bankruptcy sale of a pre-press house or engraver - you'll see older drum scanners that have to be hauled off for scrap becasue they are essentially unusable and not repairable. But there's no work for them anyway. Which is why you can buy a used but still serviceable drum scanner for a tiny fraction of its original cost.

What is a big fat 4x5 drum scan worth? Thats hard to say but as the installed base of drum scanners shrink I don't think you'll see big drops in scan prices. If anything, prices will go up. But does that really matter when the truth is that very very few people really need the highest quality scan?

Whoever started the idea that you must scan and archive a huge drum scan file of all the "good" pictures you make is nuts. Does the world need a 1 GB scan of your old falling down barn? I doubt that 10% of the "masterpieces" that grace a drum are worth the operator's time much less the cost of owning and operating a $200,000 machine. A realtively cheap flatbed scanner, well run, will accomplish most of what needs to happen for about 99% of the photos made. For your really good ones just buck up and pay what it costs. Two or three a year at $100 each won't kill you.

Edit:
You tell 'em Bruce!
(we must have been writing at the same time)

[paulr]

I doubt that 10% of the "masterpieces" that grace a drum are worth the operator's time much less the cost of owning and operating a $200,000 machine.

My dayjob is in publishing/advertising ...the industry that keeps the prepress shops alive. We order drum scans for ANYTHING that gets printed on a press. Habit, probably. At any rate, your 10% estimate strikes me as wildly optimistic. Virtually everything we have scanned is crap. So are most of the pubs that print the scans.

[Ted Harris]

Terrance, one more point. None of the high end scanners have come down appreciably in price ... at least not so you would notice although mos tare no longer over 100K these days. You are confusing consumer or prosumer flatbeds with their professional relatives. Professional flatbed scanners from Screen, Aztek, Kodak/Creo, etc. all start at over 10K and go up to over 40K. The guts of these scanners is as different fromt the guts of a prosumer scanner as are the guts of an 18 wheeler Mack truck from those of a Buick ... as are the prices and the software that drives them and the complexity of the software, etc. Finally with the high end scanners we are talking about machines that are virtually hand built to extremely tight standards and in very small quantities. The total worldwide output of these scanners in a year is less than the number of prosumer scanners manufactured in a week, maybe a day.

With the right operator the results are that much different as well. Yesterday I was working with a client and we did some comparison scans with a really tough full plate glass plate. The amount of detail that the Screen pulled out of the image matched or surpassed what he had been able to acheive with an enlarger (top line Durst 10x12 enlarger) and the results from the Microtek 1800f, while ok, weren't even in the same league.

BTW, I'm also not sure sure that much has come down in price in LF equipment. I don't see any significant downward trends in either cameras or lenses from the major American, European and Japanese manufacturers. Sure, there are new Chinese entries that are less expensive in some instances but that is another story.

Looks like we were all writing at the same time! I might quibble a bit with the point that the high end drum scanners are a less expensive technology than their drum breatheren. The CCD array's are somewhat less expensive than the PMT's used in the drums but the stepping technology used to drive the moving beds and arrays is very expensive at the level of accuracy at which they operate.

FWIW I think there is also one more company(maybe two) still making drum scanners. Worldwide, Kodak, Screen, Aztek, Fuji and Microtek all market flatbeds that compete with the drums at the high end. It is hard to say of Aztek actually manufacture their own though and Microtek doesn't sell theirs in the North American market because of prior licensing and marketing agreements with Creo.

BTW the under 100 made a year mentioned above is probably very close to the mark .... including the high end flatbeds.

[Terence Patrick]

You are confusing consumer or prosumer flatbeds with their professional relatives. Professional flatbed scanners from Screen, Aztek, Kodak/Creo, etc. all start at over 10K and go up to over 40K.

True, I forgot about the Creo used when I was in advertising. I knew a lot of the photographers who shot for us were still using film, I just don't remember taking note of how we got that film digitized for the art directors or production artists.

How much of a difference in performance are we talking in these professional flatbeds when it comes to scanning 4x5 or even 120 film versus a Microtek or Epson? Would the difference be akin to a Phase One back and a Nikon P&S?

Thanks for all the input everyone, I'm still putting a lot of pieces together about scanning film as I've really started exploring the LF world.