Low-cost scanners and the World Wide Web have led to a second outbreak of the desktop publishing revolution over the past two years. Images are everywhere, and quality -- predictably -- varies. Professionals still want to get it right, and if you know what you're talking about, the opportunities for your business are immenseMatthew JC. Powell investigatesLast week, ARN walked you through the consumer end of the imaging market with inexpensive scanners. These scanners were all below $500 and designed mainly for OCR text scanning and low-resolution images suitable for Web pages.
From there, the price of a scanner keeps going up and quality rises accordingly. For professional-quality scanning, it's hard to spend less than a couple of thousand dollars and it's easy to find yourself looking at a $50,000 sale. Who needs to spend that much? What do they get for it? This is the subject of this week's investigation.
Scanners are not the only way to get a high-quality image into a computer. Digital cameras have evolved over the past two years into a serious option for professional photo-graphers looking to save development time. However, as our discussion with David Swainston of Maxwell Optical reveals (see page 36), you have to be sure that your customer knows what its going to get from a digital camera before you make the sale.
A little bit of knowledge may be a dangerous thing, but a lot of knowledge about putting an image into digital form and what you can do with it once it's there can make you a more effective imaging reseller.
The Mac's last stand?
In the earliest days of professional scanners, the market was dominated by proprietary systems from Scitex, Optronics, Hell and others. With the advent of powerful graphics computers on the desktop, these proprietary standards were replaced by Mac OS front ends.
In most areas of digital publishing, the Mac has gradually lost market share, especially since the advent of Windows NT. NT's native TCP/IP networking and ability to handle large amounts of data very quickly and reliably have led to its widespread adoption as a server platform in professional print houses.
However, NT's lack of a colour standard has hampered its adoption by professional scanners. The Mac OS has the capability of calibrating monitors and printers so colour output will be identical, and that has led to it remaining the default standard for colour pre-press. The front ends for most professional-level scanners still rely on a Mac, and Apple still claims a greater than 80 per cent share of colour publishing.
Windows NT 5.0 will include a colour-matching technology licensed from Linotype-Hell, which may grab some of the Mac's last great stronghold of market share. In response, Apple's upcoming Rhapsody OS will include native TCP/IP networking.
More than one way to scan a cat
Well actually, a cat would require a 3D scanner -- like they use to make com-puter models for movies. And you'd probably have to sedate the cat, which may not be legal. However, scanning a two-dimensional picture of a cat is a rather more accessible activity, and chances are you have customers who want to do it. How they decide to do it and how much they want to spend on doing it can vastly affect the results they achieve.
Scanning on a flatbed is the most straightforward thing in the world: you simply place the original on the glass plate and you're ready to roll. You may wish to hold the original in place with a frame, but it's not entirely necessary -- gravity is your copilot. With a drum scanner, the drum rotates very fast, so securing the original (usually with clear sticky tape) is a good idea.
An advantage that drum scanners hold over flatbeds for mounting artwork is their suitability for what's called "oil mounting". This is where a small amount of oil is placed between the original and the drum, serving to keep the slide dust-free and reducing the visibility of scratches. The drum creates tension in the film that holds it evenly against the glass, distributing the oil evenly. Flatbed scanners are not suitable for oil mounting because there is no inherent tension in the film, and air bubbles are the result.
Both types of scanners direct light through a transparency being scanned, and both use optics that focus and direct the light. Drum scanners use photo multiplier tubes (PMTs) to convert light into a digital signal, while flatbed scanners use a linear charge-coupled device -- like the ones used in digital cameras. FlexTight scanners, distributed here by Proscan, are a hybrid of the two technologies -- bending the original to create tension, but with a CCD for the actual imaging. The result is an odd-looking beast indeed.
Drum and flatbed scanners also use slightly different tricks for magnifying images. Both types actually maintain a 1:1 aspect ratio. Drum scanners change the rate at which the optics scan across the drum, as well as the frequency with which the PMTs sample the image. Flatbed scanners focus their lenses to capture either a smaller or larger area of the image.
A relatively new player on the horizon is the slide and negative scanner. This also uses a CCD, but the body of the scanner is specifically designed to accommodate small originals. An example is Nikon's forthcoming Super Coolscan 2000, which will be announced this week. As well as slides and negatives, it has an adapter for Kodak's Advanced Photo System film, which allows you to slide a roll of developed film into the scanner in exactly the same way you would insert film into a camera.
How do you sell a $50,000 scanner?
Many of your clients might be making very regular use of bureaus for in-house brochures, catalogues and client mailings. It is possible they have not considered the possibility of purchasing a professional-quality scanner for themselves because of the high initial expense and training costs. The fact is, though, that if your customers have the volume to make it work, a $50,000 scanner could be a bargain buy compared to using bureaus. There are three distinct advantages for your customers in bringing pre-press hardware in-house: money, time, and control.
Is it cheaper to do their own pre-press work in-house than send it to a bureau? Maybe. Service bureaus have the same costs as everyone else -- capital, labour, training, maintenance, consumables -- mitigated only by volume. If your customers produce enough printed work to make their costs lower than what the bureau charges for its service, they can come out ahead.
How much volume does it take? Compare the average scanner operator's roughly $22 per hour wage with the typical service bureau charge of $45 for a 150lpi 8cm x 10cm transparency scan. If your customer is producing a couple of thousand scans in a year (eight a day spread over 260 working days), they're up for nearly $100,000 in bureau costs.
Considering their scanner operator's wage comes to about $45,000, they could have paid for a pretty good scanner.
Time is the next factor to consider. If your customer is, say, a large retailer with a big investment in print advertising, changes to ad material can take days going through bureaus. That can mean missing print deadlines and ending up behind its competitors. With pre-press hardware in-house these changes take hours. That nimbleness can prove to be a distinct survival advantage. Bureaus have hundreds or even thousands of clients. An in-house pre-press department has only one.
A final point
Finally, control. If your customers have the requisite skills in-house (or can hire them in), it's easier for them to get what they want from their own employees than it is to communicate their wishes to a service bureau. And because they can check and fix their own work much more quickly, there is a further time advantage. This allows more room to experiment or tweak for more quality.
There is a downside though. If a bureau's equipment breaks down, they generally have a redundant system in place to keep the work flowing while the damaged machine gets fixed. Your customers are unlikely to want to shell out for an extra scanner "just in case", so workflow stops the minute the scanner does.
Interview -- John Swainston
Maxwell Optical Industries recently reeled in its electronic imaging division to make it a more integral part of the organisation. Managing director John Swainston says that was because the electronic imaging people and the optical imaging people weren't working with each other, or were actually duplicating their workloads. Matthew JC. Powell spoke with him about the right way to sell electronic imaging productsSwainston: The first thing is to understand what the electronic imaging business is. To us, it is about taking pictures and picture assets which might be data files, graphics and so on, and making them useful to the client who is in need of propagating their story: distributing visual information, instructions, whatever the purpose of that picture is. To be successful in the electronic imaging business, you have to have some understanding of the utility of pictures and how to make a picture a useful picture.
So, for example, our experience so far is that computer resellers are very good at applying technology that works well with desktop publishing, because that's where the specialists have had the experience. So the Masters of Media, in the case of the Mac, or some independent PC resellers, have the expertise.
Still, they are much more comfortable with a scanner than they are with a digital camera. They are very much less successful selling a digital camera, because it's not just a device, there are a lot of other features in a camera that have to be explained and they have to be related to people's existing experience of cameras. If you don't have that connection, there's going to be that gap.
So if you're going to be in the digital camera business, which is the front end of the business, you've got to understand what cameras are used for, and what people's expectations are.
ARN: Are people's expectations of digital cameras being set too high at the moment?
In the US, there is already clear evidence of a very high return rate for digital cameras. In the low consumer end, through department stores for example, that's up as high as 40 per cent. That's because the expectation of the digital camera has been poorly explained at the retail point of sale. A digital camera of VGA resolution is not a substitute for a photographic (film-based) camera.
Kodak's DC-20, for instance, is down in the sub-VGA space and it's a great product for use in electronic communication.
So if you're into e-mail, doing graphics for your Web page or internal communications that are intranet-based, then that's a fine product. If you're going to print it, don't look to that end of the market. You have really got to look at an XGA, 1024 x 768 resolution camera to get a usable print that will give the impression of a photographic equivalence when printed on the average networkable printer.
Even then you're still nowhere near genuine photographic quality.
Yes. To get to that level, we're talking at least a 1.3 to 1.4 megapixel camera. That will give you something like a 5 x 3.5in print at about 205dpi. That's actually not an unsatisfactory image. The eye has a wonderful capacity for accommodating -- that's why motion pictures work. We let ourselves be fooled.
It's important that a reseller understands the difference be-tween the resolution of a 1.4 megapixel camera that produces a 4Mb file, and what the data content of a 35mm image is. There's roughly 35 to 40MB of data equivalence in a 100 ISO conventional film. If you go to a slower film, you're actually looking at 45 to 50Mb of data before you get grain interference. So we're miles away. We're one-tenth of the resolution, when you consider that, with a camera that costs $2500.
So it's very important when selling the primary capture device that the first thing you do is question the customer on what is the need that is to be filled here. If it's for print, I'm not sure we have the right value proposition for you. We're going to be looking at a camera typically around the $8000 to $10,000 mark, and most of those people who are successful at selling are specialist camera stores that understand the whole gamut of high-quality colour origination.
If it's for internal documentation -- bulletins that are going to be printed on a laser printer, typically you're printing in a two or three column layout on A4. You can work with a 1.3 megapixel camera, and that will produce satisfactory output at 300dpi.
And that's a $2500 proposition?
That today is $2500. Three years ago, that was a $10,000 proposition. Now, for $10,000 you're not really getting much better resolution. What you're getting is greater bit depth, better colour management process in terms of integration with the output systems, greater memory capacity, greater speed, and typically, an SLR lens.
But to deliver the same quality as the $2500 camera, the computer reseller should seriously consider partnering with companies that can sell a 35mm camera to go with the scanners they're selling. Resolution is not the only benchmark of quality. Resellers have pushed the price points down to below $200, but again the customer's expectation is they're going to get full colour pre-press 135lpi screen quality, and it's not going to happen. Optics comes into this as well as transport reliability and steeper motor accuracy.
Unfortunately, that seems to get overlooked in the selling process.
Realistically, the kind of print quality people expect from a scanner is going to cost roughly $1500, but on top of that you could pay $300 for a quite good 35mm camera and still be able to deliver comparable quality to a $2500 digital camera for less than $2000. Plus you can then use the scanner to image any other flat art.
I think that possibility is being overlooked. When people are thrusting digital cameras at people, they should also consider the other applications a customer might have and try to understand that.
You're obviously not a big supporter of the push towards lower price points.
Resellers can certainly sell on price, but you don't pay the staff's wages on margins of $10 per transaction. We believe that as this business matures, people have to get involved in the category of the $2000 to $3000 mark where several hundred dollars profit per transaction is going through. Actually, the selling time for this category is similar to the $199 flatbed scanners. So the yield for the sales people is much better in terms of value for their time. Resellers should be very keen to work with that.
Input devices -- scanners and digital cameras -- are only half the story with a photographic image. Customers want to put those images on paper, so where are the opportunities for resellers in that regard?
On the other side of the business, the output side, inkjet has become the predominant consumer product. It's also now providing a fairly serious alternative to colour laser printers. First because the print technology is becoming quicker, and second because when you introduce colour output into an organisation it tends to be fairly localised. Quite often, it's not a network product and in that case inkjet looks pretty good against the print speed of a typical network print server.
Because of the photographic quality available now from inkjet, which far exceeds the quality of colour laser, people are looking to that for short run colour brochures.
They might only produce 10 layouts for a particular client, so photographic quality colour inkjet is now very much a part of the equation.
Productivity also has to be a part of the selling process. Resellers have to understand what the customer's time horizons are, because you can get, just like the resolution issue with digital cameras, a very high level of customer dissatisfaction. That can influence your whole commercial relationship with a long-standing client.
In our experience, organisations with an outside sales force that goes out to clients and asks questions have the most successful model in the electronic imaging business. You can sell boxes, but by and large you will also have a much higher tech support post-sale if you haven't adequately understood the client's installation. If you're going to get into this business, expect significant initial startup support activity with each client if you haven't done the pre-selling enquiry process properly. We have some resellers, both in photo and computer channels, who are absolute experts in this process and the post-installation support is almost negligible.
Because they've asked the right questions, they know what devices need to be supplied in addition to the initial product. Support can kill you in this business, and the reseller has to take some of that responsibility.
Staying on the subject of post-sales, what after-market opportunities are there in imaging?
The biggest one is consumables, especially in the inkjet space. Resellers should understand that most brands of inkjet printer use specialised binding technology to get the ink to dry right, to merge right, and to spread out right. If you are trying to make extra margin by selling a third-party product to that customer, they are not going to get the same quality. They are going to be dissatisfied and your net margin will decline because you won't get a lot of returns.
The other thing is the application programs that will make the images useful. The first is an asset management program. The average roll of film is 36 images -- that's 36 files. If they're scanning three or four rolls, they've suddenly got 100 images. If you want a particular one of those, there is nothing more annoying than having n30502tj.jpg as the sole identification of what that image is. A product we handle for this is Canto Cumulus. It's cross-platform and it provides automatic filing, automatic descriptors, cross-referencing, and sorting. All of those things will increase the customer's satisfaction.
Linotype-Hell Opal Ultra
The key to the Opal Ultra is its software bundledual lens system enables accurate magnification from 35mm to A42800dpi x 1400dpi optical resolution431.8mm x 144.8mm optimum scanningRRP $17,132For further information contact Heidelberg. Contact details belowKodak DC-200An affordable, good quality camera -- but be careful your customers know what they're getting.
1152 x 864 pixels per inch resolution
threaded lens, allows attachment of industry-standard accessories
liquid crystal display and viewfinder
4MB memory card included, holds up to 60 picturesoutput as FlashPix, JPEG, PAL or NTSC VideoRRP $1250For further information contact Kodak Digital Science. Contact details belowScitex Eversmart scannerScitex recently revamped its top range of Smart scanners. The Eversmart range offers similar features, with a second stepping motor to ensure accurate scans in the Y axis.
2540dpi maximum resolution
"X and Y" scanning heads ensure consistent density in both horizontal and vertical axes15 scans per hour20 to 2100 per cent scaling (at 300dpi) in 0.1 per cent increments305mm x 432mm maximum original size (transparency or reflective) ESP $43,000 For further information, contact Media Tech. Contact details belowImacon Flextight PrecisionAn odd hybrid beast, Imacon's FlexTight bends transparencies but doesn't use a drum and has a CCD as its imaging mechanism.
4800 x 4800dpi resolution
maximum original size: 300mm x 121mm (transparency); A4 (reflective)
doesn't require tape or oil
various sizes of film holders included
For further information contact Proscan.
Contact details below
Super Coolscan 2000
High-quality images: 1*s per colour channel A/D and file sizes up to 57MB20sec scan time at 2700dpi ImageFix function -- automatically removes surface defects such as dust, scratches, fingerprints from scansESP $2700For further information contact Maxwell Optical. Contact details on below.
Canon Professional Services
Tel 1800 804 240
Fax(02) 9888 3314
Tel (02) 9318 5222
Fax 02 9313 7516
Kodak Digital Science
Tel 1800 674 831
Fax 1800 659 757
Tel (02) 9390 0200
Fax (02) 9390 0201
Tel (02) 9879 4744
Fax (02) 9879 4845
Tel (02) 9904 1244
Fax (02) 9904 1099