Demystifying Remote Proofing

Originally appeared in the IPA Prepress Bulletin magazine

November 2000

By Chuck Gehman

While remote proofing is of interest to many people, there is also a lot of confusion surrounding the process, including misconceptions about what is required to do this, and how well it works. We’ve already mentioned low cost and high quality. That alone might be enough to stop a lot of people who still think Remote Proofing requires a T-1 line, expensive, specialized hardware and software and perhaps a Kodak Approval. While that is certainly something that people are doing, there is much more to it than the very high end, and I’ll attempt to address some of the fundamental issues as well as provide some “nuts and bolts” information that can apply to both high-end users and entry level users.

We won’t spend a lot of time talking about ICC color profiles and color management, but instead will focus primarily on workflow issues and how to deliver the files to the remote site. We’re going to assume that the reader has a good working knowledge of digital proofing and the issues that surround it. The color science behind getting matching proofs on two different devices may actually be pretty close to rocket science, but fortunately, being successful with remote proofing isn’t.

Why Remote Proofing?

Remote proofing is a topic of hot debate among knowledgeable technical people in the graphic arts. Why? Perhaps because everyone has their own definition of what is considered remote proofing. We aren’t going to attempt to resolve that debate here, but to point out several applications where this technique can add great value.

The very first thing that needs to be understood about remote proofing: Unless you are already in a relationship (whether you are the service provider or the client) that has adopted digital proofing as a standard way of working, you are never going to be able to implement a remote proofing workflow. So, perhaps obviously, digital proofs first, remote proof second.

Where are the big wins for remote proofing? They are certainly in the creative process, such as between an Ad Agency and their service providers. Between a prepress house and a printer (perhaps for imposition proofs). Between a printer and a catalog publisher. Certainly, between an ad agency and their corporate client. Even between in-house creatives in a large company, or in a workflow with packaging designers for a consumer products company. I’m sure we can all think of many more useful applications for this technology. The key is that both production and creative stand to benefit from this workflow.

Workflow

Perhaps obviously, the device and the technical implementation you choose for remote proofing must match your existing workflow, and/or that of your customer. For instance, if you are using a Kodak Approval at your site, and your customer is using an Epson Inkjet, you’re going to need to generate a file that is suitable for the Epson without having to add multiple steps to the process. There are workflows and RIPs available from a number of vendors that can make that happen for you. Of course, they cost money. Some that come to mind that might be particularly useful are the Scitex Brisque, and the products from Shira Corporation and Rorke Data. We won’t discuss these products in great detail and it isn’t necessary to invest large amounts of money in these technologies to do remote proofing. If you already have this equipment and/or software, however, you’re halfway there and you may not know it.

If many additional manual steps are necessary to generate the file to be sent to the remote proofer, it has been our experience that the remote proofing project will fail. And we say “and/or”, because if you are supplying the communications lines and the proofer, you are probably going to be able to fully control the steps you will have to take to deliver the proof. This makes your life a little easier and puts you in total control of the workflow. If you supply the device and the connection, you probably won’t want your customer to use the output device for their own purposes. This will also make the whole process more expensive for you.

If the customer is using the output device for their own purposes, or let’s say, they own it and are letting you output to it from your remote site, they may want to maintain full control over the device. In that case, you may be delivering your proofs into a queue (for example, in the case of a trade shop delivering files to an ad agency’s Fiery-type device or other color digital copier front-end.) When you’ve delivered the file, you send your customer an email (we’ve implemented setups where this is done automatically) that the file is in the queue. When your customer is ready to output, they go into their screen and move your job from “hold” to “active”.


If you own the proofing device, and the communications lines, you’re going to have to decide if you want the customer to be able to print their own work on the output device. This means that you may have to keep track of how many prints the customer makes that aren’t related to your work with the customer. You’ll want to recover the costs of the supplies for any work you aren’t generating. You’ll also have a higher likelihood of media jams and service calls for the proofer, because chances are the customer will have multiple users taking advantage of the cool color printer. Finally, if the proofer becomes very popular at the customer site, you may have to wait in line to get your proofs delivered over the remote connection. This is a big issue to consider.

Communications Lines

First of all, before you say, “I’m going to use the Internet for remote proofing,” or, “Let’s use ISDN for remote proofing”, you need to consider the frequency and volume of the file transfers to the remote proofing device. The next issue will be economics: How much bandwidth can you afford to make this connection happen. This will help you determine what type of technology is right for you. After all, remote proofing at it’s most basic level is a special purpose file transfer. Fortunately, the line you put in for remote proofing can typically also be used to support other applications (i.e., a robust, high speed Internet connection will serve many purposes besides the remote proofing application.)

A rule of thumb when selecting what type and size of communication line to use is what we call “the one hour rule”. Simply put, this rule says if any proof will take more than an hour to deliver, you need to examine other options for communication. For example, with a single direct ISDN line you can deliver about 50 megabytes in an hour. A dedicated, private T-1 line (from your site to a single remote site), can deliver upwards of 600 megabytes per hour. Internet file transfer speeds can vary depending on the ISP and type of connection you have purchased, but with a Tier-1 Internet T-1, you can expect reliable delivery of over 400mb per hour.

A direct connect ISDN setup is a viable option particularly when there are strong security needs (i.e., the customer won’t let you deliver their jobs over the Internet—a rarity these days, with advanced encryption techniques and more savvy IT departments at corporations.) Combining multiple ISDN lines can achieve performance of as much as 200mb per hour.

Internet file transfer speeds can vary depending on the ISP and type of connection you have purchased, but with a Tier-1 Internet T-1, you can expect reliable delivery of over 400mb per hour.

A direct connect ISDN setup is a viable option particularly when there are strong security needs (i.e., the customer won't let you deliver their jobs over the Internet- a rarity these days, with advanced encryption techniques and more savvy IT departments at corporations). Combining multiple ISDN BRI lines can achieve performance of as much as 200mb per hour. ISDN PRI lines can provide multiples of that speed.

Dedicated T-1 lines are practical only when the customer is a relatively short distance away (private T-1 lines are priced by the distance), or the volume is so high that a line dedicated for this application just makes economic sense.

Frame Relay networks can be practical for multiple sites that are often sending to one another. A good example of this is a distribute and print type of application, but that really is a whole other ballgame. Perhaps a better example would be a greeting card company with a central design site and several manufacturing facilities. Proofs might need to be generated often at each of the manufacturing sites, or conversely at the design site from work done at each of the printing plants.

Perhaps the best communication technology of all for remote proofing applications is the Internet. Since we all need to have an Internet connection anyway (for email, digital job deliveries, research and web surfing and the emerging e-commerce applications), why not use the same pipe for this application.

For an Internet connection to work for remote proofing, it must be an “up-all-the-time” connection. Modems will not cut it, because of their low speed and low reliability. ISDN Internet services typically can’t be used, because they are “dial-up” services just like analog modems and you won’t be able to predict in all cases when to dial-up the line so the proof can be delivered.

The lowest cost of the always-on Internet services is called DSL (Digital Subscriber Line.) DSL, at this writing, is available in about 30 cities nationwide but availability is growing quickly. You’re going to want a type of connection called Symmetrical DSL (SDSL) not Asymmetrical (ADSL – the type of DSL you see advertised on a billboard for $39.00/month). The difference is that with SDSL, you have the same speed sending and receiving, which is essential for graphic arts applications. ADSL is designed for web surfing, where downloading is much, much faster than uploading because the service providers expect you to click once, then watch a web site or streaming video link come onto your screen. ADSL is perfect for home, but not useful for graphic arts file transfer or remote proofing.

SDSL is available in speeds ranging from 128K (the same speed as an ISDN line—about 50mb can be delivered in an hour) up to 1.544mbps (the same speed as a T-1 or as much as 400mb per hour.) Keep in mind that DSL availability is not guaranteed, even in the areas of the country where the service is offered. And distance and other telephone company technical issues can often prevent the service from being installed at your site at all, or may limit you to a slower speed. Typically, if available, you will get speed choices of 128K and 384K.

Cable Modems have sort of come and gone from the hype of the telecom world, with most people realizing that it’s going to be a long time before any sort of business-class, generally available service that uses broadband digital cable will be obtainable to our business locations. When and if this happens, these services do offer an “up all the time” type of service, and may be able to provide the high-performance, reliable type of service that we in the graphic arts require.

Beyond DSL, you’re looking at Internet T-1 or even multiple Internet T-1 (a popular new option called “N by T or NxT”, which combines T-1 lines into a bigger connection) lines for high performance Internet service. To do remote proofing over the Internet, you will need to use a Tier-1 Internet provider. These are the companies (typically household names like AT&T, MCI/Worldcom or Sprint) who own the large backbone networks. These companies are able to provide uncongested, reliable bandwidth for mission critical applications like remote proofing. As we discussed before, an Internet T-1 line can deliver as much as 400mb per hour to a remote site; but not if you are using a local ISP who has oversold or “oversubscribed” his bandwidth by an order of 10 to 1 (meaning he’s sold 10 times the bandwidth he has contracted for to the Internet backbone… not uncommon today.) We call this oversubscription “the healthclub model”, and isn’t going to provide great performance for your remote proofing application.

As the Internet becomes more reliable and demonstrates even higher performance than available today (something which is happening very quickly with the billions of dollars being pumped into it), we expect virtually all remote proofing applications will take place over the Internet. With new technology such as RealTimeProof (discussed later) and the Internet Printing Protocol (IPP), a new standards-based printing architecture from Hewlett-packard and Microsoft coming online, this will become easier and easier.

Delivering the Proofs

We’ve covered the most popular technologies for digital delivery. We now have a pipe to send the files from your site to the remote site. Now let’s talk about how to get the files on the output device at the remote end.

Some of the most popular and affordable proofing devices are the Imation Rainbow, the Polaroid DryJet and PolaProof (the newer, smaller format unit is especially popular for remote proofing) and the Epson 5000. Each of these units uses a different technology for imaging the proof. They all come with various software packages to handle the digital files. It is beyond the scope of this discussion to describe all the different proofing technologies and printing methods these devices use. The good news is, over the last couple of years, there have emerged three easy, standards-based ways to deliver files to these proofers.

First, is the hotfolder method. All the above units can support delivery by hotfolder, and just about every professional quality machine that we’ve investigated supports this method. This means that there is a folder on the machine that runs the RIP (whether it be a Macintosh or a Windows NT machine, or a proprietary RIP like the Fiery in the Epson.) The RIP software is monitoring that folder and when a file is dropped into it, the RIP picks it up and begins to image the file. Typically, the file that will be delivered to this hotfolder is a Postscript file. That means that at the end of the connection generating the file, the user must print to a file, and use a PPD (Postscript Printer Description) file that matches the target output device. This is extremely important, because if you do not set up the PPD to match the destination proofer, it is very possible that the print job will cause an error on the device, and you will have to re-create and re-send the file.

The second method uses the “desktop printer” on the Macintosh. Later versions of the Macintosh operating system (from system 8 on up) support this method. Basically, it is possible to drop a file onto the desktop printer, which subsequently sends the file to that device. We use a simple Macintosh-based FTP Server called NetPresenz (from an Australian software developer called Stairways Software) to deliver the files to the desktop printer. We have also used this software for “hot folder”-based applications as described above. The best thing about using this method is that you can deliver a proof to almost any output device. As long as the printer or proofer is available in the Macintosh “chooser”, you can use this method for remote proofing. This opens up a wide range of printers that can be used for remote proofing applications, from the least expensive (i.e., $199.00) InkJet on up.

There are some caveats to both of these methods. First and foremost, you need to make sure that when the file arrives and is dropped onto the hotfolder, that it is all there (i.e., a complete file transfer.) This sounds strange, but it is possible when using some software to deliver the file to the destination machine, that the hotfolder will pick up a “partial” copy of the file. This will result in the file not imaging at all, or generating an error on the proofer. A good workaround to this is to use an applescript on a Mac-based RIP proofer. Deliver the file to one folder, make sure it has been completely received, then have the Applescript drop it into the hotfolder. Applescripts that do this sort of thing are easy to create and you should be able to find one on the Internet that someone else has already created for this purpose.

Second, there isn’t a lot of error checking with either of these methods. Once the file is delivered and imaging starts, you’re pretty much holding your breath until the proof comes out. That’s the bad news. The good news is, these methods are actually very, very reliable and with a little testing you’ll generate useable proofs almost every time.

Finally, we mentioned a third method of delivering the files. This is referred to as LPR, which is a standard Unix method that has been around for many years. It is supported by the Fiery RIPs, the Windows NT-based Harlequin RIP used with the and by standard Windows NT and Windows 2000 servers with the Microsoft TCP/IP printing services (an add-on option that is free.) Any RIP that runs on a Unix machine will be able to support printing with this protocol.

Consumables Issues

Someone has to change the consumables (i.e., the proofing media, toner, ink, what have you) in the proofer. And someone has to pay for them. And someone has to maintain the device and calibrate it. This is one of the most important and neglected aspects of making remote proofing a success. Without considering this point and putting procedures in place to handle these issues, the project will fail.

If you are paying for the device at the remote end, you are probably going to have to be responsible for the consumables. You may be able to convince your customer to assign a staffer to the care and feeding of the proofer. This will be difficult, because calibration alone can take 15 minutes to half an hour on most devices. You will also have to replenish the consumables, and just getting someone at the client to tell you that they’ve run out or are getting low can be a problem. You certainly don’t want to run out of supplies when you are on a deadline. You especially don’t want this to happen on a Friday afternoon, when it’ll be next to impossible to get the supplies to the site in time for a weekend of production.



You may want to consider contracting with a graphic arts dealer to periodically visit the site. Once you have a regular workflow, and know what the frequency and volume of the output will be, you’ll be able to set up a calendar with your dealer for the visits. Keep in mind that it may be necessary to calibrate the proofer more often than you will require additional consumables. In major metropolitan areas, this shouldn’t be an issue-- the dealer personnel are more than likely in the area of your customer on a regular basis and can stop by to perform the calibration. If the customer site is more remote, you may need to go back to plan A and have someone on staff learn how to run the calibration routine.

There are going to be times when you have media jams or other problems with the device. The same dealer we talked about above can handle maintenance issues; preventive maintenance performed when the consumables are replenished can preempt certain types of problems from happening. But there will be times when the equipment breaks and you’ll need to plan for these eventualities. In a high volume, mission critical application, this might require having a second printer onsite as a standby. Or having 24 by 7 maintenance on the device, so you can dispatch a technician no matter when the failure occurs. It also might going back to using Fedex for shipping hardcopy proofs, or it might mean using one of the new product mentioned below to suffice with a softproof instead of hardcopy.

In the case of the less expensive InkJet printers, all of these issues become less important. If the printer fails, you can probably go out and buy a new one while you wait for the broken one to be repaired by the manufacturer. And it’s affordable to keep a spare onsite. Of course, what you gain in simplicity and low-cost, you sacrifice in print quality.

New products

A few new products have recently come out that should be of interest to anyone considering implementing remote proofing.

First, there is a new product called RealTimeProof (DAX currently offers this as DAXProof) from a new company called RealTimeImage, Inc. in San Mateo, California. This Internet-based service supports softproofing (i.e., on-screen proofing) including a tool set that lets you measure color (RGB and CMYK), dimensioning and to place notes on the image for discussion with other participants. It runs in a web browser with a plug-in, and is a service that you pay for as you use, rather than having to buy software or equipment. The service also delivers proofs to a remote site hardcopy proofer.

RealTimeProof has it’s own compression and encryption built-in (for speed and security, respectively), as well as support for ICC color profiles and it is very easy to use. It works with all the popular file formats (you don’t have to create postscript first… you simply upload the image or page file to the server) and virtually any output device.

Another new product of great interest is Adobe’s PressReady. The list price is only $249.00 and it is basically an Adobe RIP that runs on the Macintosh or on Windows NT. This product makes a variety of low cost InkJet printers from Epson, Cannon and Hewlett-Packard into viable postscript output devices. To quote Adobe’s literature on the product, “PressReady prints are intended to help communicate your intentions to your client and your print provider, but they are not intended to replace final/contract proofs.” Fortunately, this is one of the “big bang” applications for remote proofing: Cutting down on the iterations in the creative process.

PressReady supports ICC color profiles supporting industry standards like SWOP, Euroscale and Japan Standard, enabling you to generate output that simulates that you would obtain on a particular printing press.

Microsoft Windows 2000 also promises to provide a really great new platform for remote proofing applications. Microsoft has integrated printing with their Internet architecture in this new operating system. Printers will now have URLs, just like web sites (i.e., myprinter.mycompany.com) and you will be able to choose this URL as an output device for programs running on Windows NT. An example from Microsoft’s Window 2000 fact sheet says “a mail-order company can send its new catalog directly to the publisher's printer, provided the mail order company has permission from the publisher and the URL of the publisher's printer.” It shouldn’t be difficult to support this from Macintosh-based client computers, either, since you can already use Windows NT spoolers from the Mac, and Windows 2000 has much better support for Macintosh clients than previous versions of Windows.

In Summary

Remote Proofing is indeed a viable tool for speeding up the production cycle and cutting down on time consuming delivery of proofs via FedEx or couriers. It doesn’t take a huge amount of resources to get started using this productive workflow, and it can be a tremendous value-add in the increasingly competitive marketplace in which we find ourselves today.

On a personal note, DAX was formed on the proverbial napkin on the basis of a remote proofing project back in 1995. Back then, it was incredibly difficult to implement this type of workflow successfully. Two of the founders of DAX (one at the time working for Scitex, one a well known industry consultant) had a client who wanted to do remote proofing from New York to their customer in Boston. In those days, you really needed to be a telecom guru to do such a project. Needless to say, we’ve come a long way in 5 years!