§ Contents Glossary Articles Books Software Typeface Webmaster Links

Computer-to-Plate...we'll do it again

We promised you a warts-and-all report on our last issue, the first Australian magazine entirely printed direct to plate. If we didn't have to share the present issue with Macworld (the luddites!) just because AMW wanted to offer Publish as a "bonus," we would have been repeating the excercise. So, you can assume the experience wasn't too bad. In fact, it was enjoyable

Our report is short on CtP problems in the mechanical sense: there were none. As Andrew Leder at Sinnott Bros prepress put it, the only change is that a laser now points to a printing plate instead of a film sheet. File formats don't alter, preparation is as before--except that for workflow reasons better quality control is required.

For the prepress, CtP ends an "anomaly" introduced with the early computers and film. Andrew Leder said: "20 years ago we were platemakers and now we're back in a similar position." The sign outside Sinnott still reads "Lithographic Platemakers".

New RIP, new problems

We discovered a range of problems at prepress, but they were unrelated to CtP. Obviously, commissioning a new CREO platesetter brought the same adaptation issues as every imagesetter replacement does, with subtle differences between the old and new RIP. Sinnott operators had to develop custom settings for their shop's particular sequence of hardware and software.

At one point of proofing we had a colour dropout error on 13 separate ad pages. It turned out to be a "feature" of the new RIP, since fixed at Sinnott. Nothing new in that. We have been producing magazines for ten years and cannot remember any upgrade of prepress RIPs or software that went without similar teething problems.

So far, so good.

The end product was more than satisfactory. Everyone noticed the crisper printing, and where ad material had appeared in a previous issue or another publication, one could spot the quality difference at a glance.

Why do it?

Not for cost savings, at least not initially. The expected savings on film and materials eventuated, but in the order of $25 per page they didn't offset the capital costs and labour. In the long run it will be a different mattter.

Sinnott's Karl Balekjian nominated "control" as the greatest benefit of the new process, a view echoed by every agent in the chain, we editors included.

Real as this is, quantifying the cost benefits of better control will require careful attention to the process.

The cost of re-work drops as a result of greater speed throughout. Plates can be damaged, film can be creased, kinked or scratched. In traditional printing, this means reshooting the film, recreating plates and printing--and losing a lot of time. In CtP, damaged plates can be recreated from the RIP files. Balekjian said that reprints on the CREO took as little as 15 minutes. The process from PostScript to RIP to plates takes about 30min when all the pieces are ready.

Mixed input, mixed reactions

"One format" is an ideal but not a reality in mixed ad/ed publications. The magazine you hold is a good example. Last time, we "trained" and cajoled our Publish advertisers to provide digital ads, but with Macworld involved we had additional clients -- could we impose on AMW advertisers the same discipline? We thought not, and went back to film for the month.

The issue is responsibility. Should creative directors and page creators be trained at the source to deliver digital files, or should prepress charge fees for acting as the first-aid station? At prestigious advertising agencies the art director is guru, and it is a trusting prepress owner who thinks he can bill the client for "fixing your mistakes". In reality, prepress has always done that, they just didn't call a spade a spade. Now, they will probably bill it as "file preparation for CtP" and perpetuate the myth that CtP is somehow different. Anything to keep the peace.

An interesting comment came from Bruce Sinnott, who said: "We're going to make it easy for some clients. Basically, we won't tell them we're doing it. We'll accept and ask for film as usual, but add 'could you send your file on disk along with it?' We are adapting to CtP; the clients can join in when they are ready."

Embedded bitmaps

In our production, a common fault was PostScript graphics created with a PICT, sometimes even an RGB image, embedded. The images dropped out. Screen-trained graphic artists have little idea how the software produces files for high-resolution print.

Another age-old problem is missing fonts; we didn't escape that.

Operator error is common, but not the only source. A Seybold Report classified typical sources of error: operator, applications, RIP idiosyncracies--but added a category "errors from outer space."In a PostScript production environment, every operator encounters from time to time the totally inscrutable failure, an apparently random Act of God (or John Warnock, depending on your point of view).

So can you educate clients via better spec sheets? "We didn't get one file that actually met our specs. People don't read specs", Leder said.

Next: pre-flight checks and post-RIP proofs

We predict that proofing for colour output will stay, but real advances in the electronic workflow will have to come in the area of pre-flight checks. Adobe's promised PDF 3.0 format may be a crucial step in that direction.

Another promising technique is the post-RIP proof. Compare this with the operation of a scanner or photocopier: they reproduce the image in a format independent of the applications that created it. Gone is the PostScript code, the QuarkXPress file format, the imposition algorithms--the post-RIP proof reads the final streaming bitmap as the mechanical printer would.

Provided the post-RIP software emulates a particular printer precisely (and that's an "if"), it ought to be the ultimate proof. An Australian company, Serendipity Software, has released its Blackmagic software claiming to achieve just that. By de-screening the file, they can also output the proof to a wide range of common proof printers.

Soft proofing is touted as the future, but we should take advantage of expertise where it resides, namely with the prepress operators. Attempts at teaching clients to proof will come out second best. The ultimate solution might be all-electronic, permitting prepress to run a post-RIP proof to file and returning that file to the client for output on their in-house proof printer. In our case, couriers and the material handling of proofs were time-consuming.

How print-ready should artwork be?

CtP causes an anxiety with art creators that is overstated. In truth, nobody expects "print-ready" pages from the art director. Karl informed us that most of our files needed work on trapping and overprinting, but added that he did not really expect us to do it. "In practice it is easier for us to do it," he said.

Imposition is added at prepress, but with CtP they feel like babes-in-the-woods. "What we in prepress call imposition is very different from what the printer means," said Karl. "Now we have to do it, and it has been a pretty tough learning curve." People mistakenly refer to CtP as another form of imagesetter; it is in fact a platemaker, which means that all the precision of plate centring and imposition for a specific web press has been transferred back to the prepress stage. There is no margin for error.

Why do we need proofing?

The difficulty of proofing has repeatedly been held out as a stumbling block for CtP, but we question this. After all, "proofing" is an unsophisticated form of quality control and fairly unique to the print industry. Final inspections are not the main instrument in other industrial sectors. Could you imagine Boeing building a new Jumbo jet, sending it up in the air, watching it spin to the ground and saying "Drats! It fell down. Make another one." After-the-fact testing should never be more than the finishing touch in a quality control process.

There is software available to better implement in-production testing. Software commonly referred to as "pre-flight" testing (apt analogy) conduct diagnostics on a supplied file, reporting on missing fonts and images. The software will analyse the graphics and identify complex PostScript build-ups, redundant paths, hidden objects--all factors that cause excessively long RIP times. Poor file preparation is a significant cost not avoided by a simplistic print-and-proof strategy.

The physical and organisational separation between art agencies, prepress and the print shop breaks the production flow. The tension is exemplified by CtP but by no means caused by it.

Electronic publishing is an unusual industry in this respect also; how often does the customer produce the raw material for their supplier? The customer is expected to understand their manufacturer's equipment and tools, which in turn leads to a circular quarrel over blame when a product fails.

Digital means "big"

Sinnott's close association with Offset Alpine, the printer, deviates from the preferred CtP setup where the platesetter is situated close to the web press. The distance puts pressure on network capacity and file storage: Sinnott's jukebox storage runs to terabytes. Over time, the geographic separation will have to be overcome by installing a fast network.

That is the fascination with a digital workflow: When you see the heavy network pipe running from prepress to the press, you cannot help thinking the next step should be to extend it back to the origin--to editors and page creators. We could leverage the prepress staff's expertise, shooting a file over to ask "how's that?" and having a reply within minutes.

It sure beats sending artwork, waiting with trepidation for the couriered proof right on deadline, and then discovering magenta boobs in your Madonna feature.

Osmund Iversen, Geoffrey Fletcher, Sarah Iversen, with thanks to Sinnott Bros staff

DeskToPress