DigiBarn Printers: Personal Recollections of the Xerox 9700 Electronic Printing System

Recollections of the Xerox 9700 by Peter Preksto
(a co-founder of Intran, founder of TyRego)

The image above shows the channel-attached version of the 9700, as the tape tower isn't present. This version was simplex, 8.5x11, 18,000 lpm or 2 pages per second. Under the LS100 terminal, Xerox had placed a modified DEC PDP-11/34. An extra cage contained a few proprietary cards to facilitate the page ripping. There was a Control Data 14" hard drive (the removable platter type) on sliders. Sometimes the disk fell out of its carrier when you spun it on and lifted it
out.

The machine was based on the 9200 copier with its superb paper handling capability (this thing rarely jammed if you had a good SE for your account). The selenium belt and laser system was in the module that sticks out front--that's added on from the 9200, of course. Duplexing was added later, once the 9400 was released.

All of us who owned these amazing machines marveled at them for many years afterwards--they were so far ahead of anything else, esp from their ability to blast out bits. The next coolest thing of the type, years later, was Larry Lukis's fast experimental printers (meaning fast RIPs) at Moore Research.

...in the meantime, the research was continuing with xerography. By now, it had been decided that this was the definitive technology for computer output printing. Out of this research came the first laser printer, developed at the Xerox Palo Alto Research Center (PARC). The research took a couple of years, 1969-1971. Gary Starkweather, one of the Xerox engineers working on adapting their copier technology, added a laser beam to it to come up with the laser printer. He named this machine “SLOT”, an acronym for Scanned Laser Output Terminal. The digital control system and character generator for the printer were developed by Butler Lampson and Ronald Rider in 1972. The combined efforts resulted in a printer named EARS (Ethernet, Alto, Research character generator, Scanned laser output terminal). The EARS printer was used with the Alto computer system network and subsequently became the Xerox 9700 laser printing system.

In 1978, the Xerox 9700 was introduced to the U.S. and the world – the first of its kind available commercially. It printed 120 pages per minute (ppm) and is still, even today, the fastest commercial laser printer. But, in its day, the 9700 was huge, both in size and price. On the upside, though, it generated $1 billion a year for Xerox’s xerographic printing business.

Xerox 9700 and the Role of Gary Starkweather

The story of the Xerox 9700 brings us to the fabled Xerox Palo Alto Research Center (PARC), which was established in 1970. When the 9700 was introduced in 1977 it was hailed by Xerox as the first major product to be "enabled" by PARC research. The key person behind this program at the Center was Gary K. Starkweather, often described as the father of laser electropho-tographic printing.

The following exceprts from a 1997 interview of Starkweather by Frederick Su tells a bit of the 9700 story as he lived it.*

Why don't you talk a bit about the imaging development at PARC?

I came to Palo Alto in 1971 to build a laser printer. It took a lot of memory to store the images and provide font storage. We built a prototype in approximately a year and managed to prove out the fundamentals of the scanner and xerographic exposure technology.

How closely is the laser printer related to the office copier first developed by Chester Carlson?

We were able to use the copier technology by replacing the lens that imaged the copy sheet to the photoconductor with a laser beam scanner that exposed the photoconductor. A laser scanner is just a pointwise method of exposing the photo-conductor. The same toners and developer packages used for copying products could be used with the laser scanner systems we developed.

Other folks, like IBM, had to create their own electrophotographic technology specifically for the printer they built. This required a much heavier investment.

Didn't the patent on xerography expire after 17 years?

There were a lot of upgrades that came along that kept the patent pretty much in a fresh state. There were new additons and capabilities and features that came along in the xerographic system...you certainly could build a 1959 copier, but who'd want to?

Anything else you want to say about the development of the laser printer?

It was more fun than anyone had a right to expect. Xerox was very generous in its support of research. There is one person that needs to be noted in all of this history and that is Jack Goldman, who was head of research at Xerox. The Palo Alto Research Center was his idea. He deserves enormous credit for being so visionary and putting it where he put it and for setting the goals that he did. I think there is over 50 billion dollars' worth of business that's been generated from the technology that came out of PARC.

* Reprinted with permission from OE Reports, a publication of SPIE, the International Society for Optical Engineering, Bellingham, WA USA, November 1997 issue. At the time of the interview Starkweather was the Imaging Architect for the Windows NT platform at Microsoft.

I was working for Burroughs at the time of the release of the Xerox 9700 and was at the Hanover Messe (forerunner of CeBIT) when the 9700 was shown for the first time (1977 OR 1978?) - but not by Xerox, by Burroughs!!

Burroughs had OEM'd the printer from Xerox and had one installed on their stand where it was one of the most popular displays due to the fact that Xerox were sending all their customers to us to see it since that did not have one!

Unfortunately it was down for a not inconsiderable time and I helped the lone field engineer in keeping it running while at the same time devouring the documentation and becoming adept at printing my own pages - I actually printed my own business cards at the show!

The engineer was trained only on the print engine and one morning we arrived to find that the associated PDP-11 based controller would not boot, it complained about bad memory or some such. We had no documentation on the controller and the engineer had not been trained on it, neither had I but that did not stop me from trying to fix it!

We opened up the box and I immediately noticed a blinking LED, I asked the engineer about it and he made some vague statement about it indicating that the system was operational. Together we examined the hardware and found a metal washer which had lodged itself between the CPU card and its connector. We fished out the washer and tried booting - same symptoms.

I postulated that perhaps the washer had blown something on the CPU board, so I wanted to try a replacement. I called the local DEC office and explained the situation. They kindly offered to provide a replacement to test my hypothesis, so off I went to collect it. The replacement CPU did not change the symptoms and I had to return it, with thanks, to DEC.

Our second thought was perhaps a power supply fuse had blown. We checked the backplane and fond that the power distribution was clearly marked. We checked each voltage and, yes, one was down. But which supply corresponded to the missing voltage? We did not know. So, we carefully dismantled each of the four or five supplies and checked their internal fuses. All were good! Reassembling the system we verified that we had not broken anything, nor unfortunately had we fixed it.

On the checking of voltages and dismantling of the power supplies, I recall that I was on the floor flat on my back with the system backplane just a few inches above my face probing with a lead while the engineer sat next to me reading off the voltages. When pulling out the (heavy) supplies they practically dropped onto my face!

The printer had been down now for the best part of two days and the salesmen were getting antsy. I suggested that we try and find an engineer at the DEC stand, just a few meters from us, and see if he could provide some insight. Being the only German speaker with at least some knowledge of the problem it came to me to implement the plan and it was with some trepidation I approached the DEC stand. I need not have worried, they were all very friendly and someone quickly volunteered to come and take a look. He came, he saw, and he conquered! On seeing the blinking LED he remarked : look your memory parity is showing a problem, just flick this switch here to disable it and all should be well. And so it was for the rest of the show.

I will always have the most fond memories of that Hanover Fair, some 30 years ago. I recall calling my boss back home to ask for vacation time so I could stay beyond the time he had allowed me. It was many years later when I finally managed to get to what was by then CeBIT and whether it was me or the show, I did not get the same feeling of excitement I recalled from my first visit. I have since visited as an exhibitor and just a touch of the old camaraderie still remains, but it has become much more a marketing show than the engineering display of old.

That first experience with a laser printer left a deep impression on me, and it was with great anticipation that I saw the first announcements of the first Apple laserwriter, in, what 1985, and I purchased one soon thereafter and I could finally develop the skills I had learnt almost 10 years before.