PDP-9 Systems Group DEC

This page is dedicated to the Engineering Techs, Engineers & Production staff of the PDP-7 & PDP-9 Systems Group

Though not directly related to Video Interchange, Digital Equipment Corp is where I got my start. There's a lot of info on the web concerning the PDP-7 & PDP-9 in general, but none of it written by anyone who was actually there at the time. I thought about a separate site dedicated to the PDP-9, but instead, decided to add it as a separate page to this site. Over time, I'll add far more info, as well as more about the skilled group of people who designed, assembled, de-bugged and maintained the systems. It was a unique group who not only worked hard, but also "played hard". I'll slowly add some of those stories as time allows..... A lot is written as recalled from memory which perhaps is getting a bit "fuzzy" with the passing years. For the present, it's not all that well organized, since material is added as more is recollected. Thanks to Ed Reed for his inputs.

While still attending college, a friend with whom I commuted with in and out of Boston each day, told me about this 'little" company in Maynard, Massachusetts he heard of that was looking for summer help. Being broke and having trouble scraping up tuition for next year, in May of 1966 I decided to apply.

After a surprisingly lengthy interview with personnel, I was hired ! Initially, I could never understand why DEC spent so much time interviewing for a summer "dweeb" position.... That only became clear much later... Personnel said nothing about where I would work or exactly what I would be doing, and since I admittedly knew almost absolutely nothing about digital computers, I figured I would be placed on an assembly line or assigned to assist in providing some other mundane need. As far as computers were concerned, I was qualified at the time to operate perhaps a "broom" but perhaps little else.... Personnel gave no clue of what DEC had in store ..... Instead of being assigned my very own mop & bucket, Herb Millman (PDP-7 Systems Checkout supervisor) comes down to clam his new "prize" after a general orientation by the personnel dept, and I find myself assigned to the PDP-7 Systems Group on the 5th floor of Building #5.

Herb relates that I had been selected for training as an engineering technician to learn everything I could about computer systems that summer. I was immediately placed with Ed Reed who was to become my mentor. Whether Ed volunteered for this thankless task or not, remains a mystery to this day, but whatever the situation, Ed had his work cut out for him, as I hadn't a clue what I was doing. Thus Ed was responsible not only for getting his own system up and running, but now had the added burden of teaching everything from PDP-7 Systems architecture, machine level programming to operation of Tektronix's most advanced oscilloscopes. A full blown system would have occupied an entire large room, and they expected me to build one from scratch, de-bug, repair and maintain it in addition to being able to do machine level programming ??? The sucker had more switches, lights, wires and modules than I had ever seen before.... The simple lab computer courses taught in school could not have prepared any mere mortal for this... I was in shock - completely & totally overwhelmed ....... That entire summer, I was far more a burden than an asset, as there was a LOT to learn. By the end of that first summer, I was given my own basic machine to build, de-bug and get ready for QC inspection (albeit with LOTS of help from everyone)..... But by then, I had learned the basics of machine level programming, how to read a flow chart and use every feature to it's max potential on Tektronix's most advanced scopes at the time.

Unlike motherboards of today where you simply "plug em in" and everything almost always works, large systems consisting of 100's of thousands of discrete parts always had a multitude of numerous problems - normally totaling around at least 100 or so for even a basic system. It was almost a rare instance when at the first "power on", smoke didn't billow out from some dark depth of the system. It was the checkout tech's job to diagnose & resolve each problem and get the system ready for shipment. Hard failures were often simple to find, but it was the intermittent problems (ie: fails only once every 20 hours or so) that took by far the most time to diagnose. Doing so required an intimate knowledge of the diagnostic software, machine level instruction programming to make a tight "failure loop" that would yield a bright scope trace, & required a thorough understanding of how the system operated. A simple one bay system generally required about a month to get operational, while larger 10 to 20 bay systems often required 6 to 9 months.

Anyways, come the end of that first summer, DEC offered to let me work Saturday mornings to assist and continue the learning process. The following summer (1967), the PDP-9 was introduced (the successor to the PDP-7) and thus began a new learning process, as the entire group transitioned from assembling/debugging the PDP-7, to the brand new latest generation PDP-9.

Although everyone was responsible for getting their own assigned systems operational, there were times when all that was needed was someone else with a fresh approach. Generally after 3 days of making no headway on a stubborn problem, you'd ask whom you thought the most knowledgeable on that component or peripheral to "give you a hand". It was an unwritten rule to help out someone else "stuck", for it would only be a question of time before you yourself would find yourself in the same boat.

Though everyone was well versed & rounded in systems diagnosis and repair, just about everyone had their little "specialty". (In no particular order)

Jack Learson was the undisputed MagTape "Giant". Jack has probably forgotten more about magtapes than most general engineers could even hope to know in total. Just by listening to the "klickity-klacking" of the IBM 57A pinch roller solenoids and air column sounds, Jack could diagnose almost any tape handling problem without ever having to lift a scope probe... Anything I ever knew about magtapes came from Jack, and the knowledge he imparted became even more valuable when I went in to the field... Jack eventually went on to get his full BS degree from Northeastern University studying nights, and left the Systems Group for Product Engineering. Jack retired from DEC in 1997 and is now living in Amherst, NH. Jack was gifted with the ability to teach and instruct. (If he could teach me, he could teach anyone !) Thus when we first spoke after not seeing one another in nearly 35 years, it came as no real surprise that he went back to school to obtain his teaching certificate. Though "retired", Jack now substitute teaches 12th grade math and science classes.

Joe Zeh and Ed Weingartner were the early EAE Giants...... The only people I'm aware of that knew an EAE more thoroughly, were Larry Seligman and Ron Wilson who designed it.... Ed I believe went on to be a manager and Joe later went onto get his BS and passed the Engineering Review Board to became fully certified as a Systems Engineer. Joe Zeh later went on to become Vice President and General Manager of ATI Research, Inc (the manufacturer of popular PC graphics boards) Joe later "retired" to pursue his dream of woodworking and currently operates Swamp Road Wood Works in Worthington, MA. .. Swamp Road Wood Works hand crafts home and office furniture that is functional, pleasing, and unique. His artisanship yields heirlooms worthy of generations past and future. Web: http://www.srww.com

Al Michaud and Jack Williams both had an uncanny knack to diagnose and "resurrect" dead vector and point plot displays. Aligning the deflection circuits to get the vectors to display perfectly formed characters, was an art form that only Al and Jack seemed to possess.... While we all could align a vector display after a day or so of "cursing", Jack or Al would have one "purring" in about 20 minutes. Most of us secretly suspected they were also equally skilled in Alchemy and the "Black Arts" and perhaps even belonged to the same Coven, as no living mortal could ever properly align an early vector display that fast..... I don't recall what ever became of Al or Jack....

Herb Millman who was also a skilled engineering technician in his own right, was promoted and moved up on the management ladder (and rightfully so). Most managers distanced themselves from their subordinates, but Herb Millman though our supervisor, was always perceived as still being "one of us". Herb had that unique ability to do both...... Herb simply managed from a position of respect. Herbie as we called him, was respected and admired far more than we all probably let on and that he ever might have surmised........

Ed Reed became the corporate PDP-7 Giant. Other than perhaps the engineers who designed it, no one knew the PDP-7 better.... Ed later went on to PDP-9 Engineering working under Jim Milton. With his skilled technical ability to take preliminary engineering drawings, schematics and then write service manuals for products not even in production yet, Ed went into technical documentation. There he wrote many of DEC's technical service manuals for products in the final stages of development. Ed stayed with DEC until it was acquired by HP. Sadly, Ed passed away from a heart attack - Saturday, July 12, 2008. Ed was my mentor in my early days at DEC. We stayed in touch all those years. I will miss him dearly...

Ralph Dieter worked at DEC as a Product Design Engineer over a 25 year period and retired in 1992.

Bob Reed was DEC's first first technician with a badge number of 11. Bob "saw it all" from the very first days of DEC & went on to become manager of PDP-9 Systems. Bob was the most honest "straight shooter" and supportive manager I ever had the pleasure to work for. Though as a group we never let on, Bob Reed was well respected and admired far more than he might have ever suspected.

Ed Hilton was the PDP-7/9 diagnostic programmer responsible for the writing of the software diagnostics for DEC's 18 bit product line. Ed and his wife Joan are now living in Anchorage, Alaska. Ed operates a home renovation business while Joan has her own successful antique business.

Bob Pooler passed the Engineering Review Board and was promoted to Systems Field Engineer. Bob left left DEC in October, 1978 to form Air Image Technology, Inc which later became A/V Services, Inc . Bob has since "retired" (?) and now operates Video Interchange. Bob presently lives in Waldoboro, Maine.

I remained friends with Ed and we stayed in continuous contact all these years... I will miss him...

You may wonder why there is nothing negative to say about that period from 1966 to 1969....... If there was, the creation of this page never would have been undertaken.... The truth is, there was nothing really negative to say of any consequence....... We all got along splendidly, and everyone "played nice" ! (for lack of a better way of describing it).......

Course there was the time that Herbie got his office filled to the roof with line printer paper one weekend............ (ok... so we had a slightly unusual way of showing our respect for Herbie) and another time that Al Michaud and Ed Weingartner had a paper tape punch "Chad Fight" ( those 10's of thousands bit's of paper holes collected by the paper tape punch hopper) where Herbie forced both Al & Ed to work an entire weekend (without pay) to clean up the mess they made. (the result of a "war" that got slightly out of hand that employed the use of a "Hurricane Fan" & 100,000 or so pieces of paper "chad".... not unlike one of those escalating feuds between Bugs Bunny & Yosemite Sam !!!)

Needless to say, there were many other "battles" & pranks, but they were always in good if not slightly "warped" fun..... Truth is: we were an independent lot, that perhaps caused Herbie first order migraine headaches on more than several occasions... But when push came to shove, each & every one of us would do whatever was necessary to get their systems operational and reliable for the customer. In today's climate, it might be a hard concept to grasp & might even sound "corny". But every system was built by us from scratch...... right up to flipping the power switch on for the very first time, til final customer/QC acceptance. We each "lived" with that system for months..... Thus each system became our own personal "baby" that had our name on it....... we gave it "life" & it was "Our's" and no one else's..... not even the customers' til it left our sight.......even after the customer took possession and "paid the bill", some of us still viewed it as "ours"...... If it meant working 30 hours straight to get it out on time and ready for customer acceptance, then that's what we'd do...... Herb would never even have to ask.... It was never written or mandated much less even verbalized..... Instead, it was something personal and far beyond the scope of any personnel manual, company policy, or official edict from above.... Perhaps a different work ethic then, as every PDP-7/9 system shipped was a thing of intense personal pride .... Some of it can be attributed to the corporate culture of the time........ but in retrospect, looking back on it far more objectively with nearly 40 more years of accumulated wisdom, that group was unique & I now suspect both Herbie & Bob Reed had a lot more to do with it........ (Probably also explains why we required so little management & supervision......)

How that work ethic & group spirit came to be, still remains a mystery to this day. I've often wondered about the group dynamics in later years and could never put my "thumb" on how it came to be. It just "happened".... It was as if it were a type of infection; and once "infected", it's effects lasted a lifetime. Consequently it comes as no surprise that every one from that early period made out well later in life.

Anyways, why Digital invested so much in training, remains also partly a mystery..... especially without any guarantees anyone would ever stay upon graduation. Any engineering graduate in those days could work wherever he so chose and all the computer manufacturers were in heated competition for technical/engineering graduates. I no longer recall the personnel manager who conducted my original interview & decided that I was the right candidate.... other than his first name was Joe..... Perhaps it will come to me later..... However DEC's "gamble" paid off (for both of us) & I went to work full time with DEC. Despite many offers from the likes of IBM, Burroughs, HP, Control Data, Bell Labs etc., it wasn't even much of a decision.... and one I never regretted. DEC especially in the early years, was a great company - with a great bunch of people to work with.... In reality, it wasn't even a job.. we got paid well for having fun and getting to "play" with some of the world's best high tech "toys" ! Upon graduation from college, Bell Labs made me a tremendous offer. But long before then, I was already "bitten & infected" by the DEC culture. Anyways, I was now assigned to PDP-9 Systems full time.

I always regarded the period (at least from my personal perspective) from 1966 to 1976 as being the "fun" years at DEC. It was one of those rare, unique and wonderful & rewarding experiences that "kids" entering the job market today could not even begin to relate to. DEC in that period is the way I want to remember it.

In later years being in business for myself, I had the opportunity to provide on site services for hundreds of corporations often working closely with their marketing departments. Within minutes, I could almost without fail be able to determine the financial health of the corporation within the first several minutes without ever seeing a balance sheet. All it took was to sense the corporate "energy level". Some projected all the energy of a deserted morgue (and usually had balance sheets to match the perception). By comparison, to walk into DEC in those early years of explosive growth, was an experience that's almost impossible to describe. It resembled total chaos with paper airplanes flying about, chad wars, fishing lines from the windows of Building #5 into the Mill Pond looking for the "Big Strike" - loud laughter, forklift trucks moving systems about - the PA system constantly paging key personnel, whirring fans, Magtapes, Dectapes, Impact Line Printers pounding away etc etc etc....... Being located in a wooden Woolen Mill, any forklift several floors away could be felt and heard as it rumbled along the wide oak floors. It was a constant visual and audible din of activity. Even in the "normal" so called "quiet" office areas, it seemed more like borderline pandemonium (or so it appeared) often with paper airplanes of every conceivable design being launched from behind cubicles amidst a din of ringing phones and IBM Selectric's. What seemed like insanity gone unchecked, instead was the manifestation of a tremendous bubbling energy that couldn't be contained.

It was a unique experience, and I shall forever be thankful for the training & investment that DEC provided......... Today, an opportunity such as that would be almost unheard of......

Without that casual "tip" from a friend, life would have taken a different course. It's often the small things that make for major course changes in life ......

The PDP-9

The PDP-9 was an 18 bit machine with an 8k core memory (expandable to 32k) which included as standard peripherals, a 300 Cps paper tape reader, 50 Cps paper tape punch, and an ASR-33 Teletype/keyboard...... I/O transfer rate was an impressive 18,000,000 bits/sec with a fast 2 µs add time. The system was fully expandable to include options such as, Direct Memory Access (DMA), Real Time Clock (RTC), Automatic Program Interrupt (API), Extended Arithmetic Element (EAE), Memory Parity Option, Memory Protect, DecTapes, Mag-Tapes, Burroughs drums & Disks, Analex Line Printers, GDI Card Readers, AD09 A to D Converter and Vector displays as standard fare. A special Time Share version of the PDP-9 was aptly named the PDP-9T and consisted of 2 PDP-9's controlled by the "T" box.... Dual core processing is nothing new - the PDP-9 supported dual core time sharing more than 35 years ago ! ..... Harvard University's Computer Based Lab (Harvard CBL) in Cambridge Mass purchased the first one and remained in use until the early 80's.... The open architecture allowed interfacing of almost anything else imaginable by the Special Systems Group, which designed one of a kind interfaces.

The rear door supported 3 large logic wings - 8 rows of modules, consisting of 40 modules per row. The top wing housed the 8k core memory, the middle held the CPU and the lower wing, the i/o section. An additional 24k of core memory for a total of 32k was housed in an optional oversized bay, usually directly to the left of the main console. Core memory was no bargain in those days - by current standards anyways, as a mere 8k of 18 bit memory cost $18,000 (and that was in 1967 dollars !)

In those days, the quad wide double height G920 ROM module that housed the 64 word 36 bit control memory, wasn't called a ROM. Instead DEC called it the Read Only Programmable Memory or "ROP Memory" (nicknamed and pronounced "rope"). Remove the aluminum protective cover that also held the two magnet halves secure for each of the "bits", and you would observe bundles of thin wires running through 36 double pole permanent magnets. It was "programmed" by the way each wire passed through either the 1 or Zero pole side of the magnets. As the addressing lines were pulsed, either a one or zero was returned based on the path the address selected wire was routed though the magnets. The bundles of wires looked like a ropes, and many of us (myself included) thought that Rope (pronounced rope - like in what you tie things with) was actually it's real appropriate name. The hard wired memory controlled the logic flow of the machine without any time wasted on unnecessary time states. The ROP memory used for the PDP-9 was actually the early forerunner of the microcode used later in DEC's highly successful lines of VAX/VMS machines and today in virtually every CPU. Today, ROM/Microcode controlled machines are taken for granted, but the PDP-9 was the first commercially successful design of anything in it's price league to employ ROM flow control. Without all those unnecessary time states, dramatic increases in throughput for the same cpu speed was realized. Where the PDP-9 really shined, was the speed of it's arithmetic operations with its optional EAE (Extended Arithmetic Element - a $5,000 option). Efficient ROM based flow control made the PDP-9 the fastest "number cruncher" of any machine in it's class at the time, and was quickly purchased by just about all the major research facilities the world over demanding serious number crunching power at a relatively bargain basement price.

Flow control was able to cut cpu processing times considerably, but there still remained the bottleneck of slow core memory cycle times. The main advantage the PDP-9 offered over the PDP-7, was the speed of it's core memory. Previous designs required 4 wires per ferrite core while the new 2½D memory design required but 3. Memory cycle time for the PDP-9 was reduced to only 1 µs as compared to 1.75 µs for the PDP-7. Since memory cycle time was the main limiting factor in system throughput, the PDP-9 because of it's much faster memory was based on just that metric alone, 1.75 times faster !

The PDP-9L

In an attempt to capture a wider segment of the market, the PDP-9L was later introduced. Priced about 1/2 the cost of a standard PDP-9, the PDP-9L unlike it's "big brother", proved

to be a marketing flop. The PDP-9L did not have the reliable PC09 high speed reader/punch assembly included as standard fare

on the 9, but instead a "slow as molasses" & somewhat unreliable reader/punch incorporated as part of the ASR-33 Teletype that used mechanical sensors to detect the holes. Loading a 3" thick fan-fold program that took perhaps but 40 seconds or so to read on the PDP-9, was more of an endurance test on the 9L, and one could easily walk away and head to lunch as it slowly "chugged" along reading a whopping 10 - 6 bit character lines/sec. The teletype reader instead of using optical photo sensors, used instead 8 mechanical "fingers" that activated micro-switches to detect the holes, which proved to be "rough" on the paper tape. Even a newly punched tape would sometimes generate read errors. Even if the tape did manage to read in correctly when new, after 20 passes or so, the paper tape would be worn out from the punishment inflicted by the mechanical reader... The standard memory was also reduced to a "measly" 4k of core. Engineering cut a few corners too many to keep the price low, and even the systems techs hated working on the things - especially with it's "cheesy" console. Apparently, so did the potential customers, and despite being almost half the cost, relatively few were ever built... DEC didn't make many marketing blunders in those days, but in my personal opinion, that was one of em !

The PDP-9i - The System that Never Was...

Towards the end of the 9's life cycle, there was actually talk of placing the PDP-9 CPU on a chip to provide an upgradeable pathway for the large installed base of customers that had made significant software investments. That in itself made a lot of sense...... With much faster solid state memory now available, the full potential of a Micro-coded flow controlled machine could finally be realized.

Rumor had it that several experimental chips were made for engineering to "tinker" with, but there was justified fear expressed by marketing, that a PDP-9i might actually seriously compete with DEC's other highly successful & lucrative PDP-10 product line....... A PDP-9i with an integrated & embedded EAE (Extended Arithmetic Element) would be a serious "number crunching" contender at a substantially lower price point.... Whether marketing put the "Kibosh" on it or not, remains a mystery.... but those were the rumors circulating at the time.

A standard Reel of Dectape would hold approximately 250k of 18 bit data. (That was quite a bit back in 1966 !)

The optional RD-10 Burroughs disk was a fixed head design and held a whopping 1 Megabyte - the disk platter alone was approximately just over 3 feet in diameter and weighed approximately 45 lbs.... Too many parity errors, necessitated removing the platter and polishing the surface with a special polishing wax. The disk spun at 1800 rpm resulting in a fast 17.6 ms latency. (Pretty impressive for 1968 !) ...Transfer rate was 2.7 mb/sec. Don't have the exact spec, but the drive must have weighed close to 300 lbs..... Cost: $120,000 (and that was in early 1970's dollars !)

The original DEC logo was blue and not red. The entire Corporation in 1966 existed entirely in the Maynard Mill... Official address: 146 Main Street - Maynard, Massachusetts 01754

One quick way to tell if it's a PDP-7 instead of a PDP-7A, is that the original PDP-7
had the I/O panel above the memory frame just left of the console. By comparison,
the PDP-7A had it instead over the I/O bay just to the right of the main console.
This was done since most PDP-7's built were substantially larger than just
a basic system. Thus all of the I/O components (input/output multiplexers etc)
were now installed in a large frame at the bottom of the expansion cab now
to the right of the main console. All peripherals then plugged into that frame, which
resulted in shorter cable runs.

The organization & management structure of the PDP-9 Systems Group was somewhat unique in large corporations. The "chain of command" pathway to the top, was incredibly short.

Engineering floor techs reported to Herb Millman (PDP-9 Systems Area Supervisor)
Herb Millman reported to Bob Reed (PDP-9 Systems Manager)
Bob Reed reported to Bob Lane (Vice President)
and Bob Lane Reported directly to Ken Olsen (President, CEO & Chairman of the Board)

That was it ..... even a lowly PDP-7/9 floor engineering checkout tech was but 4 "hops" away from CEO & Chairman of the Board ! (ok... so for whatever trivial reasons, none of us never made it beyond the first two "hops"......... No matter....... the point being that flow chart wise, we at least perceived ourselves as being close to the top !).... Everything about the PDP-9, even the management structure, was amazingly "compact" and efficient. (I guess you might say that there were no wasted "time states" there either) !

Bob Lane -VP in charge of PDP-9 Systems
Bob Reed - Manager PDP-9 Systems
Herb Millman Systems Area Supervisor ("Herbie")

   Don White
   Ron Wilson
   Larry Seligman
   Jim Milton
   Don Vonada
Ralph Dieter

Software Engineering

Engineering Techs

Ed Reed
Ed Weingartner
Jack Williams
Jack Learson
Al Michaud
Joe Zeh
John (Jack) Duffy ("Duff")
Bob Pooler ("Reppy")
John Majeran
Bob Lane
Ron (Greg) Bahnmiller
Lane Palmer
Harry Nelson ("Bunkie")
Joe DeMaria
Bob Lashua ("Lash")
Barry Eknenroth (Ekenroth ?)
George Dollish
Nunzio Celona
George DiNardi

Wiremen

Bob Towle ("Corky") PDP-9 Wireman
Richard Kruge PDP-9 Wireman
Don LaVallee PDP-9 Wireman
Don Lind Wireman Supervisor
Dennis Credit PDP-9 Wireman
Louis Raymond PDP-9 Wireman & QA Inspector
Ron Cohen PDP-9 Wireman
Bill Whitby PDP-9 Wireman
Ken Norton PDP-9 Wireman
Larry Conley PDP-9 Wireman

Stockroom

Dick Bruce
Don Lind

Administration

Rory Trocki - Systems Group Secretary
Lynne ? (Bob Reed's Secretary)

Quality Control

Don Zereski
Dick Jones

I took this picture while we "torture tested" an entire dual PDP-9 system for use aboard a Navy vessel at a test facility in Florida for Vitro Systems, Inc. - Circa 1973

Components were first subjected to simulating the roll and pitch of the ship, as the Navy was concerned that the gyroscopic precession forces on the large heavy spinning disk platters might pose some "interesting" reliability problems. Kind of a crude way of testing for pitch and roll as pictured here, but it worked ! Turns out, that wasn't a concern.... The worst torture was yet to come.....

To the rear & somewhat hidden, was the country's most powerful shaker table at the time, used to generate powerful vibrations in the range of .1 Hz to 20 kHz.... For our testing, the signal generator was programmed for sweep frequencies typically put out by the ships engines, shaft and propellers.

Just as an aside: The shaker table was nothing more than a over glorified audio speaker, but any other similarity, pretty much ended there. As I recall, the table supported objects weighing up to 2,000 lbs and was driven by a 120 Kilowatt audio amplifier (yes.. that's with a "K") that required 480 volt 3 phase power. So powerful was the audio amp and "speaker", that the testing lab room had to be devoid of glass of any kind. Operations above certain power thresholds required a key operated manual over-ride, due to safety concerns of personnel as well as certain resonant frequencies causing damage to the building structure itself.

Though we never operated anywhere's close to those power levels, the entire system was vibrated & literally "pounded" to the point of failure. After repaired (my job) and design changes made by Jim Milton of PDP-9 Engineering, it was the most "rugged-ized" dual processor PDP-9 ever built !

The testing lab's chief engineer gave us both a "crash course" in the physics of harmonics. I still recall him emphasizing that "The apparent mass of an object effectively approaches zero at the point of harmonic resonance". I was always intrigued by the concept & to this day, I still think about that and its' most interesting implications........ especially useful today in the recovery of vintage 78 records !

Jim Milton of PDP-9 Engineering is in the blue shirt with Cal Howrigan - Chief Engineer of Vitro Systems on the far left. Somewhat fitting I suppose, that the engineering Rep from the US Navy can be seen simulating a large "wave" by pushing from the rear !

Sadly, Jim Milton passed away....

Since the PDP-9 was primary designed around research needs, many of the applications required additional engineering and interfacing. Most larger systems were unique in their integration. As such, folks in Field Engineering got to be involved in some pretty exciting work..... some of it for the military and perhaps still classified.

Both Ed Reed and myself finally moved to Corporate and Regional Support where we both got to see a good part of the country, resolving problems beyond what the local Field Service Reps and their regional support networks could handle. Being the technical resources of last resort and with over 700 complex PDP-7 & PDP- 9 systems in the field in critical applications, we were both kept quite busy hopping from one "crisis" to another. Ed remained the undisputed PDP-7 corporate "Giant", while I fielded most of the PDP-9 support calls....

*******

When originally conceived, DEC figured on a 150 unit PDP-9 production run. Instead, so successful was the PDP-9 System, that 445 systems were built while in full production.... far surpassing all expectations. Though 445 was the official production run number, even after official production ceased, strong demand for the PDP-9 continued - despite release of the newer more powerful and even less expensive PDP-15.

The PDP-9 had such a loyal set of followers and reputation for reliability, that the product simply refused to die. To put that in perspective, I wonder how many folks today would opt for an Intel ® 486 over a Pentium. (absolutely none, I would dare say).

The Traditional Products Line (TPL) would continue to build, refurbish and ship PDP-9's on special order for another several years to come. The last serial number I personally documented was # 527....... Why the additional 82 systems that continued to be built by TPL were not officially counted in the final total, remains a mystery.....perhaps privy to only the inner dark workings of the marketing group.

*******

The purpose of this page is to reflect back on those amazing times at DEC - not on what was to prove to be their unraveling....

There are plenty of books written and articles on the web concerning DEC's demise, all with their own slant on what went incredibly wrong. I won't dwell on those issues other than to briefly add my own 2 cents worth....

The business model/management analyst experts will continue to sift thru the "bones" to try & identify and quantify reasons of what went wrong at DEC perhaps for years to come - depicting with all kinds of spreadsheets, 3D graphs and charts the fine details of DEC's demise..... but when all is said and done, what it ultimately comes down to is that the PC had exactly done to DEC as DEC had wrought to the very same industry only 30 years earlier..... That is; make an effective product that could be sold at a good profit far below what the competition had to offer...... It's that simple......

Perhaps amazingly ironic that in the end, DEC management was unable to grasp the very concept that made them so incredibly successful in the first place....

But the simple reality is that none of us are infallible.......... "20-20" hind site is easy...... any "jerk" can indulge in the luxury of making those calls (myself included)....... The net and book publishers are awash in post-mortem "analysts", all claiming to be "smarter" than the rest of us mere mortals. Anyone can do a post-mortem on what went wrong, but the real trick is trying to predict the future turn of events before they happen.....

Thus I shall always hold in Ken Olsen in the highest regard..... He provided us all with an incredible life experience and opportunity........ wonderful memories that could not even be remotely comprehended by today's generation. Without Ken, none of us would even have such wonderful memories.....

The simple fact is that Ken Olsen managed to turn a small 1200 sq ft operation in the Mill into a 14 billion dollar multi-national corporation that gave even "Big Blue" a run for it's money. That was No small feat....

Though DEC sadly never achieved it's full potential, the simple reality is that without Ken Olsen, I'd never be writing this nor would so many thousands of employees have had such fond memories of those amazing times .......

Consider just a few of DEC's achievements as related to the internet alone:

DEC in 1977 was the first commercial company to connect to the Arpanet (the predecessor to the internet).

In 1985, DEC was the very first company to register an Internet Domain Name.

and if all that wasn't enough, DEC was also the very first Fortune 500 Corporation to even have a web site !

DEC was also instrumental in developing the TCP/IP standards. The backbone of the Ethernet in wide use today.

If anyone recalls other names, has additional pictures or knows the whereabouts of others; please email me: Bob Pooler mail@videointerchange.com with subject heading "PDP-9"


The G920 Read Only Programmable Memory - ROP Memory. Architecture was 64 words consisting of 36 bits An aluminum top cover secured by 9 nylon screws, held a foam pad whose purpose was to compress & hold the top ferrite halves in place. As the foam aged, it lost it's suppleness, resulting in loosening of the cores and corresponding sensitivity to vibration.


Vitro Systems	Bob Pooler in background - Circa 1968

Harvard CBL PDP-9T- Circa 1977 Scott Bradner Only 2 PDP-9T's were ever built

PDP-7 / PDP-9 Systems - DEC

The PDP-9

The PDP-9L

The PDP-9i - The System that Never Was...