EXSYSCO and National Advanced Systems (NAS)

This Page last updated on .

My name is Mark DiVecchio, I was working at Burroughs Great Valley Labs near Paoli, PA in 1978. I wasn't real happy there. One of my co-workers put this advertisement on my desk one morning (thanks Chuck!) :

In 1978, the company was called EXSYSCO.

(This page was created from my memory and my experiences at NAS. It also relied on copies of the company newsletter and other documents that I saved. I'd like to add more points of view and experiences - email me with yours.)

I flew out to San Diego in March, a few weeks after I sent them my resume. They put me up in a motel in Del Mar which was only a few miles from their offices in the Sorrento Valley area of San Diego. NAS was a PCM - a Plug Compatible Manufacturer - of IBM clones. There were clones, not like the IBM clone on your desk, but clones of large mainframe computers - computers that you could actually walk inside of. Here is an article about IBM and the PCM's.

I met Howard Sachs. He was the VP of Engineering. Howard is fairly famous being the subject of a cover story in the IEEE Spectrum. Howard left NAS in 1979. He went to Cray and then to Fairchild where he developed the Clipper microprocessor. In the 1990's another former NAS engineer, John Petry, and I did some contract work for Fairchild using the Clipper chip.

I interviewed with Paul Jeffs who was a senior engineer there and with Yee Lee who was an engineering manager.

I guess they liked me as they offered me a job in early April.

So I took the job with EXSYSCO and by the middle of May 1978, I moved to San Diego. I shared a cubicle with Paul Jeffs (Paul later left with Howard Sachs to go to Cray). Paul was my mentor, even though he might not have known that. He guided me through my first project, the redesign of the XAR box.

EXSYSCO got its start from a project at Digital Scientific Corporation (DSC) called the META 4/370. DSC previously designed and sold a computer named the META 4 - it was a MECL 10K clone of an IBM 1130. They started work on an IBM S/370 clone in 1974. DSC and its marketing arm, Digital Leasing Corporation (DLC) did not have the funds to complete the project so, on 5 Dec 1975, they sold the project to EXSYSCO, a wholly owned subsidiary of National Semiconductor Corporation (NSC). Some history on this sale and its results can be read in this document from the State Board of Equalization of the State of California : 82-sbe-285.pdf. As part of the original sales agreement, DSC was to get NSC stock based on all computers shipped before 31 July 1977. As it turned out, the first shipment was delayed because of design problems. DSC never received any payment for the sale.

The META 4/370 became the AS/5.

The AS/5 was an IBM clone, compatible with the IBM S-370 model 158 which IBM introduced in 1972. They were marketed by a company named ITEL. They were built of Motorola MECL 10K, a fairly fast ECL MSI logic family (and the same circuit family that I designed with at Burroughs). The boards were arrays of sockets about 2 feet wide and 3 feet long. They were wire-wrapped and interconnected by cables. The computer clocking scheme was hard to believe. There were clocks everywhere, each with a different timing. If worst case path delays caused a failure, the clock was delayed by using a longer wire! We ran all sorts of power supply voltage margin tests to try to find these worst case paths. Of course, in the next computer built, other paths were worst case and other failures occurred.

Click here for another article about the Augat wirewrap panels.

My first job was to redo the design of the XAR. The XAR was a floating point (FP) arithmetic unit attached to the CPU. Microcoded instructions interpreted the S-370 FP opcodes and passed the operands to the XAR. The XAR did the operation and made the results available back to the CPU. The XAR was very unreliable and I was to redesign it. The major job was cleaning up the clocking system. I could only take that so far. Here is a timing drawing that showed the clocks that remained AFTER I completed the redesign.

ALD's were Automated Logic Diagrams. IBM sold these. They were complete schematics for the S-370. EXSYSCO basically duplicated the logic of the S-370. By the time that I joined the company, they were already shipping the AS-3 (roughly equivalent to the S-370 model 138) and the AS/5 (equivalent to the model 158). I don't remember what the difference was between the AS/3 and the AS/5. A few months after I joined, during the summer of 1978, the company celebrated its 100th computer shipment. At several million dollars each, that was a tidy sum of money. (But those prices didn't last for long!)

Most of the technical people in the company previously worked at NCR in Rancho Bernardo. NCR did not like EXSYSCO and, I believe, sued us for hiring away their engineers and technicans.

The XAR was an unusual design - it was microprogrammed and the arithmetic unit was PROM based. It didn't have adders or multipliers. 4 bits of the first operand along with 4 bits of the second operand were used to address a PROM. The output was then the arithmetic operation. These algorithms were already developed by IBM and I did not have to change them. My major contribution was changing the architecture from an asynchronous clocking scheme to a synchronous scheme. My redesign was successful. I don't remember when it was placed into production but it must have been in 1979 (I remember that Howard Sachs was still the VP).

Here is the 1979 NAS phone directory.

EXSYSCO was in four buildings. Building 1 was the engineering group, building 2 was the administration and finance groups, building 3 was the test floor (pictured below) and building 4 was the field engineering group. The whole 4½ years that I worked there, my office was in building 1. Click here for a map.

I'm still in touch with Len Krumpen. We went skiing together for 20 years. Bev Norris died in 1985, shortly after NAS closed down. I don't know what happened to the rest of these people.

At some point while I was there, EXSYSCO was fully merged into National Semi (I think we were named the Advanced Systems Division). Later when ITEL exited the computer leasing business their Data Products Group became part of us and we became National Advanced Systems (but were still a wholly owned subsidiary of National Semi).

The ITEL deal was not all roses:

ITEL got out of the business because IBM put a lot of price pressure on the PCM's. Here is a paper that I wrote for one of my MBA classes at the University of San Diego that described that policy: An Examination of IBM Pricing Policy in the 60's and 70's. Here is a graph from that paper that illustrates the point.

These photos were from a 1980 (ca) National Semiconductor Annual Report

"National has become a major manufacturer of computer systems. IBM-compatible mainframe computers made at the Company's facilities in San Diego, California are used in more than 100 locations throughout the world. Excluding IBM, National is the largest maker of IBM-compatible mainframe computers in the world."

"Experience in producing and marketing IBM compatible add-on memory systems and a 2-year manufacturing history in IBM compatible mainframe computers as encouraged National to take the next measured step of selling IBM compatible mainframes direct to end users. The San Diego Facility that has shipped over 300 computers will be the source of National's new NS8500."

You can see the 2x3 foot wirewrap panels which were the circuit boards. If I recall correctly, my XAR was the bottom board on the first door on the right side of the computer. IBM had designed, and we duplicated, an ingenious serializer which read out hundreds of internal signals and displayed them on the CRT.

Left: Charlie Zapata, Right: Rick Hetherington

Charlie Zapata got his15 minutes of fame and a photo with the boss in the local NAS newsletter.

Another test floor shot. Ed Davis tells me that this is him. See his email below.

My cubicle before the term Prairie Dogging was coined

This was my cubicle in building 1 in 1981.

AS/6100

We started a new design with gate arrays around 1979.

The new design was to compete with the IBM 4341 which IBM announced in that year. This machine was not a clone but a design from scratch. IBM no longer sold ALD's (schematics). We had a codename for the project but I can't remember it (might have been S-203 or at another time - 'Shark'). I recall that the computer weighed in at 2,200 pounds.

It was MECL 10K and used Motorola MECL 10,000 Macrocell Gate Arrays. We designed about 30 unique chips.

Each gate array was mounted on a carrier board.

The carrier board was then plugged into the Printed Circuit Board (PCB). This was done because of the difficulty (at the time) of routing signals between the pins of the gate array if it were on the main PCB.

This is the main memory ECC board for the system. Each gate array processed 8 bits of a 64 bit word. All 8 gate arrays on this board were identical.

If I recall correctly, only two of the gate array designs had to be redone because of design errors.

One of our designs, a 4 bit (or was it 8 bit?) ALU slice became a standard product, the 10800, sold by Motorola. Allen McAfee was the lead designer on that one. The 10800 was laid out with drafting tape on a drafting table - no CAD here (see email below from Joe Militello). I remember checking the design layout using a yellow highlighting marker on a big piece of drafting paper. (whoa! - stone knives and bearskins.) No simulation either - but it worked.

Click on the scan of the cover of the Motorola MECL 10,000 Macrocell Gate Array Design Manual to download a scan of the entire manual (about 18MB). Written by Jerry Prioste of Motorola..

The AS/6100 was announced in April of 1982.

AS/6130 Technical Documentation - this is still sitting in my garage.

A series of articles about the AS/6100 was written by me and other engineers in the design team.

Unlike the AS/5, the AS/6100 was a completely synchronous design. Clock distribution was done by a carefully balanced clock network. Between the boards, the clock was distributed by equal length clock coax cables.

Here is one of the clock distribution coax cables.

Clock distribution coax cable

1982 September - First Customer Ship (FCS) for the AS/6100

The control console showing the boot up screen.

This is the first machine we shipped. Its the big blue box. The AS/6130.

This is our engineering model. It's the one we spent 3 years working on. These were real printed circuit boards which plugged into a backplane. During debug, we had engineers working four 12 hour days covering 7 days a week. Debugging went on for over a year. At least we had a logic analyzer (bottom left). This workarea was in building 3.

The design was synchronous. The black wires that you see interconnecting the boards were the clock distribution network. There were strict rules about clock distribution so that every flip-flop in the computer was clocked at the same time.

This is a picture of some members of our engineering group which worked on the first shipped machine. That's Allen McAfee in the lead. followed by Julie Lilly. In the back, I recognize Chun Lee and Kirby Spalding. I don't recall who debugged this particular computer. It may have been the engineering group or it could have been the test floor crew.

This is the ceremony which marked the first shipment. That's the division general manager, Bob Spencer, addressing the "troops". That's me (Mark DiVecchio) on the right. The blue box (not the dumpster!) is the AS/6130. The customer for this computer was Scripps Clinic and Research Foundation in San Diego.

We installed a half-dozen or so of these. They were working great. Here is a the serial number plate for machine number 2. I don't recall why I was able to get this unless it was because of the engraving error, it had to be redone.

Here is an article that I wrote about the AS/6100 for NSC newsletter.

This is from the NAS newsletter:

Around December of 1982, the president of NAS left

Kvamme ended up as a big shot in the computer business in the Bay Area working at Kleiner Perkins Caufield & Byers.

1983 January - The layoff.

The end came on Monday, January 31, 1983. We had shipped a half dozen or so AS/6130's but in 1983, the division closed and we were all laid off. We later learned that our AS/6100 development was in competition with Hitachi (who supplied NAS with larger IBM clones to sell). The company management decided that they could buy a similar computer from Hitachi rather than manufacture the AS/6130 in San Diego.

The field engineering group was not laid off and it merged with Hitachi to form Hitachi Data Systems (HDS). They continued to service the AS/5 computers out there and they began to service the Hitachi computer. They have been in San Diego ever since but the business has changed a lot.

I talked with one of accounting guys who was kept on to liquidate the division, he managed to "liberate" one of the circuit boards for me. That is the ECC board pictured above.

This photo from a party that the former engineering group held about October 1983. The party was held at my condo in Mira Mesa. By then, I was working at Megatek.

1983 - Here is part of the design engineering group. Taken at a get-together sometime after the layoff.

Back row (left to right):
Jerry Falkman, Gary Griffin, Dave Bilak, Chuck Stead, Julie Lilly Clauss, Mark DiVecchio, Kirby Spalding

Front row (left to right):
Buck Titherington, Sarkis Artinian, Tom Sarnowski, John Petry, Chun Lee

(back then not a gray hair in the bunch but you should see us now)
Buck Titherington passed away in 2010.
I've lost touch with Sarkis but still talk to the rest.

Missing: Stu Kroll, Colin Isenman, Ralph Mullins, Steve Getz, Clay Dahlberg, Chien Nguyen, Dae Woo Lee, Amir Asvadi, Ken Randery, Shen Wang, Tony Valenzuela, Mike Fowler, Jay Wheeler, and Yee Lee.

Newsletters

Here are scans of just the cover of a few of the NAS newsletters. I (as a typical pack rat) saved 5 years worth of these.

1978 Sep
1979 Oct
1980 March
1981 Aug
1982 Jan
1982 April

Other NAS story lines

IBM 3033 Clone

NAS also started a project to clone the IBM 3033. A building was leased, building 5, to house the development group. Click here for a map. I knew a few people there but they were mostly new people. I think the project was named 'Lightning'. That project ran for a while and then was canceled (approximately from 2Q 1978 to Apr 1980) resulting in the layoff of about 120 people . They had built a prototype but it was not working. It was an MSI design also cloned by using IBM ALD's.

Just cloning a circuit as designed by IBM was not a simple matter. I don't think the company understood the issues. I figured out much later that the biggest technical problem was that IBM's MSI chips were architected differently than MECL 10K. In many cases, we had to use two chips where IBM used one. This caused a lot of propagation delay problems.

I/O Processor

There was a project to develop some kind of I/O processor. I think it was part of the 3033 clone. That project was also canceled. Our AS/6100 team managed to snag a few of the engineers before they were laid off. Names that come to mind are Ralph Mullins, Kirby Spalding and Chuck Stead.

Campus Point Buildings

At one point in 1979, NAS had great plans to expand into a new set of buildings on a nearby plot of land called Campus Point. That never happened and SAIC ended up there.

Other IBM Clone Companies

IBM mainframe clones were also manufactured and sold by, Amdahl, Storage Technology Corporation, Trilogy Systems, Magnuson Computer Systems, Fujitsu, Hitachi, and probably a few others.

ITEL's Final Days

ITEL finally went belly up in 1981. At one time, because of its rapid growth, it was expected to become a case study for the Harvard Business School. I guess it still could have but now as an example of what not to do.

The New IBM Computers

About two weeks before the layoff in January 1983, I bought an IBM PC. It had two 360KB floppy drives and 64KB of RAM. The display was a simple composite video based TV monitor. It cost me over $5,500. The times had started to change......

Emails

Date:            Thu, 18 Sep 2008 19:16:56 -0400
From:    "William \"Buck\" Titherington"
Subject:         How ya doin'

Hi Mark,

Just happened to run across your web page. Thought I could fix the fact that we lost touch.

I've been here and there since NAS & Megatek, and finally back to the east coast, Rhode Island (EMC & LTX), Conn. (to care for my in laws), and now moving to Maine. Divorced again, married again (to my high school sweetheart) for about 20 years now.

Like you, I'm "between jobs", would like to have a bit more fun before I hang it up forever. Things are disgustingly quiet so I joined a Volunteer Fire Dept. to stay a little busy.

That's about it, stay in touch,
--
Buck Titherington

Buck,

Wow. Good to hear from you. It has been a long time.

As you could probably tell from my web pages, I've 'toughed' it out in SD all these years.

After NAS, it was a short stint working again and then I went off on my own. I had a small (usually 1 person) company named Silogic Systems for about 20 years. I did engineering consulting and logic design. In the early 2000's I went to work for one of my clients, AMCC, worked there for 3 years and then retired.

About half of the people in the old NAS photo are still in SD. The rest - who knows. Chun, John, Dave, Chuck. Gary and Julie are still around SD.

Kirby and Ralph are up in the bay area. I worked with Gary at AMCC . I see Dave Bilak at a Star Party that I do once a year.

If you want any emails of these people, I have a few.

I will forward you an email that I received in 2006 from a Tammy D'Imperio. She wrote me looking for you and I've been saving her email with the faint hope that you would surface - and you did.

Mark

I learned today from Joe Militello that Buck Titherington passed away in 2010.

Here is his obituary.

You will be in our memories, Buck.

Subject:         RE: Buck Titherington 1942-2010
Date:            Mon, 16 May 2011 16:07:44 -0700
From:            "Falkman, Jerry (ETS-Elec)" <Jerry.Falkman

Honeywell.com>

Mark,

Sorry to hear about Buck. He was a good guy.
Thanks for the update,

Jerry Falkman
Sr. Engr., ETS Engineering Tucson, Honeywell Aerospace

Subject: Re: Buck Titherington 1942-2010
From: Gary Griffin <whenslunch

cox.net>
Date: Mon, 16 May 2011 18:27:54 -0700

I remember traveling with Buck to Raytheon in Boston to work on cross-talk issues with those old wire-wrapped boards at National. He was a fun guy to have around.

--Gary

From: w.bonneau

att.net
Subject: Been A long Time
Date: Sun, 07 Jun 2009 23:37:43 +0000

Mark,

I was searching for an old reference for EXSYSCO/NAS in San Diego and came across your web-site. I do not think that you will remember me but here goes; I am Walt Bonneau and was a Associate Engineer/Control Console Tech while at NAS from 1977-1980. As with you, I moved to Megatek in 1980 and remained there until 1984. I left San Diego for Texas where I worked for Texas Instruments for over 10 years. From there worked for Sony Corp for nearly 4 years. Created my own company which I sold, and joined Cubic Corp for the last 11 years.

I was simply amazed to see your web site and the family history as well. I took notice of the visit you took to Idyllwild in 1999. My wife and I have land up there and plan to build on it next year. I just remembered when I was a very young engineer that you were always very helpful.

Anyway, just wanted to say hi,

Walt Bonneau
Escondido, CA.

Subject:    Exsysco
Date:    Wed, 8 Jul 2009 03:00:35 -0700
From:    "Moyer, Jack" <jmoyer49

sbcglobal.net>

Mark,

I decided for some reason to Google "Exsysco" and see what I would find.

Imagine my surprise to get your website and crawl through those memories from so long ago. I was the Jack Moyer you responded to in the advertisement and signed your original offer letter. Yes, NCR hated us and used to call Floyd Kvamme to tell us to stop hiring their people (especially the test floor which was virtually all ex-NCR) - Floyd would call me and I'd stop for a week or two then back at it. I still keep in touch with Bob Spencer from time to time and with Tim Harris who followed me to San Diego after I returned to Santa Clara for another HR role.

Great web page - thanks for the memories.
Regards,
Jack Moyer

Jack,

Thanks for the email. You really are a voice from my distant past.

As young engineers at NAS, we had a chance to work in an exciting industry. I'm amazed at how young everyone was (at the time).

For about 20 years after the layoff, we did a mini-reunion every January for anyone who worked at NAS. I'm still in touch with many of the engineers and technicans that I worked with.

May I use your email on my web page? I can leave your email address off or use a home email address if you prefer.

If you like, you can expound on your memories as well. It would be interesting to hear how HR operated to staff the company.

Mark

Subject:         RE: Exsysco
Date:            Wed, 8 Jul 2009 18:23:35 -0700
From:            "Moyer, Jack" <jmoyer49

sbcglobal.net>

Mark,

Feel free to use my email in your website.

HR was primarily recruitment focused during my time. I was transferred to Exsysco in November '76 and returned to Santa Clara for NSC in February of '79. We would grow from 50 to 1100 people during my time and zero to $100 million in revenue - a real wild ride. As you can imagine we were hiring at an unbelievable pace and totally focused on getting people on board. In early '77 HR was Debbie Watson and me - I added Terry Mick in late '77 and then grew to a total of 7 or so people by the time I went back to Santa Clara. Randy Bresee who was our controller after Mike Woodward left and I are still friends and often reflect on the hard work and great fun. We were all young, energetic and at a great time in our lives. We worked and played hard. There were organizational and performance issues, and our share of drama, but all in all the experience survived the test of time.

I forwarded your site to Bob Spencer and he is going to go through some of his old material and forward anything that is of additive value.

Thanks for the note back.

Regards,
Jack Moyer

From: "Ed" <davis9580

verizon.net>
Subject: Excellent Exsysco Website!!!
Date: Sat, 20 Mar 2010 09:15:02 -0700

Mark,

I think I remember you. I left Exsysco in 1978, but I stayed in touch with the people. That unknown Tech in your photographs is me. I worked with Paul Hastings, Bill O'Neal, Dennis Daniels, Buck Titherington et al in the back of the Digital Scientific building before it was sold (given) to Exsysco.

Here's what I looked like back then:

I talk to Buck Titherington once in awhile. He moved to Boston. I spoke to Hanan Potash a few times after he started his Cray clone type company with George Kenoshita. I also worked with Chuck Laustrup a couple years here at Intel in Oregon. He remembers working with all those guys too. Old Bob Hinkle shot a guy in a bar in Arizona. And crazy Bob Shows!

What a flood of Memories!

EXCELLENT WEBSITE!!!

Thanks,

Ed Davis
Chief Architect (retired)
Intel Corporation

To: "Ed" <davis9580

verizon.net>
Sent: Saturday, March 20, 2010 9:33 PM
Subject: Re: Excellent Exsysco Website!!!

Ed,

Thanks for your email. I'm not sure that I remember you. I started at EXSYSCO in May of 78.

Which group did you work for?

I got an email from Buck a year or so ago. I just added it to the web page. I still hear from Dyke Summers every once in awhile. I remember Bob Shows. I haven't heard the names Hanan Potash and George Kenoshita in 25 years.

After the layoff in 1983, we got together on the aniversary of the layoff. This happened for almost 20 years but we haven't had a reunion in a while.

A lot of the guys in Field Engineering ended up working for what remained of EXSYSCO which became Hatachi Data Systems so we saw them occasionally at the reunions.

I've added your email to the web page. Let me know if that is OK with you.

Mark

From: <davis9580

verizon.net>
Subject: Re: Excellent Exsysco Website!!!
Date: Sun, 21 Mar 2010 10:00:18 -0700

Mark,

I noticed I forgot to answer your question and I also noticed in my email that I had said "Brad" Daniels, I believe his name was Dennis. He was Manager of Field Service, Paul Hastings worked for him.

I worked for Stu Kroll and Bill O'Neal. I was a lowly Tech back then still going to San Diego State and UCSD. I joined Digital Scientific, Dave Ahlgren, et al on the Meta 4/370 project around Oct./Nov. 1975. I really enjoyed working in the back of Digital Scientific in a small group of only 5 or 10 people.

When we transitioned into Exsysco and moved to Roselle St. the company exploded to over 1000 people, and the fun was gone so I left in early 1978 just before you arrived, but I remember your name vividly, because a Stock Broker friend of mine was Michael DelVecchio and I remember telling him about you and how your names were so similar.

One other thing regarding that email. I recently retired from Intel, so you may want to put an (ret.) after my title. Here's a presentation http://bt.pa.msu.edu/TM/BocaRaton2006/talks/davis.pdf

I left for Michigan State, where I did my PhD work in high energy Particle Physics. We had just gotten 80 processors to run on a single die when I did this presentation in Boca Raton.(The Polaris project)

Thanks very much,

Ed Davis

From: "jclauss04@cox.net" <jclauss04

cox.net>
Date: Thu, 1 Apr 2010 14:51:34 -0700
Subject: A voice from the past....

Wow, love your website. Reading about National Advanced Systems surely brought back a lot of memories...thank you so much! I was also very skinny then!

I think the last time I heard from you was about Derek May...anyway, glad to see you and Sally are still enjoying retirement as I would like to someday. Have you gotten all traveled out?

Anyway, after I left AMCC, went to Brooktree (2 yrs) then to Schumacher (3 yrs) then to Cymer (7 yrs) and at present I am with Qualcomm and tomorrow I will celebrate my 3rd year anniversary. Needless to say I will be here till I retire which hopefully will be next year. Of course, my tax guy thinks I am crazy. Both my sons also work here at Qualcomm, they've been here 25 years, collectively.

I lost those photos you sent me the last time we had a get-together with Ralph, Allen, John and Gary. The hard drive on my computer crashed and everything was lost. I was keeping in touch with Ralph for awhile but lost his email.

Keep in touch,

Julie Clauss (formerly Lilly)

Date: Sat, 07 May 2011 18:21:51 -0700
From: "Chips Etc." <info

chipsetc.com>
Subject: AS/6100 processor question (pics attached)

Hi Mark,

We enjoyed your website, lots of great computer history and photographs!

We are researching a piece of NAS AS/6100 computer memorabilia in our chip collection. It is a LGA ceramic processor we believe is from around 1982.

Was this type of processor made by Hitachi or Motorola? (picture of the item is attached). It looks similiar in style to some of our Amdahl air-cooled processors.

Thanks in advance for your help in identifying this processor, we want to make sure we get our description correct on our site.

regards,
Jane (Chipsetc.com)

Jane,

The chip that you have does appear to one from the AS/6100 that I worked on in the early 1980's.

If you look on my web page for the AS/6100, you can see several photos of the chip. We mounted the chip on a daughter board.

The chip was a gate-array and was fabricated by Motorola using their MECL 10K Emitter Coupled Logic Circuitry.

I may have a data book from Motorola on the gate-array family.

Mark

Thanks for all the help identifying this as Motorola, now im wondering - why wouldn't NAS use a chip manufactured by National Semiconductor? Was it due to patents or did national semiconductor not have anything like the MECL to offer it's NAS folks at the time?

Great hobby, learning alot from folks like yourself, I will defintely check out that data book...thanks for offering that to all to see...

Jane

Jane,

Motorola was the only source for the high speed ECL circuitry. Signetics was a second source (I think they had a sourcing agreement with Motorola).

When you get the photos on your website, let me know and I will add a link to them from my web page.

Mark

We do have some pictures up now including close-up's of the chip, they can be found here <http://www.chipsetc.com/national-semiconductor.html>.

thanks again
Jane

From: "Militello, Joseph G" <joseph.g.militello

intel.com>
Date:    Mon, 16 May 2011 11:09:56 -0600
Subject:    Hi Mark!

Hi Mark,

Hey, long time no see! I found your website and I'm so jazzed to see all the familiar faces once again. I worked at NAS under Dave Adams, in CAD
as a 23 year old pcb designer, and also did gate array layout. Remember those 133x enlarged cell maps on mylar, and especially the 'beast' that I was
able to squeak by with the last 3 hook-ups - "The Mover"? I remember the bottle of champagne and t-shirt that engineering gave me for
completing it. I have so many great memories of working with great people there at NAS. That place at the time with that select group of
people was one in a million!
I'm working for Intel now outside of Portland, OR.
Wish you the best,
Joe Militello

Enclosed a pic to jog your memory. I looked more like Barry Gibb in those days ...   ;^)

Date: Wed, 20 Jul 2011 16:55:20 -0700
From: <techtrain

cox.net>
Subject: NAS website, memories

Hi Mark-
About 6 months ago I found your great NAS website. You did a fantastic job!

I worked there from July 1981-Layoff. My first job was in the memory dept. and later on the AS6100. It was an amazing experience, and at one point I actually believed the company would be around for a long time--an old fashioned job! Here are a few memories:

To prep for the AS6100, we had one INTENSE month of training on theory of ops and troubleshooting of the wirewrap predecessor. I remember dreaming about the logic diagrams. Do you have any in your book--the ones with gates rendered with ASCII symbols in rows and columns, all kind of looking the same? I'd love to see one of those on your site.

Before going to the system debug dept., I brought up the service processor and power system--lots and lots of Amperes! Occasionally exciting! The ID tag in your photo is 35A at 208V = 7KW+. As one of our trainers said, a bad programmer could turn it into a $500K room heater!

BOARDS:
The great fire: In early 1982, 2 boardsets ($100K+) were destroyed by a burn-in chamber failure. One giant step backward.

Many of the devices were not layed out on std.on 1/10 in. centers, so automated testing was difficult. In the summer of '82, they used brute force and hired dozens of temps to ohm out every node connection. The defects were passed to techs to sort out due to the rework wires and cuts and drills! We spent hrs. looking through layers of artwork to track these down.

Business Stuff:
At one of our monthly meetings I asked why we didn't use National ICs (as did a previous poster on your site). They were right there in those beautiful data books--why not use "vertical integration"? The response was that our chips didn't meet engineering req'ts!

At these same meetings, there were charts that showed us losing money, but at a lesser rate than they anticipated, so it was o.k.!

National had a great stock program for the common worker. The only thing that really worked out--I bought in at $19 in 1982 and sold for $57 in May of 1983!

FCS
The first production units were very difficult to bring up--huge amount of rework on boards and backplanes, not always done right! It seemed like we were way behind. I recall the original clock speed goal was around 56nS, but ended 10-20% slower. Seemed like we were in trouble. I worked the 2nd shift (4 day week thing) and when I came to work on FCS day, from the view of the parking lot from the I-5 freeway ramp, I thought the crowd was there because of a layoff meeting! As soon as I saw the champagne glasses, I knew it couldn't be that. You might say, as a 2nd shifters we were kept in the dark. Once the boards were cleaned up, it went pretty smoothly, but little did I know that it was too late!

Hitachi
Right after I started in in the memory department, they started building Hitachi memories. The most striking difference was that they soldered the memory chips (vs. socket them for replacement, anticipating failures). Also, the poor smokers had to step outside.

Names-
Gary Griffin and Ken Davis were a great help in production. Later, I worked with Dyke Summers (Celerity Computing) and Chuck Stead (Overland Data). I still meet people who worked at NAS.

It was a privilege to be near the cutting edge so long ago--it gives me a perspective that makes the new developments in technology so mind blowing--the "youngsters" simply take it for granted!

Thanks again,
Dane Tovey

This site prepared and maintained by Mark DiVecchio

email : markd@silogic.com

DiVecchio HOME
Frazzini HOME
Site HOME

This Page last updated on  .