UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

The $10 Xilinx Box, the FPGA Board, and the Man from HP Labs: An Xcell Daily triple mystery story

by Xilinx Employee on ‎08-16-2017 12:05 PM (62,418 Views)

 

It was a sunny day last week in Sunnyvale but I was down the street in Santa Clara killing time before the exhibits opened at the Flash Memory Summit. (The exhibits never opened, but that’s another mystery.) I found myself drawn to one of my favorite Old Silicon Valley haunts, the recently relocated HSC Electronics (formerly Halted Specialties). As usual, I drifted to the darkest corner of the store, to the half-price table, to see what treasures awaited. That’s where I found the old Xilinx box—and a triple mystery.

 

 

The Mystery of the $10 Xilinx Box

 

 

HSC Box top.jpg 

 

 

The first things that I saw on the box were the Xilinx logo and the old company tagline: “The Programmable Logic Company.” That told me that the box was old because Xilinx has a much broader portfolio these days with a broader tag line to match:

 

“All Programmable.”

 

 

HSC Box End.jpg 

 

 

 

A shipping date on the box confirmed its age. It was shipped on March 19, 1998 from 2300 55th Street, Boulder, CO. That’s the old address for the Xilinx facility in Colorado before it moved to its present Longmont location in the early 2000s. (That address is also around the corner from one of my former employers—Cadnetix—but that’s a different story too.)

 

The next thing that caught my eye was the box’s destination: 1501A Page Mill Road, Palo Alto, CA. All engineers that have ever worked for HP (including me) have that particular address burned indelibly into their brains. That’s the old address for HP headquarters back in the Bill and Dave days and it was also the address for HP Labs.

 

Xilinx had shipped this box to HP Labs. To someone named Dave Moberly—a different Dave.

 

I peeked inside of the box and found a board marked “FPGA Demo Board.” It had two Xilinx devices on it: a 2nd-generation XC3020A and a 3rd-generation XC4003E. I also found a pair of short cables.

 

 

 

HSC FPGA Demo Board.jpg

 

 

 

As the Editor in Chief of Xilinx’s Xcell Daily blog, what are the chances that I would leave this box on HSC’s half-price shelf? None whatsoever. This was a box full of questions waiting to be answered—about the box, the board, and David Moberly—the man who had ordered this box.

 

I paid HSC’s price for the box: $10 (after the half-price discount).

 

 

 

The Mystery of the FPGA Board

 

I didn’t work for Xilinx in 1998 so I was not familiar with this demo board. Google was familiar with it, of course. This board was used to create designs with the Xilinx Foundation Series development system. The Xilinx Foundation Series of tools, introduced in 1996 shortly after Xilinx merged with NeoCAD, was an early development platform for VHDL- and Verilog-based design that ran on Microsoft Windows 95 and Windows NT. Here’s what the 1998 Xilinx data book had to say:

 

 

“The Foundation Series provides a complete, ready-to-use design system for the design of Xilinx programmable logic devices. The Foundation Express System incorporates advanced synthesis technology from Synopsys, and provides design entry (schematic and HDL), VHDL and Verilog synthesis, simulation, and device implementation tools for a broad array of FPGA and CPLD devices targeted for low density and high volume applications.”

 

 

According to the 2Q 1999 issue of Xcell Journal, Xilinx had shipped more than 10,000 Foundation Series seats in the two years following its introduction.

 

These days, FPGA-based design using Verilog and VHDL are part of the natural order of things but back in 1998 the transition from schematic-based to HDL-based design was still taking place, driven by the increased device density made possible by Moore’s Law.

 

And that brings us to the XC4003E FPGA on the Xilinx FPGA Demonstration Board.

 

 

 

HSC FPGA Demo Board 4000E Closeup.jpg 

 

 

Xilinx introduced its 3rd-generation XC4000 FPGA family in the early 1990s and the XC4000E series, fabricated using an “advanced” 0.5μm process technology, first appeared in 1996. This is the point where FPGA technology really started to ride the Moore’s Law curve, as you can see from this graph taken from Steve Bailey’s article titled “FPGA Technology Drives Design Software,” which also appeared in the 2Q 1999 issue of Xcell Journal:

 

 

 

Soaring FPGA Device Complexity.jpg 

 

 

 

The year 1998, when Xilinx shipped this box with the FPGA Demo Board to HP Labs in Palo Alto, immediately followed the knee in the device complexity curve. That’s the point when FPGA device densities began to make HDL-based design for FPGAs very, very interesting—compelling in fact.

 

To learn more about the XC4000 series and its place in Xilinx history, I had a conversation with my friend Austin Lesea who currently works as a Principal Engineer the Xilinx CTO Office. Austin joined Xilinx during the last days of the XC4000 program to work on the last generation of XC4000 devices: the Xilinx XC4000XV FPGA family, which was fabricated with a “very advanced” 0.25μm process technology.

 

Austin explained that the XC4000 family elevated FPGAs from glue logic to something considerably more important. Among its many advances, the XC4000 family introduced the concept of LUT RAM—use of LUTs for on-chip memory—and fast carry chains. Rapidly evolving process technology allowed the device designers to cram enough LUTs on the chip to build the necessary multipliers for DSP work and these features were especially important for DSP applications where you could build multipliers from the LUTs and fast-carry chains and use LUT RAM as local memory to hold multiplication coefficients.

 

As it turns out, that’s precisely why David Moberly at HP Labs was interested in Xilinx FPGAs.

 

 

 

The Mystery of the Man from HP Labs

 

David Moberly is the last part of this little mystery story. His LinkedIn bio shows that he worked at HP Labs from 1978 to 1988 and then again from 1992 to 2000. When Agilent Technologies split from HP in 2000, Moberly followed to Agilent Labs and worked there for another year. So when Moberly ordered the Xilinx FPGA Demo Board in 1998, he was working his second tour of duty at HP Labs.

 

I found a mention of David Moberly’s work in the December 1988 issue of HP Journal, in an article titled “A High-Speed Optical Time-Domain Reflectometer with Improved Dynamic Range.” The article describes HP’s first optical TDR, the HP 8145A. The article’s acknowledgements says:

 

 

“We would especially like to acknowledge the contributors on Steve Newton's team at HP Laboratories: Scott Foster, Robin Gifford, David Moberly, Moshe Nazarathy, Rick Trutna, and Paul Zorabedian.”

 

 

 

HP 8145A Optical TDR v2.jpg 

 

 

HP 8145A Optical TDR

 

 

 

The HP 8145A incorporates a DSP Engine, detailed in another article titled “Data Processing in the Correlating Optical Time-Domain Reflectometer” that appears in the same HP Journal issue. The article states:

 

 

“The Digital Signal Processing Engine is a powerful 32-bit digital signal processor with an architecture designed especially for the correlating HP 8145A Optical Time-Domain Reflectometer. With its 200 MSI ICs, the DSPE outperforms any available microprocessor or signal processor for its tasks.”

 

 

That sentence gives you the scope of the DSP challenge in this instrument, designed a decade before David Moberly ordered the Xilinx Demonstration Board.

 

I wanted to reach out to David Moberly to discuss his subsequent work and ask him more about his work with FPGAs. Sadly, I soon discovered via Web search that he’d passed away earlier this year—on March 9, 2017. That probably explains why his Xilinx box ended up at HSC.

 

However, that’s not the end of the story.

 

I decided to contact the HP Labs team manager mentioned in that 1988 HP Journal article, Steve Newton, who had become the first Director of Keysight Labs when Keysight split from Agilent in 2014. I connected with Steve via LinkedIn and then asked him about David Moberly.

 

Steve told me that the HP 8145A Optical TDR project was his first project as a project manager at HP Labs. He also confirmed that Moberly had worked on the HP 8145A’s signal-processing board. That’s the one that used 200 MSI chips to implement the HP 8145A’s DSP Engine. He also said that Moberly was “generally at the leading edge of things.” That drive to the leading edge took Moberly to Apple in 1988 but he returned to HP Labs in 1992. Newton wrote of that later period: “I recall that he was a very vocal champion of the use of FPGAs for signal processing.” He also wrote: “David was a very kind and sensitive person.”

 

Unfortunately, that’s the end of the trail. So far.

 

 

Postscript: The Xilinx DSP story has certainly continued to evolve since that FPGA Demonstration Board was shipped to David Moberly in 1998. The Virtex device family, introduced in 1998, immediately followed the XC4000 device families. The first Virtex FPGA family was closely followed in 2000 by the Virtex II FPGA family, which added hardened 18x18-bit multipliers (and much, much more) to the programmable-logic mix. Those multipliers have proved invaluable for high-speed DSP. The newest members of the Xilinx Virtex device families, the Virtex UltraScale and Virtex UltraScale+ FPGAs, integrate massive DSP capabilities. The Kintex FPGA series—derived from the evolving Virtex architecture—now includes the Kintex UltraScale KU115, which has 5520 DSP48E2 slices. Each of those UltraScale DSP48E2 slices incorporates a 27x18-bit multiplier. (See “The UltraScale DSP48E2: More DSP in every slice.”)

 

I call the KCU115 FPGA the “DSP Monster.”

 

 

Comments
by Scholar ronnywebers
on ‎08-17-2017 05:07 AM

cool ... my first Xilinx device was a XC3064A, which I used for my thesis in 1994. I did build a 6 channel (uncompressed) digital audio network, running over a single twisted pair, using TDM & Manchester coding/decoding & PLL clock recovery (external PLL at that time :-). 

by Newbie endric.schubert@missinglinkelectronics.com
on ‎09-12-2017 09:44 AM

That brings back some fond memories from the "early" days of FPGA design. Thank you for this nice piece!

Endric - Missing Link Electronics

by Newbie cherylbesden
on ‎01-31-2018 10:13 PM

Greetings - I am the widow of the mystery man from HP Labs, David Moberly.  When David passed away in March I had much to "dispose of".  I made regular trips to HSC.  I am so happy to hear that some of David's collectibles are appreciated by engineers in the area.  He would be happy to know this.

 

My son came upon this story on the Xilinx site quite by accident this morning while riding the train from Palo Alto to SF for work.  He was so surprised to see his dad's name and this story!

 

Cheryl Besden

by Xilinx Employee
on ‎01-31-2018 10:59 PM

Cheryl,

 

Thank you so much for leaving the comment. I'm glad you're pleased with the story.

 

--Steve Leibson

 

by Newbie tomshoup
on ‎02-01-2018 12:01 PM

I was David Moberly's next-to-last manager at Agilent and had hired him into his last job at HP/Agilent from elsewhere in HP, around 1999.  HP Labs at 1501 Page Mill Road was incubating a new medical business that was far different than the existing capital equipment medical business.  We were designing patient monitoring equipment to be used to monitor patients with congestive heart failure at home, to be offered as a service to Medicare HMOs.  We built a simple set of instruments to measure weight (yes, we built a bathroom scale, but it was an HP bathroom scale), blood pressure and heart rate, and a single lead ECG rhythm strip.  We introduced this product to the market in 1999 at the Heart Failure Society of America at their annual meeting in San Francisco that year.  After about 18 months we had ~5,000 patients under monitoring for a marquee customer.  Philips, which bought the medical business from Agilent in 2001 still offers a current generation of this equipment and the associated service.

David was one of several people who basically came up to me and said "I've always wanted to work on medical equipment but didn't want to move to Boston (location of HP's medical business), so here I am."  He was one of the most inquisitive people I've ever met, irritatingly so sometimes, but always in an endearing way.

I left Agilent in 2002 when they sold the business to Philips but kept bumping into David and learned he had a love affair with FPGAs.  One product he was over the moon about was mixed-signal acquisition systems that are basically an FPGA that provides 2 to 4 channels of analog 'scope function, 16 channels of logic analyzer input, some power supply output, spectrum analysis, arbitrary waveform generation, and other features, all using a laptop as the user interface.  Somehow David became the go-to guy to test drive these because of his blogging about them, so manufacturers of such systems would send him new devices to play with and hopefully write them up in his blogs.

David and I ended up working together again early in 2013.  I was a contractor at a medical-device company and the company needed help writing software test plans and protocols.  I introduced them to David and they approved me hiring him as a subcontractor.  We worked together for about another year then.  David was retired then and wasn't really looking for work, but was intrigued with the product and the chance to work again.  He told me afterwards that he paid for his daughter's wedding with that gig.  He also gave me a wonderful thank-you gift: a brand new Analog Discovery by Digilent, one of those FPGA-based mixed-signal acquisition systems.  I've used it a lot in my current consulting work and think of David every time I use it.  David's fondness for these instruments rubbed off on me and we exchanged a round of e-mail as I lusted after the Picoscope 3000, the Cadillac in David's words; I still lust after it but haven't taken the leap. I know if I buy it David would be proud, and if he was still alive would probably want to borrow it to take it apart.

Through mutual friends I knew David had had a bout with cancer.  We met for coffee after he was in remission and true to form he told me in detail about his treatment, with his usual level of fascination at the technology, almost to the molecular level.

After David died, I knew his wife Cheryl was faced with the too-common task of being the widow of a pack-rat engineer who had excess storage capacity: think containers.  So I volunteered to help Cheryl sort through the Moberly archives and introduced her to HSC Electronics as a way to recycle some of what David had collected.  Having known David, and ending up sorting through his stash with Cheryl, I could easily imagine the glee he must have felt when he acquired all those goodies we sorted through.

David had a wonderful career:  MIT education, Apple, Trimble, HP, HP Labs, Agilent, Philips.  A small team of engineers, with a couple of Davids in it, could probably build anything and run it with an FPGA and a couple of AA batteries.    

Labels
About the Author
  • Be sure to join the Xilinx LinkedIn group to get an update for every new Xcell Daily post! ******************** Steve Leibson is the Director of Strategic Marketing and Business Planning at Xilinx. He started as a system design engineer at HP in the early days of desktop computing, then switched to EDA at Cadnetix, and subsequently became a technical editor for EDN Magazine. He's served as Editor in Chief of EDN Magazine, Embedded Developers Journal, and Microprocessor Report. He has extensive experience in computing, microprocessors, microcontrollers, embedded systems design, design IP, EDA, and programmable logic.