music bio gigs Bourland's Technology Page Einstein Z. DogPhotos
Technology? I though Bourland was a musician?
Guitars, grades and graduation
The peaks and valleys of silicon hills
Bourland rides the lightning @ Dell
Transcending the hierarchy from software to silicon
The wonderful world of analog mixed signal design
A new lease on life @ AMD
Original Bourland technology page

Technology? I thought Bourland was a musician?

Bourland is also an electrical engineer. Having spent most of his early life taking stuff apart, most of which wasn't likely broken to begin with and probably never worked again. Bourland just liked to know how things worked. It is true, music was his first love; it was not his last. Music introduced young Bourland to guitar effects and amplifiers and he definitely wanted to know how they worked. Cruising along playing music, skate boarding and generally slacking in school there was one constant: MATH. Bourland was good at math. Though he rarely won the math races on the chalk board, Bourland was methodical and deliberate. There was a certain intuition that he couldn't always explain and a disciplined approach to problem solving. As fate would have it, when Bourland was a senior in high school, an engineer from NASA came to his class for career day. Bourland's science teacher was worried Bourland wasn't going to go to college and choose to chase a dream of fame and fortune in music. Bourland had just received his SAT scores and while they were a little lacking in English and such, his math scores were very good. As it turns out, the engineer from NASA was also a musician and brought a guitar, an amp, guitar effects and an oscilloscope to the class with him. Finally, there was someone right in front of Bourland explaining how electric guitars, amplifiers and effects worked! He brought up the waveforms on the scope and explained the decay in the guitar signal and inductors for pickups and just about every damn question Bourland could ask in an hour or so. Holy crap!  Bourland was going to college!

Guitars, grades and graduation

So off to college the young wild eyed Bourland goes. Rule # 1, guitars and grades don't get along. After a horrible first year, Bourland took his guitar home and left it there. His grades improved dramatically but that would not be the last challenge. Academia didn't always suit Bourland. He wanted to understand the core theory not just mechanical memorization and regurgitation. Still, his math skills got him through and when he finally hit his upper level electrical engineering classes he discovered he was also good at analog. While the entire economy and world were racing along towards a digital future, there was Bourland finding out he was good at analog. Engineering students were beginning to flock to digital design. Today the ratios are horrible with only 1 out of 20 electrical engineering students choosing analog, But as things work out, the world is analog. No way around it. Nature doesn't deal in 1's and 0's. So being good at analog turns out to be a wonderful thing and as the digital revolution keeps turning, it keeps creating new needs for analog to translate all those 1's and 0's into something we can understand. At Texas A&M, Bourland meets a professor by the name of Dr. Randy Geiger and from that moment forward, Bourland knew he wanted to design analog integrated circuits. Microchips. Microelectronics. Silicon. Cathedrals of Sand. But it was not going to be a direct path. Completing his senior project, a bi-quadratic, adaptable, current mode filter fabricated using the MOSIS program, Bourland graduated having made A's in every class related to analog including Dr. Geiger's ELEN 474 VLSI class.

The peaks and valleys of silicon hills

It was 1990. The semiconductor market was in the dumps. Jobs were scarce and Bourland wanted to live in Austin, Texas so he could play music again. While working as a co-op at Datapoint in San Antonio, Bourland discovered a writer named George Gilder from one of the chief architects while doing research for a paper titled The renaissance Engineer. Not only describing how Bourland saw himself but also describing the need for modern engineers to be well versed in a wide range of disciplines and social skills. The best design engineers are those that can understand the customer's needs, often times because they are a user/customer themselves, and solve the problems of use not just technology. To do that one must know stuff. The classic case of the Palm vs. the Newton. The Palm had that perfect blend of usability and technology like catching a wave. Good surfers miss waves sometimes and great companies miss the ball. Fast forward look and that damn ipod! There was some article in Forbes by George Gilder called You Ain't Seen Nothing Yet. And Bourland wanted to know what it he hadn't seen. That led to the book Microcosm by George Gilder and in fact Bourland  nearly failed a class the following semester because he couldn't put that damn book down. Bourland was reading about how the quantum revolution was overthrowing the industrial revolution. How quantum physics and semiconductors were overthrowing the world of Newtonian physics, planes, trains and automobiles in the world's socio-economic and political landscape. As a side note, this is also where Bourland discovered supply side economics through the book Wealth and Poverty also by George Gilder. Wealth and Poverty has been said to be the basis for Regan's trickle down economics or voodoo economics as it's known by by it's critics. Bourland was officially a capitalist, believing in the entrepreneurial spirit of enterprise. He would start his own company someday.

Bourland rides the lightning @ Dell

By chance, as Bourland looked for a job in Austin, Texas hoping to get on with Motorola, he applied at this little company called Dell Computer Corp, formerly PC's Limited. Michael Dell was some kid Bourland's age that had started a company in his dorm room building personal computers for his friends. Bourland was the first new grad engineer hired into the verification team. Back then the plan was: install every hard drive, network card, video card, software and operating system ever known to cause a problem and see what happens. There were a total of 100 engineers in R&D at Dell and more work than you could get done. Bourland held various positions at Dell from test to design but in the end Dell was a marketing company that happened to have engineers. Some brilliant engineer at that, but still marketing made the decisions and they weren't always that great. Intel ruled with an iron fist. We used to say "Intel doesn't have customers, they have hostages." But to his credit, Michael Dell knew he had to be allied with Microsoft and Intel. He killed Dell Unix. He ran every other CPU core out of the building and he built a relationship, while tenuous at times, he built a second to none, BFF, relationship with the two companies that had all the cards. No doubt Michael Dell is one of the most adept businessmen minds on the planet. When Intel said use this 386SL and 360SL in your new notebook instead of that 386SX even though engineering was ready with the SX and it would take a new circuit board and BIOS to get the SL architecture working. Worse yet, the SL had lots of errata. But Michael knew he had to find some leverage against Compaq. The story goes that when Compaq started using AMD 486's, Intel suddenly didn't have 586 samples and documentation for Compaq and Dell beat Compaq in 586 shipments not only by being first out but in volume too. Dell went from something like100 million/year to 2 billion/year between 1990 and 1994 and passed Compaq, eventually becoming the # 1 PC company in the world. Even with the long hours and challenging work environment, it was an amazing ride. A one point in time, Bourland would share a cube with none other than Chris Goggans, Erikb, editor of Phrack, of famed LOD vs. MOD hacker wars and the Steve Jackson games raid. It was a crazy time in the world and in Bourland's life. Fast company, fast cars, rock and roll and fast women. In the end Bourland wanted to do analog IC design and when Motorola SPS came knocking with a nice job offer, Bourland had sold his  meager 300 shares/options of Dell stock and headed for greener pastures. At the time, Bourland looked like a genius. He sold his stock near an all time high of 49/share and not long after it was down around $13 as Dell hit it's first losing quarter in history. Bourland used the money to buy his first oscilloscope, a 200MHz Tektronkix TDS 310, and a 65 Chevy pickup. Oh yeah, and that hat he always wears, but that's another story.

Transcending the hierarchy from software to silicon

Bourland had arrived. He was finally at a semiconductor company. The one he had always wanted to work for, Motorola SPS (Semiconductor Product Sector). The good guys. The one's with the white hats. A CPU architecture without a colon in the middle of the address space. Analog integrated circuit design! Oh yeah. But not so quick young man. Fist stop: product engineer on the MC68HC11C0 microcontroller unit (MCU). Yes, ladies and gentlemen product engineering. What the hell does a product engineer do? Well. basically everything after the designer is done. Tests, tests and more tests. Guarantee that the parts get to the customer fully tested and without huge yield loss. Basically cleaning up other people's messes, but it was still great. Bourland would go down to the cafeteria and there were engineers everywhere. You sit down at a table, any table and they were talking about capacitors and MOSFETs and resistors, diffusion resistors, poly resistors and diodes, PLL's (Phase Locked Loops, oh so analog) and ADC's (Analog to Digital Converters). He was in heaven. Playing guitar in The Texas Philistines and working at Motorola. Life was good and it got better. He gets a write up in the Austin Chronicle for his Growler guitar effect. But Bourland still wanted to do analog integrated circuit design, damn it! And he was gonna find a way to make it happen. Spending about 2 1/2 years as a product engineer, a great way to get introduced to the world of semiconductors by the way, Bourland was itching to get into design. So he did what any good engineer does when he wants something, he gets a job offer from somewhere else. In this case it was AMD. AMD had just bought NextGen and that team was looking for a verification/design engineer. Bourland would have to leave his beloved Austin, but damn it he wanted to do design. When Greg White, the general manager of Motorola SPS HC11 group, got word, he called Bourland into his office. He says, "Bourland what's it going to take to keep you here at Motorola?" You see Bourland had already done his homework. Mike Gladden had just moved into the office down the hall and he was the analog mixed signal design manager. Bourland had already stopped by to ask Mr. Gladden what it would take to get into analog design, would he need a masters degree? Gladden said, "hell, Bourland all I need is a req, we can put you to use right away!" Armed with that, Bourland told Mr. White, "I want to make this much money and work for Mike Gladden." Without hesitation, Greg White said "consider it done." It was 1997 and Bourland was officially joining an analog mixed signal design team. Life was real good! Of course there was the occasion tinge of pain as he watched Dell stock split, over and over and over again. Ouch! Man, that was an expensive O-scope and vintage truck. Don't forget the hat. That beautiful cowboy hat.

The wonderful world of analog mixed signal integrated circuit design

Life in the wonderful world of microcontrollers and CMOS semiconductors was good. Working for Gary Daniels, Greg White, Brian Wilkie, Pat Laviolette and Mike Gladden was as good as it gets. Pat Laviolette was a design manager that believed in a disciplined, methodical design flow that was heavy on reuse, verification and test including Digital Virtual Tester. Product engineering no longer had to wait for silicon to see if their test programs worked and test engineering was part of the design team.  It was the engineering equivalent to Camelot. "But not so fast young man", Gladden tells Bourland "before you get to do design we have a job for you. You see, we have that HP83K ATE tester and we need you to generate the patterns and the test program to evaluate this PLL and ADC." "Great," Bourland thinks, "I finally get into design and I'm back doing product engineering and test." Of course, it was an excellent way to learn about PLL's and soon Bourland was doing design and had already made himself useful to the team. First design job: PADS. Microchips have to interface to other microchips on a printed circuit board (PCB). They need power pins and IO (Input/Output) pins and analog pins and clock pins. Those pins on the packaged silicon need to get to the silicon and in this case that was visa vi a bonding wire, a landing pad of top layer metal on the silicon chip or die. Oddly the landing pad is just a slab of metal. There's a lot of electronic circuits and serious considerations like ESD and Latchup attached to those pads. Not sexy like a Low Frequency Receiver, more like fundamental. Blue collar design engineering. You get the point. Over the course of 11 years in design Bourland worked on many projects and saw the company go through  good times and difficult times. He watched as the company that he knew and loved  (Motorola) lost its way, spun off the semiconductor unit and neither were ever the same. My bet is the semiconductor unit now Freescale will find it's way . With Rich Beyer and a new found focus on analog, maybe life will get back to normal. Bourland had the good fortune to work on the following projects and collaborate with some brilliant engineers while at Motorola/Freescale

  • Designed I/O pads and pad ring for a 144 pin 32-bit MCU. Bourland's first design project. Needed new 3V pad driver with 5V tolerant input. Took over for an engineer that had left the company. Project was delivered on schedule, with first pass success. Pads are an excellent introduction to CMOS integrated circuit design. They are custom circuits, meaning the layout process is manual. So an engineer becomes familiar with all the layers. CMOS semiconductors are doped silicon (active) regions on a doped silicon substrate. Above that are oxides, poly-silicon gates, more oxide (oxides are use as insulators in semiconductor processes) and more layers of metal with vias to connect between metal layer. Each of these layers has to be physically drawn on the computer, a process known as layout. In the case of pads and pad rings in a Bulk CMOS process, ESD and Latchup are serious considerations with diminished ability to simulate and predict behavior prior to tape out.
  • Developed verilog RTL control block for a wheel speed timer circuit for an automotive ABS MCU. Tasks included verilog code development, synthesis, place and route, clock tree insertion and evaluation pattern development.  Worked on an amazing team. The analog designer, Patrick Falvey,  was top notch and a is still a close friend. The verification engineer was a real hoot, hippie beads on his cube entrance, naps under his desk, yoga and crazy loads of holistic medicine in his cube cabinets. Strangely, for an enlightened being he sure could be confrontational. Regardless, he was good. We were the long pole in the schedule, the critical path, the whole project and yet we delivered on schedule. This project is where I gained a deep respect for the leadership of Mike Gladden. He nailed that schedule with in a week or so. While other managers played schedule chicken waiting for someone else to slip their schedule, Gladden came out and said the WST will be done on this date. And after being on the critical path the entire schedule, the WST was the second  macro done and fully functional on fist silicon.  Bourland was living a dream. RTL, synthesis and automated place and route?  It was the silicon compiler from George Gilder's Forbes article about Carver Mead. 1000's of lines of verilog RTL code get synthesized into an actual circuit and laid out automatically. This was cool stuff but still Bourland wanted to do analog design and he would get his chance but it would mean parting ways with Mike Gladden.
  • Designed a charge-pump PLL to integrate the loop filter and add built in test modes for TSMC 0.25u process.  PLL included delta-Vbe current reference, transconductance amp, differential ring oscillator and a phase-frequency detector.  Developed verilog RTL and synthesized feedback divider. Bourland finally had an analog design project. It meant leaving the strong leadership and wonderful fellowship of Mike Gladden's team and starting a new team with Art Collard. The project turned out to be a nightmare as engineers in the India Design Center kept leaving and several senior engineers on the US design side left during the project.  The sponsor of the part and leader of the Sea of Gates (SOG) System on a Chip (SOC) methodology , Chase Studor, left the company taking some very talented engineers with him. Bourland did what he always does, deliberate, methodical execution. Eventually the part taped out (taping out is a term that actually goes back to a time when the design database and mask sets representing the physical process layers and shapes were put on tape and sent to the fab). The PLL came back and worked just fine. Bourland had achieved first pass success on his first PLL design. He was officially a successful analog designer.
  • Debugged a broken crystal oscillator amplifier. The crystal amplifier was new and the designer had left the company. The PLL worked just fine with an external clock source but turn on the crystal amp and the MCU was dead. Nothing. No clocks. There turned out to be two problems. First, the PLL and crystal amplifier required a current and/or voltage reference from the bandgap. The bandgap had a digital pulse starter instead of an analog self start mechanism. The crystal amp had to generate the clock for the the bandgap pulse starter but it needed the bandgap to start. A system level issue with an easy metal only fix to get the pulse from the POR (Power On reset) signal. The second problem was that AGC (Automatic Gain Control) was broken. Again a simple metal only fix was found. Both were verified with a FIB (Focused Ion Beam used to edit microelectronic circuits after fabrication). This was a particularly difficult time for Bourland. Can you say Re-Org? A reorg had happened. Art Collard was let go and Bourland had a new boss. The part he had worked on had been transferred back to Pat Laviolette and Mike Gladden and Bourland wasn't going with it. He had a new boss, Jim Feddler, and a new boss's boss, Chase Studor. Yup he was back. And now he was the big cheese design manager. Out of respect for Mike Gladden and Pat Laviolette, Bourland fixed the crystal issue instead of working on his new assignment under Jim Feddeler. Not a great way to start under a new manager, partuicularly if his name is Jim Feddeler, but the time it took Bourland to fix the issue was a fraction of what it have taken womeone unfamiliar with design and that person would have been Bourland's buddy You Got The Falvey! Not something Bourland could do. Over the years Bourland developed a reputation as someone who just got it done.
  • Developed a Matlab model for pll loop stability analysis including 2nd, 3rd and 4th order loop filters and running both linear time-invariant and discrete time analysis.  The Matlab program includes 2nd, 3rd and 4th order loop filter component calculator based on desired phase margin and Gardner ratio. The model was used to successfully address a wide range of customer issues from reference spurs on an RF transmit channel to simple filter component calculations and parameter tweaks required for process transfers.  Performed PLL inter-modulation analysis of a wireless USB transceiver. This is a period of time where Bourland worked with Jim Feddeler to overcome the initial adversity in their relationship and learned that Chas was a pretty damn good person.  Life was good again. Hell Bourand had even impressed the Fedd with a deep pretty strong understang of phase noise in a wireless PLL. Reference clock spurs, intermodulation, thermal noise, 1/f noise and the like. Deterministic vs. random noise.There was old Ludwig Boltzmann in every formula for random noise perturbation functions. Bourland learned about state variables and worked closely with Taiwan and Suzhou design engineers fixed a range of issues and mentoring junior engineers while solving complex issue remotely. Bourland was back and ready to take on the world. Once asking Mario Rivas in a group communication meeting why Motorola didn't have an 802.11 solution and why weren't we working on a universal  MAC layer. Only to be ridiculed by the senior technical contributors. Why would Mario care about a MACC?" Thinking Bourland was talking about Mutliply Accumilate instructions. When what Bourland was talking about with the Media ACcess layer. It turns out that MAC layer in wired and wireless ethernet (802.3 and 802.11) share a large amount of logic and it was likely one could create a single MAC layer that could support both and likely also support 802.15 (Bluetooth). Leaving the risky physical layers off chip on the first round, one MCU could go into three vertical markets. Meanwhile the physical layer for 802.3 could be developed in house, put on a test chips. Initially to be sold separately, then integrated to create a single chip MCU. Upper management decides to buy 802.3 IP and integrate it rather than develop in house. It was rewarding that a few months later, Mario had a brown bag lunch and all he could talk about was 802.11. Those same senior technical contributors had to eat crow. Mario left to go to Philips to work on and Bourland went back to the grind.
  • Led a small team to develop two low power internal clock sources and a verilog RTL control block for low power MCU operation. Tasks included mentoring two new design engineers, spec development, system level architectural definition, RTL development, test development and driving the schedule for all modules.
  • Developed a fully differential, low power, low frequency magnetic inductance receiver with auto-zero and built in test modes. Worked with a global project team to develop two similar modules from concept through implementation. Tasks included schedule development, architectural investigation, spec development, spice simulations, test chip development, schematic capture and layout, and silicon evaluation.  Delivered on schedule with 1stpass success. The coolest project period. Designed for remote tire pressure sensing, the low frequency receiver was able to receive 2mVp-p Amplitude Shift Keyed (ASK) Data to wake up the MCU and the movement detector could sense when the car was moving. Bourland uses his dimplomatic skills to improve the relationship between the Austin and Brazil analog design teams, life is good. Bourland turns down another handsome offer from AMD and is granted U.S. PAT # 7,236,014
  • First place winner of the 2006 Freescale Free Your Mind Design Challenge: developed design challenge guide lines and format now being used for the FTF Design Challenge. Entered a design for an automated, Zigbee enabled, solar powered green house called the Green Machine. Left Freescale to join AMD prior to the prototype phase. This was part of an initiative at Freescale called Free Your Mind. The Free Your Mind team is Bourland's fondest memory of Freescale and was without a doubt the # 1 thing that made leaving Freescale so difficult.
  • IO pad library development supporting low leakage 0.18um and 0.25um technology.  Tasks included schedule development, spec development, coordination with global teams in Suzhou, China and Munich, Germany, mentoring jr. engineers in Suzhou, test vehicle development and oversight of evaluation (ESD, Latch-up, AC/DC performance), generation of all chip integration views (.lef, .lib, .v, .gds, lvs) and timing characterization/simulation. The project that drove Bourland out of Freescale and onto greener pastures once again. A thankless job where Bourland worked 7 days/wk, 80-100hour/wk for six months and quit playing music to deliver the smoothest new pad library release in multiple generations in terms of SOC integration view generation and product quality silicon on the first pass in two technologies. But still thank you Mr. Feddeler, a brilliant engineer and technical leader that just should not have been Bourland's manager. He used to tell Bourland regurlarly, "Bourland, you should be my boss." He never knew how right he was but Bourland didn't know it at the time either. Feddeler used to say, "Bourland, you're better at getting other people to do their work than anyone I've ever met." But as with most brilliant people, Jim Feddeler coun't take his own advice. And to be honest it's something that plagues the analog design community. As rare as analog designers are, analog designers with people skills are ever more rare. When it comes to managment, you need people skills. Unfortunately to be successful in analog design you need experience and deep technical knowledge. So analog design managers are often the most technical guy or gal on the team. Sometimes it works, sometimes it doesn't. In this case, it finally didn't work for Bourland. Word is, things have improved and there's no doubt Bourland is nonetheless grateful for all that he learned while working with the brilliant engineers on Feddeler's team and around the globe. It was not an easy thing to leave.

A new lease on life at AMD

Bourland finally takes a job offer from AMD. The best career move in the last several. Bourland goes from working on 0.5um CMOS design in 2004 with PLL VCO (Voltage Controlled Oscillator) speed of 120MHz, to 0.045um or 45nm in 2006with VCO speed of 5-10GHz. Bourland found a culture at AMD that totally fit. More on this soon. For now lets just say Bourland is working on 32nm and having the time of his life.

Ok, so some of this is half baked. We'll do a re-read and apply proper editorial filtering like spelling check and add links. Hopefully the readership will be low because its getting published. Hope you enjoyed the read.

The original Bourland tehnology page

Technology has played a wonderful role in world of music. In fact, during the early days of our developing understanding of the physical sciences, say 1600-1800's, it was music and sound that were the technology of the times. Amplification was mostly acoustic, but that all changed with the advent of the vacuum tube in 1879. The triode was not really invented until about 1906, but it was Thomas Alva Edison that paved the way with that simple little light bulb. Anyhoo, what always amazes me is how dramatically electronics has influenced the way we relate to music and sound. It is a harsh reality that the abstract nature of the acoustics link in the audio chain has obscured its importance. The sudden availability of controllable gain ushered in 125 years of electronic audio experience that changed our relationship to music forever. The feeling of comfort and familiarity when using microphones, amplifiers and loudspeakers is entirely lacking in how we relate to the acoustical (i.e.. physical) environment in which those devices operate. Acoustics is the science of sound with two distinct natures: physical and psychological. Sound as the fluctuating molecular densities in air is physical; sound as perceived by the ear is psychological. Electronics widened the gulf in our understanding between listening to the music of Bach or Pink Floyd and those darn fluctuating molecular densities in air. Bourland wants to narrow that gap with devices like theremins on steroids and The Bourland Beat Box, an improved stomp box being created specifically for John Pointer. John may be one of the few folks capable of actually driving the crazy contraption.


But Bourland is into more than just closing the gap between the acoustic and electronic worlds. Here on the technology page there are links to cool stuff like Wi-Fi QoS , audio effect circuit design, great audio codecs, historical information like the story of Ludwig Boltzmann and new technology like Carver Mead and Neuromorphic IC design. Check out George Gilder's latest rant on Wi-Fi and how he believes his dearest Qualcomm has so much more to offer. Soon Bourland will have information on what he calls the Gildered Age. The age of abundance. Basically, Bourland plans to use this page to give his take, for what it's worth, on a wide range of technology subjects.  The audio circuit design portion of this site will eventually move over to Bourland Audio Labs, once he has some noteworthy audio electronics to offer. Bourland will also give updates and report on the trials and tribulations of a circuit designer, entrepreneur, artist, musician, mad scientist as he goes through the pain, sweat and toil of creating a successful new product or two.  Expect an update soon!

Many engineering companies today seem to have lost the creative edge. Many of today's established engineering firms struggle to imagine anything outside what they already do and worse yet it seems they have lost the intellectual curiosity required to be innovative. Imagination, ingenuity, creativity, discipline, competition and good fortune combine to make some focused people very wealthy, but it all starts with an idea. Bourland has had many ideas, but he is not wealthy. He is not the CEO of a technology company. Creativity alone is not enough. Focus and discipline are required. Striking the balance between form and function, performance and cost, usability and development time are well and good, but at the end of the day it comes down to discipline, vision and faith. Vision is not taught in engineering school. Sure engineers learn discipline by solving complex math and physics problems through long nights studying, but the visionary is different. The visionary sees things not as they are but as they imagine them to be. The whole build it and they will come idea. This is why Bourland believes in supply side economics. People didn't know they needed a cell phone or an mp3 player or a Bourland Beat Box for that matter, but someone out there had an idea and the focus, discipline and faith in themselves and others to make it a reality. That is the significance of Ludwig Boltzmann's story. If he had the kept the faith for a little longer, he would have lived to see his ideas proven true rather than hanging himself thinking he was a failure. So when Bourland struggles and fails, when he thinks "Dang, I must be an idiot!" He just thinks to himself "six more months and I might be a genius".