Product Architect and Hardware/Software Engineer with Eastman Kodak since 1980.
| Resume... | Contact... | ||
|
The More links go beyond the terse resume text to fill in the rest of the story. I hope you enjoy it. I certainly have! |
|
mail@jemcgarvey.com |
Creating great products requires both a high level vision and a low level attention to detail. The union of these views is the strength I bring to product development: Imagination to conceive the finished product in the user's hands, expertise to execute the details to perfection, and the productivity to deliver on time!
Elegant design means the optimal tradeoff of hardware and software to minimize development time and unit cost while maximizing reusability and maintainability. Simplicity and consistency are the keys!
| Technology Experience | |
| Languages | C/C++, C#, Visual Basic, HTML, CSS, PHP, Javascript, Java, Assembly |
| Operating Systems | Windows, Linux, Android, uItron, uC/OS, pSOS, MSDOS |
| Tools | Visual Studio, Eclipse, SVN, PerForce, GNU |
| Interfaces | PCMCIA, CF, SD, MMC, USB, Firewire, SCSI, I2C, SPI, RS232, NTSC, PAL, ATA, NMEA |
| Formats | XML, SVG, TIF, JPEG, PCX, WAV, IPTC, CVF, SVF, COFF, ELF, FAT, DCF, DPOF |
| Processors | VideoCore, OMAP, DM3xx, C64x, MPC82x, ARM, PowerPC, x86, 80196, i960, 8080, Z80, AVR |
| Hardware | CCD/CMOS Imagers, FLASH, PLDs, FPGAs, LCDs, Batteries/Power Management |
See a video of Jim at kodak.com
Kodak Professional DCS Digital Camera lead engineer from 1990 to 2004.
You've seen thousands of images from DCS cameras - in the newspaper, magazines, catalogs and online.
Probably the best known is the image on the Heroes of 2001 stamp, captured by Thomas E. Franklin using a Canon D2000 camera, a rebranded version of the Kodak DCS520, designed and built in Rochester, New York.
NASA has used the cameras on most manned missions since the early '90s... NASA Cameras. The National Air & Space Museum displays the DCS 460 camera that John Glenn used on his historic return to space in 1998.
DCS cameras were the only DSLRs - the only professional quality digital cameras available until 1999.
In 1990, my digital camera team moved to Kodak's Professional Photography Division and transformed the Hawkeye II design (more on that below) into the Kodak Professional Digital Camera System, the first commercial DSLR. The team faced the new challenges of medium volume production, reliability, FCC certification and publicity.
While we moved the first DCS into production, I worked on the next generation. The goal was a much lower price and a much smaller camera. The DCS 200 was the first commercial single unit handheld DSLR, a completely different product using a much simpler architecture. The project sprinted from back-of-the-napkin stage to shipping in only 10 months!
By 1993, the team was seasoned, our process refined, and the long run of overlapping design and production cycles was well underway. All DCS cameras were manufactured by Kodak in Rochester - just outside the doors of the design offices... DCS Factory Tour
We could respond to factory problems quickly and we learned a lot about design by being so close to production.
The next quantum jump came with the DCS 520 camera in 1998. The first DSLR with on board color image display, it was also a jump in image quality and rapid capture performance. This camera set the bar for the competition that would soon come from Japan!
Growing competition encouraged expansion into new markets. The DCS ProBack line included four digital backs for medium format studio cameras. Only the front plate and a flex circuit were unique to models for Hasselblad, Mamiya and Contax camera bodies.
These camera backs used Kodak's 16 megapixel CCD - a 36mm square chip that produced spectacular images - and required the added horsepower of a TI C62 DSP to process the images.
The last generation of DCS cameras used the first sensor to fill the 35mm camera frame with 14 megapixels.
After 14 years of work together, the original DCS design team was nearly intact. The cooperation and coordination of that team through so many great products was a pleasure - and a standard for the rest of my career!
The DCS Story is a brief history of the product line that I wrote when Kodak exited the business in 2004. It includes details of every Kodak Professional camera model - and some of the twists and turns of pioneering a new kind of product.
Advanced Development architect and software engineer from 2005 to present in Kodak's Consumer Digital Imaging Group.
A colleague and I addressed the growing complexity of consumer camera firmware and the demand for richer graphics by creating KNUI, a data driven UI engine and development tool. KNUI dynamically renders a UI from an XML logic description and an SVG skin, allowing rapid prototyping, rapid product development.
KNUI is fully skinnable and includes excellent support for multiple language localization. The Windows development tool runs the UI on the desktop and includes extensive debug and graphic layout support.
KNUI was deployed in 16 Kodak consumer digital camera models from 2006 to 2009.
From 2008 to 2010, I led a team creating KIP, a Kodak digital camera framework for Linux and Android based smartphones. Code was created for the TI OMAP processor, using the Zoom Mobile Development Platform as a development and demonstration device.
KIP included still and video capture management, advanced image processing, automatic exposure, white balance and focus, and a complete user interface.
Intellectual Property analyst and advisor. From 2004 to 2005, full time support of Kodak digital camera patent licensing and litigation.
I hold eleven US Patents for digital camera technology. Click More to see them.
Kodak people pioneered the digital camera and created an enormous body of patents disclosing all of the fundamental technology of a digital camera. Since the late 1990's, Kodak has very profitably licensed this IP to most other digital camera manufacturers.
I have supported this program throughout those years, providing well documented prior art from the professional camera line, and technical consultation to legal counsel. In 2001, I received Kodak's highest honor for inventors, the Eastman Innovation Award, in recognition of my contributions in the early development of digital cameras.
More recently, successful litigation with cell phone makers Motorola, Nokia, Sony, Samsung and LG has required more of my involvment, including a year of full time IP work. In 2009, I testified before the International Trade Commission, and prepared a video and live demonstration of prior art DCS camera functions.
Hawkeye II Digital Camera lead engineer from 1987 to 1990 in Kodak's Federal Systems Division.
In 1986, Kodak announced the world's first megapixel CCD image sensor, the KAF 1400. It was a monochrome sensor with 1.4 million pixels.
In 1987, I designed the first Digital Single Lens Reflex (DSLR) camera under a government contract. The EO Camera used the KAF 1400 sensor mounted in a custom back for a Canon F-1 body. A logarithmic amplifier, 10 bit A/D, DRAM image buffer, a Conner 3.5" 100 MB hard drive, battery and most of the electronics were housed in a separate shoulder pack.
When the EO Camera was completed in 1988, we built several demonstrator DSLRs. This one is the Hawkeye II Integrated camera, the first single unit DSLR. This camera used a Nikon F3 body and stored images on removable, battery powered DRAM cartridges.
These cameras included SCSI and NTSC video outputs and a simple character based UI with soft keys.
The DCS Story contains much more detail on these cameras.
The Hawkeye II Tethered camera returned to the shoulder pack design to accomodate the varying requirements of government customers.
NASA flew this camera on Space Shuttle mission STS-44 in November of 1991 - the first DSLR in space.
In 1990 we built a Hawkeye II with the first color megapixel CCD, the KAF 1300. This camera became the prototype for the Kodak Professional Digital Camera System.
Selective Image Printer redesign lead engineer from 1984 to 1987 for Kodak's Federal Systems Division. Complete control system redesign for high speed random access contact printer for satellite imagery - dramatically improved reliability, simplified maintenance and lowered machine cost.
The design of the first version of this machine was well under way when I joined the group in 1982. I developed the firmware for the bar code reader that allowed the machine to randomly access images for printing. The machine handled film up to 9.5" wide on rolls up to 4000', slewing up to 500 fpm and positioning within a fraction of a millimeter.
15 DC motors handled film and shutter movement and illumination control. Motors were controlled by a complex network of digital logic, analog circuits and servo amplifiers. Reliability was very poor, and with over 200 pots to adjust to maintain stability of the servo systems, maintenance was onerous.
During many weeks spent at the customer site debugging the machines and training the maintenance staff, I developed my own design for the control system, never expecting that it would actually be proposed for the upcoming construction of additional machines. For a small additional cost, we completed the new design on time, and retrofitted the old machines. The customer was delighted!
Motion control was the central challenge in the redesign. I thought it essential to eliminate the analog servo amplifiers and their many adjustment pots, so I developed a switching amplifier using the newly available power MOSFETs. This amplifier delivered up to 50 volts or 12 amps, yet was small enough and efficient enough to fit on a single circuit board along with a microcontroller and an interface for an optical encoder to provide speed and position feedback. The machine used 15 identical boards - only the EPROMs were changed for the different motor applications.
Integrated motion control boards are widely available today, but not in 1984. This may have been the first.
My father became QA Director for the Singer/Control Division near Chicago in 1978. I found a ideal summer job in his department as a technician, doing the work of a development engineer.
In 1978, I designed, built and programmed a life test fixture for refrigerator defrost timers. Up to 96 timers were mounted on the test stand and operated at 10X normal speed by increasing the AC line frequency and voltage. The computer monitored the operation of the timer switches. Thus years of life test could be simulated in months. I used an IMSAI 8080 computer and programmed in assembly.
In 1979, I wrote more assemby code, this time for an end-of-line motor tester. We replaced an expensive IBM System/7 computer with a microcomputer. This fixture was used in Singer's Crystal Lake, Illinois plant, where motors were test very briefly as soon as they were assembled.
Audio/Visual Equipment service technician for the summers of 1976 and 1977 at Marbaugh Engineering Supply Company in Indianapolis.
At the start of each summer, I faced two rooms filled with a jumble of A/V equipment, most of it from schools. The work ranged from a half hour cleaning dust, pencils and chewing gum out of a record player, to the all day overhaul of a 16mm movie projector. By the end of the summer, the equipment was stacked neatly, ready to go back to school.
This job gave me an appreciation for well made machines - and a disdain for poorly made ones!
Taylor University, Upland, Indiana. Class of 1980. BS degree in Computer Science, Systems Analysis and Business Administration.
I started at Taylor as a physics major and planned to transfer to Purdue later for an EE degree. But the computer science program caught me instead. Taylor had a very practical program, to turn out productive business programmers. We even punched decks of COBOL to run on a General Automation computer. The courses on operating system design and structured programming were the best.
My real niche was in the chemistry department, in a course on interfacing computers to scientific equipment. We used a PDP 8 to build an automatic titration system. Later, as a lab assistant, I ported that design to an Apple II.
My career goal was clear now: Combine my love of electronics and programming to use microprocessors to control machines with well structured, multithreaded, maintainable software.
North Central High School, Indianapolis. Class of 1976.
Electronics... Microprocessors... Programming... It all came together when I was in high school.
Let me back up a few years. My father was an electrical engineer and confirmed DIY-er. He had wired electric locomotives in the 1930's and was an electrician on a ship in World War II.
My brother was a budding mad scientist and mechanic who built various motor vehicles, including a bucket truck for house painting. He assembled a ruby laser in 1965, and the next year built a hyperbaric chamber for a science project on "suspended animation" or "keeping a dog alive during prolonged hypothermia."
So I grew up in the workshop, surrounding by all things electrical and mechanical.
Dad worked as QA manager for the Electronic Timers Company in Warsaw, New York. In 1965, he had an IBM 1800 computer installed to do end of line automatic test for appliance timers. It was the first computer in our small town, and it was not doing payroll or billing, it was dealing with motors and switches. Computers were meant to control machines!
In 1970, we moved to Indianapolis. I was applying my electronic skills to model railroading. In high school, I began programming in BASIC on a time sharing terminal and wrote a submarine war game. It was a little early for robotics clubs, so we had no opportunity to control machines with our programs!
At work, Dad's engineers were applying the new microprocessors to automatic test systems, to replace those old and very expensive IBM computers. He brought home a TTL Data Book and manuals for the microprocessor boards they were using and I was hooked. I couldn't afford a microprocessor chip yet, but I started playing with TTL and dreaming of the things that could be done...
Personal Projects
My Retro Telecine website describes the telecine system I developed in 2009 to digitize my father's 8mm Kodachrome movies. This project merged my interests in photography, electronics, software, optics and old machinery. The emphasis was on image quality - see the Video Album
This site was also a writing project. Writing for the web is a hobby in itself, with the satisfaction of immediate publication and the pleasure of tracking visits from and interacting with people with the same interests around the world.
Other web projects...
Mid Life Cello journals my experience learning to play the cello after age fifty.
God is Kind is web writing from the heart with a spiritual perspective.
My home electronics lab.
In 1997, my sons and I began rebuilding a 1978 Ford F150 pickup. We pulled the cab and box, sanded and painted the frame, fully rebuilt the engine, and reassembled. It even worked when we were finished!