|Page (1) of 1 - 08/10/17||email article||print page|
Semiconductor Pioneer George Perlegos Honored With Lifetime Achievement Award(August 10, 2017)
SANTA CLARA, Calif., Aug. 10, 2017 (GLOBE NEWSWIRE) -- The Flash Memory Summit today granted its 2017 Lifetime Achievement Award to George Perlegos, founder of three successful semiconductor firms, for a portfolio of IC memory innovations that have contributed to the development of virtually every leading edge electronic product on the market today. Mr. Perlegos contributions include:
N-MOS EPROM N-MOS EPROM (Intel (INTC), 1974). Gordon Moore said the erasable programmable read only memory (EPROM) was "as important in the development of the microcomputer industry as the microprocessor itself. 1 From computers to pacemakers, the processors that power them need programs to tell them what to do. In the old days programs were stored on large expensive external disks. In 1970 Dov Frohman developed the first EPROM (Intel 1702) using a P-MOS floating (unconnected) gate structure. In 1974 Mr. Perlegos designed the Intel 2708 N-MOS EPROM based on a new process that used channel injection and a single dual layer polysilicon cell to create a dense, fast access non-volatile memory. The device did not require power to maintain its contents and could be easily erased for reprogramming by shining UV light through a window on the package. When paired with the Intel 8080, the first commercially successful integrated processor, low-cost, low-volume systems became feasible. Electronic system design migrated from the assembly of a myriad of hardware-only components to the use of software as the core of the design. This development constituted the dawn of the embedded computing era. (ISSCC, Feb 18, 1977. Patent 3,938,108, filed Feb 3 1975)
+5 volt only EPROM (Intel, (INTC) 1976) - Early EPROMs required 3 different voltages to operate (-5V, +5V and +12V), necessitating an array of external power supplies. Mr. Perlegos exploited a self-aligned polysilicon gate structure and ion implantation in the channel to enhance channel injection, enabling single 5V operation. Intels 5 volt 2716 eliminated the need for the extra components, reducing both the size and cost of end-products. System design was simplified, and battery powered products became a realistic possibility.
16-Kbit EEPROM with single-byte erasure, 10,000 read/write cycles. (Intel, 1978) - Previous non-volatile data storage memories were small (<1Kbits) and could be erased and re-programmed only a few times. Building on work done by Dawon Kahng (Bell Labs, Patent 3500142) Mr. Perlegos utilized two-layer poly stack-gate technology and very thin oxide (<100A) tunneling for both program and erase to create Intels 16 Kbit 2816, non-volatile data memory. For the first time single bytes could be changed without having to erase and re-program large memory blocks, and with an endurance of 10,000 erase/write cycles local data storage no longer needed battery back-up. In addition, with the elimination of the UV light erase requirement, the device package could be made smaller and the device could be erased in-situ. Henceforth portable electronic systems could update themselves in the field which greatly simplified and accelerated the growth of electronic portable devices. (Patent 4,203158. Continuation of Feb 24, 1978. ISSCC, Feb, 1980. Patent 4,266,283, Feb 16, 1979.) (Figures 1, 2, 3)
5-volt in-system programmable/erasable EEPROM (SEEQ, 1981). First generation EEPROMs required high voltage (+12V) pulses to erase/program necessitating the addition of an additional power supply in system. At SEEQ Technologies, Mr. Perlegos utilized oxynitride dielectrics and an integrated voltage multiplier to eliminate the need for the external 12V supply and support the tunnel programming from a single +5 volt supply. These advances simplified system design and reduced programming power requirements to reduce battery size and costs. They remain in use today, throughout the flash memory industry, in virtually every embedded control system (smart phones, laptops, cars, IoT devices, etc.). (ISSCC Feb 11, 1982) (Figure 4)
First 5-V Only NOR flash memory. (Atmel, 1988). Although EPROMs were low cost they could not be erased and reprogrammed in-system. EEPROMs allowed in-system changes but were expensive. Mr. Perlegos used proven tunneling erase/program technology to introduce the first 5V only Flash memory which allowed large blocks to be erased at once, providing a solution for the nascent cell phone market. With the capability to change the program code remotely Flash memory allowed the early cell phone makers to update their products after shipment which accommodated the evolving standards.
Worlds first flash microcontroller. (Atmel, 1994) Under Mr. Perlegos leadership, and employing many of the previously mentioned innovations, Atmel combined its Flash memory process with a CMOS logic process to produce the first Flash based microcontroller (MCU), the 89C51. By separating the program and data storage memories, the 89C51 allowed remote updates in-situ while maintaining permanent data storage. To take full advantage of the capabilities of on chip Flash and EEPROM memory, the Atmel team optimized a new microcontroller architecture (AVRTM) to produce a next generation capability of embedded intelligence. Integrating the Flash based MCU with specific peripheral circuits has led to low cost single chip solutions for a growing universe of embedded systems which continues today with the IOT industry. (Patents 5493534, 1999 and 6032248, 1998) (Figure 5)
In 2000, seven years before the existence of the smart phone, Mr. Perlegos predicted, ?Youll want your phone to do everything. Youll want to see a picture on your phone someday and youll want to look at the stock market and be able to buy something. And what everyone wants to be able to do is E-commerce, which means images and security. As we bring images to phones youll need a camera. . . . Youll be able to scan your finger and it will recognize you and turn your cell phone on. Nobody else will be able to open it (World Report Silicon Valley, 28, October 2000)
Smartphones would not employ commercially viable fingerprint authentication until 2013. E-commerce was nascent; Amazons U.S retail market share was less than 1/10% Today it is over 5%. Cell phones were flip phones that ONLY made calls and sent texts - no pictures, no Internet, no e-commerce, no security.
According to Alan Niebel, 22 year Flash veteran and CEO of WebFeet Research, ?On the 10th anniversary of the iPhone, it's appropriate to credit the technologies that made smartphones possible. George Perlegos vision, hard work, and technical acumen contributed directly and substantially to the way we live today, and the way we will live in the future. The explosion of the Internet of Things is another example. Many of those ?things that are controlled using the Internet are based on self-programming Flash MCUs, one of George's patents. That allows them to sense, take action, communicate and take instructions via the Internet."
Today, Mr. Perlegos is thinking about the next 20 years of innovation in technology and medical research.
Photos accompanying this announcement are available at:
About George Perlegos. George Perlegos is trustee of the George and Angeliki Perlegos Charitable Trust, which provides grants to promote scientific advances toward the treatment and cure of progressive diseases such as IgA nephropathy and degenerative diseases (Parkinsons, ALS, MS, etc), and to promote research in technology directions. He is also Chairman of the Perlegos Capital Management LP investment group. He attended San Jose State University (BSE, 1972) and Stanford University (MSE, 1975). He resides in Fremont, CA.
Mr. Perlegos founded three successful semiconductor companies: SEEQ Technologies (1974), Chips and Technology (1984), and Atmel (1984). Atmel was acquired by Microchip (MCHP) in 2016.) Mr. Perlegos holds 16 patents for innovations in memory technology and device physics.
Additional information is available at www.georgeperlegos.com.
1 A Revolution in Progress: A History of Intel to Date, (Intel Corporation, 1984), p. 22.
CONTACT: Nancy B. Green The William Baldwin Group+1 650 856 firstname.lastname@example.orgJohn BryantGeorge and Angeliki Perlegos Charitable Trust+1 408 857 email@example.com