8- and 16-bit processors: state of the art

Micro Minis
By Jack G. Ganssle, Embedded.com
March 13, 2003 (1:01 p.m. EST)

With chip companies packing more peripherals and functionality into 8- and 16-bit processors, it's almost impossible to keep track of what's out there. But don't worry; we got someone to do the work for you. Here's Jack Ganssle on the current state of the art.

Rumors of the death of 8- and 16-bit processors have been greatly exaggerated. Over seventy-five percent of all processors sold in 2002 were 8 or 16 bit. What follows is a look at what's new with these small parts. According to November 2002 Embedded Market Survey, 43% of the magazine's readers use Microchip's PICMicro parts, 55% use 8051/52/251/AVR, 36% go for Motorola's 68XX family, and Zilog's Z8/Z80/Z180 devices account for about 15%. (These numbers total more than 100% as some folks used more than one processor in that year.) In the 16-bit world, 41% use 8086/186/96/196 devices and 21% employ th e 68HC12/16.

Most of these architectures are more than 20 years old. New designs have a hard time competing, probably due to the wealth of support for older CPUs and the large pool of developers who are well versed at using them.

Though 8- and 16-bit processors were once the computing engine found in all smart products, today they are used in the realm of deeply embedded applications, ones that typically have a lot of I/O. Peripherals are important—just look at the astonishing number of variants for some devices. Some 200 different 8051 variants offer almost any mix of memory and peripherals you can imagine.

Though timers, parallel, and serial I/O, and other traditional functions remain the staple of peripherals, many devices now offer sophisticated special-purpose built-ins like LCD drivers, plum controllers, and fast A/D converters.

Communications are key to many devices, even very simple ones. On-chip I2C, not found on 32-bit chips, is common in the 8-/16-bit world. CAN, which is sti ll popular in Europe, is also available. Even USB controllers exist on-chip, as in Hitachi's H8S/2215.

Internet support takes the form of on-board network controllers on a few parts, notably Zilog's eZ80F91. Third-party vendors provide protocol stacks to connect even the smallest processors to the Internet.

Most intriguing is the move to "virtual" peripherals. Triscend's E5 is an 8031 surrounded by an FPGA; the designer configures peripherals using Triscend's tools. Need four timers and three serial ports? Customize the peripheral mix to your requirements.

Cypress's 8-bit Programmable System-on-Chip (PSoC) is similar, but allows the designer to redefine the peripheral mix as the program runs. Like a shapeshifter, the device adapts dynamically to changing requirements. Analog blocks let you configure devices that are traditionally found in external circuitry, such as programmable gain amplifiers and filters.

Memory
Memory requirem ents are skyrocketing in all computing applications. Microcontrollers with just hundreds of bytes of ROM are still common and quite useful for simple applications. More complex ones are served by the growing capabilities of new CPUs. Zilog's eZ80, for instance, is an 8-bit processor that addresses up to 16MB of memory.

Even the venerable Z8, the heart of most TV remote controls, is available with 64KB of on-chip flash and another 4KB of RAM.

Microcontrollers are finally moving from one-time programmable (an EPROM with no erasure window) to flash. Though flash remains a more expensive technology, being able to program a device after it's soldered onto the circuit board reduces manufacturing costs. As program sizes grow so do bug counts; the ability to update code without disassembling a product gives flash a powerful appeal.

When we hear the word "security," we immediately think of a new federal department—or of network vulnerabilities. Yet, in the embedded world, many of us need to protec t our intellectual property and the dollars spent creating a product's code base. Some microcontrollers address this need by making it impossible, or at least extremely expensive, to read code stored in flash. An example is Dallas Semiconductor's DS2252, an 8051-compatible device that stores program code encrypted with a 64-bit key. The SDI pin, short for self destruct input (no kidding!), will, if asserted, delete the key. A circuit that detects tampering can drive SDI, leaving the product brain-dead but reverse engineering resistant.

Squeezing size and power
Twenty years ago, I saw a Motorola 68HC05 running off of two lemons wired in series. Since then batteries haven't improved much, but portable computing has exploded. Most processors now offer low-power modes. The February 2003 issue of Embedded Systems Programming contained a nice article by Nathan Tennies ("Software Matters for Power Consumpti on," p. 20) about using these modes effectively.

Five-volt parts are still common, but don't lend themselves well to devices running from a pair of AA cells. CPUs today run from a wide range of voltages: 3.5, 3.3, 3.0, 2.4, and even 1.8. Lower voltage reduces the power (measured in amp-hours) sucked from the battery.

A plethora of low-power modes are available. In sleep mode, Epson's S1C88 family uses 0.3µA (typical). Awake, it needs only 14µA at 32kHz and 2µA at 4MHz. Eight-bit H8 devices use a barely measurable 0.1µA when sleeping and just over 1µA at 32kHz.

Power limitations are part of building small systems. It's hard to believe that a few decades ago a small computer weighed tons. Today's surface mount technology puts processors into packages whose pins appear as fine as spider webs.

Infineon's C163-L 16-bit microcontroller comes in a 100-pin TQFP configuration—typical of small form-factor devices—measuring just 14mm on a side, with 0.5mm lead pitch (leads too small to see if you wear bifocals). At 1.4mm thick, it's useable in super-slim applications like PCMCIA cards. The Rabbit 3000 comes in a variety of packages, including a 10mm2 TFBGA just 1.2mm thick.

Signal processing
Analog will never go away, but designers seem intent on pushing the digital part of a circuit as close to a sensor as they can manage. This is partly possible because we can create digital filters in software, a task that burns enormous amounts of processing power.

TI pioneered DSPs. Their TMS3201x chips are 16-bit DSP devices used in a huge range products. But, in a twist, Microchip Technology is now making Digital Signal Controllers, a family of 16-bit DSP microcontrollers. All come with on-chip flash, RAM, A/D converters, pulse width modulation, and even EEPROM. They contain from 16KB to almost 200KB of flash.

Other devices, like TI's MSP430, are hermaphrodites, traditional CISC CPUs with the addition of the DSP's multiply an d accumulate (MAC) instruction. A MAC instruction performs both a multiplication and an addition in a single cycle, which speeds many signal processing algorithms. To get an idea what's out there, see Embedded.com's special report on DSPs by Don Morgan.

The right fit
The vast proliferation of processors makes it likely one exactly suited for your application is available. And if not—well, take advantage of the technologies that let you design your own peripheral mix.

Jack G. Ganssle is a lecturer and consultant on embedded development issues. He conducts seminars on embedded systems and helps companies with their embedded challenges. Contact him at jack@ganssle.com.

8- and 16-Bit microcontrollers and embedded microprocessors
The following list of 8- and 16-bit embedded microprocesso r and microcontroller vendors can also be found in the Embedded.com Buyer's Guide at www.embedded.com/bg along with vendor-supplied product specifications. The online Buyer's Guide enables you to run detailed side-by-side comparisons of their products. You can sort by data bus width and architecture.