Subscribe to EDN

Blurring Lines Between 8- and 32-bit Microcontrollers?

February 22, 2010

… or is the 16-bit microcontroller family dead?

Over the years, the lines have blurred between the 8-, 16-, and 32-bit microcontroller families with designers confused to when they should select which line. In Robert Cravotta’s recent article entitled “Recognizing Technology’s Inflections” he says that

“… Freescale’s Flexis and Atmel’s AVR lines blur the line between 8- and 32-bits by sharing common IP (intellectual property) between the processor groups.”

Although this is very true, and I agree completely with Robert’s statement, I wanted to make clear that there are several clear distinctions to when a designer will select an 8-bit vs. a 32-bit. But the blur is when one thinks of the 16-bit microcontroller family – a technology that is slowly ‘dying on the vine’ since the 8- and 32-bits are moving into the 16-bit space.

Let’s start with the 8-bit microcontroller family. Many engineers use 8-bit microcontrollers for applications that have typically required 16-bit microcontroller performance. These applications require a microcontroller that has scalable system performance with reduced cost, smaller size, more flash memory and lower power. A practical application may be a soda vending machine (since we all drink soda). A soda machine, let’s call it a Coke machine, may need an 8-bit microcontroller, but with a large program requirement, and still require ultra-low power performance. The Coke machine’s task is to add up the total number of sodas dispensed on a daily basis and send the data to headquarters so the delivery person can restock the sodas accordingly. Many companies today deliver microcontrollers that may fulfill these historical 16-bit application requirements of – standby, sleep for lower power consumption and smaller footprint – but with an 8-bit microcontroller. This is a typical example of how the 16-bit market is being serviced by the high-performance 8-bit market.

There is typically no doubt in a designer’s mind when he needs a 32-bit microcontroller for his application. For instance, applications in the automotive, industrial, biometric, audio, communication and web server space have intensive data computing and typically require 32-bit microcontroller performance. These applications require extremely robust computing performance yet low power consumption is always a concern. In the past, designers may select a 16-bit microcontroller to suit these requirements but with the integrated functionality of a complex-deep sleep mode in the latest 32-bit microcontroller families, they are now able to select a 32-bit microcontroller rather than a 16-bit and use the additional performance of the 32-bit while still reaping the low-power performance and smaller footprint. In addition, the 32-bit microcontrollers also include more memory and interfaces such as high-speed USB for the system designer.

Although there is a blur between 8- and 32-bit microcontrollers due to the shared IP, I believe both families are moving into the 16-bit space. In short, I believe the 16-bit microcontroller is a shrinking market with the 8- and 32-bit encroaching in its market space.

 

Pete Jones, General Manager and Managing Director, Microcontroller Business Unit, Atmel

 

 

I encourage you to read all of the posts for the robust design series; maybe they will inspire you to share your observations. I would love to be able to consolidate different perspectives and lessons learned with regards to robust design practices here. I suspect there are some valuable lessons to be gleaned from comparing such stories.

 

To make following this series easier (especially as multiple series overlap each other), I am including the index below to previous posts, both for this and the guest post channels.

 

Guest posts:

2010, February 15: Wireless Everywhere and Programmable Designs

2010, February 8 : Wireless baseband inflection point – SDR as a technological breakthrough

2010, February 1 : Of Windows, Newton’s, iPad’s and 10GBASE-T

2010, January 28 : Technology inflections : digital signal processing

 

Posts made here

2010, February 15: Inflection Points : Wireless and Programmability

2010, February 8 : Inflection Points : Wireless and SDR

2010, February 1 : Inflection Points : timeline (networking)

2010, January 28 : Inflection Points : timeline

2010, January 26 : Inflection Points

 

Posted by Robert Cravotta on February 22, 2010 | Comments (7)

February 24, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
George Freeman commented:

Hello John. Yes, I'm the George Freeman that took charge of the specification you referred to. Shanghai is where it's happening. I'm getting married again to a very young lady in China. Let me know if you will be in China in April. I will set out an extra pair of chopsticks for you. I tried the LPC1100 with the LPCXpresso. It's a great combination! I've decided to also buy the LPC1300 LPCXpresso board for another project. It's amazing how inexpensive MCUs and embedded tools have become within my career. Stay well.


February 24, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
John H Titus commented:

Hi George, are you the George Freeman that worked on the IEEE low voltage specifications in the '90s? I heard you are living in China now. Can you let us know how the LPCXpresso worked for you? If you recommend it, I will definitely try to use the NXP tool and the M0 product on my next project.


February 23, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
Dorgish Pomotomo commented:

Bobby have you collected or have you seen any data for the code eff comparisons of the M0 core that you could direct me to? This core appears to be very promising for our new products (old parts using 8bit micro's) guess we should go and pick up a board!


February 23, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
Frank Landing commented:

Patrick Heath is Strategic Marketing Manager at Microchip Technology, a dominant player in 8- and 16-bit MCUs but with no success in the 32-bit world. Yes, that's the same Microchip who tried but failed to aquire Atmel in 2008.


February 22, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
Bobby Caracus commented:

I was using microchip for a project and switched to the M0 and found it way better for code eff. and performance not to mention power. I know George Freeman, we worked together in the past. I agree with him you won't be dissapointed with the LPC1100.


February 22, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
George Freeman commented:

Patrick Makes some good comments, but he sounds like a product manager for a company with a 16-bit MCUs to sell. Unless ARM and NXP are making stuff up, the Cortex M0 products seem to be extremely code and power efficient. Just purchased a LPCXpresso development board from DigiKey. Can't wait to try it.


February 22, 2010
In response to: Blurring Lines Between 8- and 32-bit Microcontrollers?
Patrick Heath commented:

Geez! This is just the sort of wishful thinking and spin control that you would expect from a manager of a company with no 16-bit products to sell! In reality, 8-bit products fall short in many uses, such as "green applications" like motor control and digital power conversion because they can't efficiently handle sensor data being fed to a 10-bit or 12-bit A/D converter. In these applications, high-speed control loop execution and high-resolution are required to eek out the top efficiency that customers demand. Truncating the data to 8-bits looses this advantage. Conversely, its a biased and inaccruate position to believe that 32-bit machines designed to move 32-bit data quickly will be better then 16-bit devices at the control applications that they are optimized to deal with by design. Padding 22 zeros on the front-end of every 10-bit A/D input provides absolutely no advantage. In these applications, 32-bit devices have lower performance and take more program and data space than 16-bit controllers. This is why the 16-bit controller market is expanding.

POST A COMMENT
Display Name
captcha

Before submitting this form, please type the characters displayed above. Note the letters are case sensitive:

Advertisement
Advertisement
Advertisement
About EDN   |   Site Map   |   Contact Us   |   Subscription   |   RSS
© 2012 UBM Electronics. All rights reserved.
Use of this Web site is subject to its Terms of Use | Privacy Policy

Please visit these other UBM Canon sites

UBM Canon | Design News | Test & Measurement World | Packaging Digest | EDN | Qmed | Pharmalive | Appliance Magazine | Plastics Today | Powder Bulk Solids | Canon Trade Shows