I'm not sure about the glowing reviews here, but I fail to see how this book can be much help to current embedded designers? While the copyright is 2004, most of the book reads as if it was written in 1994 or even earlier. The book talks a lot about discrete logic, something that's rarely (or barely) used in modern embedded designs. Likewise, the author largely assumes external memory, which is also increasingly rare.



Most people, today, use SPI, I2C and/or CAN serial interfaces within (and even between) embedded devices and peripherals. The book doesn't provide useful coverage of any of these (or even mention some of them). It's as if the author's last real world embedded development work was with the 8051 a few decades ago? It's the only processor he goes into any detail on and not even the current System on a Chip versions.



The book does contain a useful primer to basic electronics and microcontroller theory. But like the rest of the book, much of it is out of date relatively to current embedded design and the same material is covered much better in other books.



The only reason to POSSIBLY buy this book, in my opinion, is the chapter on worst case timing and loading analysis. But even that section is showing its age. Signal levels, PC board design, bus/signal termination, impedance matching techniques, etc. are largely done differently today than what the author suggests. For example, do you know anyone using TTL discrete logic in their current embedded designs? Apparently both Ken Arnold and Newnes/Elsivier think someone does? You can have my copy back Amazon.