Most Helpful Customer Reviews
|
|
7 of 8 people found the following review helpful:
An excellent manual for Embedded Systems., 1 Oct 2001
By A Customer
This is a great book, It takes you from the basic principles of embedded systems through to running multiple processors whilst using a variety of available 8051 processors. The book takes an in-depth look at time triggered architectures, something I haven't yet found anywhere else.There are extensive examples showing the reader how to use the various schedulers, protocols and applications that are considered. Not only are there plenty of code examples, but also suggested circuits for each application. These are easily merged, adjusted and integrated making the book extremely useful when designing embedded applications. The book maintains a consistent layout throughout with the use of patterns. The patterns create such a structure that the book can either be used for dipping in and out of as a reference guide, or for extensive reading from start to finish. The included CD saves the reader having to type in the examples, whilst the applications included are good evaluation versions that allow the reader to test out their ideas before deciding what to buy. The wide variety of topics covered, the application of examples to real-life situations, and the consistency of layout is what makes it, for me, such a useful book that is worth having on any desk, or in any library.
|
|
|
Hang on a minute ..., 26 Jul 2006
As author of this book, I'd like to make two comments.
First, this is a book about TIME-TRIGGERED embedded systems (I think the cover makes this fairly clear ...). There are other ways of developing embedded systems but - many people would argue - time triggered approaches are a good solution, particular if you need to have predictable system behaviour (e.g. medical systems, automotive systems, aerospace ...). If reliability isn't an issue for the type of embedded systems you wish to build, then buy another book.
Second, the book describes a flexible, time-triggered co-operative (TTC) scheduler. This scheduler is then used in most of the examples. In my experience, users have found this to be a good introduction to TTC designs. However, if you don't want to use this solution, this book also discusses alternatives to this scheduler (and a related book - Embedded C - discusses the use of this very simple scheduler in more detail).
If you are considering the purchase of this book, I hope that you find this review useful.
(PS: I am forced to award a star rating with my review).
|
|
|
A student reference, BUT, 11 Jun 2006
Having read the 'gang of four' book I was delighted as an embedded developer to find a 'patterns' book specifically aimed at embedded engineers. Having now read the book it is, I'm afraid, a big disappointment.
The book starts with a nice description of the 8051 processor, and later includes some good information on different interfacing techniques. These sections of the book form an an excellent reference, especially for the novice engineer.
However, the ideas about software structure that are forcefully presented and repeated throughout the whole book seem one sided and misguided. Of coarse the techniques described have their place in certain applications - where appropriate - but other techniques are rubbished for no good reason.
The central theme of the book is that of a cooperative scheduler approach to software design. The scheduler presented however is flawed in its complexity (although very simple, it is much more complex than the funtionality warrents) and its usefulness restricted to a small number of trivial application designs. The author states that this type of cooperative scheduler is too often ignored by the engineering community press in favour of preemptive schedulers. He goes on to suggest that this is for commercial reasons, nobody wants anybody else to know how easy it can be as less money will exchange hands. I however have a different theory. There is not much written as there is not much to write about, although this author manages many hundreds of pages.
I would further suggest that the scheduler initially presented, with its massive usage restrictions, would better off being replaced with a single switch statement (with each 'case' representing a relevant time and simply calling the correct function for that time). This would be simpler, faster, more understandable, more easily visualized, easier to setup, easier to maintain, would not necessitate the user of function pointers, and not loose any functionality.
Interestingly, one of the people thanked for input during the writting of the book seems to be of similar opinion to me.
In summary then, as a novice reference I think this book is very good and recommended, but don't take the software ideas presented as being anything other than a very slanted opinion.
|
|
|
|
(2)
£31.34
See all Product Description
(3 customer reviews)
(
