Applied Numerical Analysis Using MATLAB

Rating:
Summary: Good for reference, but don't expect to learn from this book
Review: I used this book in a college level course, and let me tell you it is close to worthless. I gave it two stars because someone might be able to use this for some kind of whimpy reference book. Don't buy this expecting to learn much from it though. The first few chapters are ok, but later on it just glosses over many things you need in order to understand stuff.

Rating:
Summary: Don't expect to learn from this
Review: If you need a reference for various numerical methods then this book may be of use to you, but do not expect to learn anything from it. You will not be taught 'why' these methods work, or 'how' they came about--it simply lists a bunch of formulas with a few meager attempts at derivations here and there. If you want to actually understand this stuff you'll have to go elsewhere. If you just want a reference...I would still go elsewhere.

Rating:
Summary: Good Reference Book
Review: This book is exactly as the title says. There are a lot of great sources on applied mathematics and it covers a lot of good material ranging from iterative methods to PDE's. Matlab is the software of choice and the book does a good job of making use of Matlab. The detraction to this book is the lack of rigor. The book operates under the misconceptiong that applied math doesn't have to be very analytical. As a reference book it is good resource, but as a textbook a little bit more analysis needs to be put in. This is a nice fresh idea for what Numerical Analysis should be, but could use a bit more beefing up in the analysis side. Hopefully this will come with later editions.

Rating:
Summary: Fantastic! Stupendous!
Review: Whether you use Matlab or not, if you need information about numerical methods, this is the book for you. It makes the standard, Press, et al, "Numerical Methods In.." look like a first-grade primer by comparison.

I'm writing my own book on numerical algorithms for embedded systems, so I know whereof I speak. _GOOD_ books that are both readable by ordinary mortals, and usable for serious computing, can be counted on the fingers of one hand. Most are either too pedantic and obtuse, or too simple and shallow, or directed at a small subtopic of the larger area of numerical methods. Somehow, through a combination of simple, no-snow examples, lots of exercises, and Matlab programs, author Faucett manages to include virtually every aspect of numerical methods. Not only that, but she includes them in depth, in a way that's easy to follow and exercises that drive every point home. In my travels, I've seen a lot of references to such things as Runge-Kutta and Adams-Moulton methods, solving stiff ODE's, solving for eigenvalues and eigenvectors, and Householder and QR transformations, to name a few. It's rare to find easily intelligible explanations, with derivations, for any _ONE_ of them. To find them all in one book is astonishing.

Clearly, by the title, you get the impression that having Matlab will make this book more valuable to you, and it will. However, don't get the idea that the book is only for users of Matlab. Whether you have it or not, you will not find another book so loaded with gems of knowledge about numerical methods.