Customer Reviews

Windows NT Device Driver Book, The: A Guide for Programmers

5 star:		(9)
4 star:		(6)
3 star:		(4)
2 star:		(1)
1 star:		(1)

 

 

I have read all, but one of the Windows 2000/NT device driver books and found this one to be the best for beginners, by far. It is very well organized in a top-down approach. I though the concepts were well presented and easy to understand. After readin two other books on NT device drivers, this one finally explained things clearly for me.

The book is also very hands-on. It describes an example device drive in full source and develops it through each chapter. You can actually build the driver as you go and the driver is developed as you would develop your own driver.

I do want to make it clear that this book is not a definitive refence. It does conatin errors, but the approach for teaching Windows driver development is its strength. You'll want to get OSR's and Wlater Oney's Windows driver development books to get a good foundation.

This book is also not wordy and quick-reading. I would suggest is, if you are trying to grasp NT device drivers.

I am pasting this review that i had originally posted in 2001 at other websites.

This is one of the best books i have read. I have several years of experience in C/C++ and some experience in windows programming however I am a 'absolute beginner' with Windows device drivers and kernel mode programming and i was able to grasp almost everything the book spoke about with relative ease and with NO confusions. I would recommend this book to anyone who has absolutely no background in device driver programming but wants to make a career doing just that. I must admit that the books assumes that you are versed with C/C++ and some understanding of programming paradigm for windows, but it is pretty obvious since this is NOT a book to teach a programming language nor it is aimed at teaching regular windows programming. It is a complete 'NO NONSENSE' book that deals with topics right upto the point. The contents of the book flow gracefully explaining each and every step with precise detail. The author seems to have made the best possible effort to explain the basics before jumping directly into details. And that does help an absolute beginner. This book does NOT cover details about device specific drivers but it does help you reach a point where you are confident that 'you will understand' whatever you research on your own. A very good book... have seen very few of these types lately.

For neophytes, writing a Windows NT device driver is often an extremely frustrating experience. I first wrestled with the DDK when I was asked to write a device driver for NT (version 3.5). The DDK documentation was wrong in several places. Thankfully, it has improved, but still leaves a lot to be desired.

Years later, Art Baker was the first to produce a readable driver development. It was, and still is, an excellent primer for those interested in developing device drivers for the NT operating system.

The information is still relevant for those of us who must maintain legacy NT drivers.

Excellent job revising the NT Device Driver book for Windows 2000. My group used the previous version, which had numerous errors and slow reading. I have also read all other books on this subject matter. It is obvious Mr. Lozano understands his audience. The book demonstrates expertise on the author's part. I especially enjoyed building an actual driver as I read through the book. Great presentation!

I look forward to more of Mr. Lozano's books.

While a relatively understandable introduction to the topic, it cannot serve as a reliable reference given the number of errors contained in it. Moreover, the author is working on a second edition, due at the end of the year; I can't recommend spending money on the current form.

At the time it was written, it was the only thing available and valuable for that alone; but there are better alternatives today. Get the Anthony Mason/Peter Viscarola book instead.

This book is a more concise, better organized version of the DDK documentation, minus the reference material. It does a good job of painting a picture of driver architecture and gets you familier with the terminology involved. However, it does not provide much insight or real-world tips, and in fact the regurgitation of Microsoft definitions and propoganda gets a little tiresome. This book does not go into much detail and is not a reference book. After you've read most of this book you can easily rely on the DDK documentation and never flip through these pages again.

A couple specific gripes: 1) If the reader has never written a driver before, they've probably never worked in kernel mode before, so more general information on kernel-mode programming issues would have been appreciated. 2) Though this book does not go into much detail, the forward did promise a chapter on USB and IEEE 1394 available on the book's website. I was not able to find any such chapter on the website, and haven't received a response to my email requesting the information (to be fair I've only given him a couple days).

I write NT drivers and kernel extensions for a living so I have already been baptised so to speak. Even so, because Microsoft has been so tight lipped about NT's internals, there is much I have to learn. And, indeed I did by reading Art's book.

I just finished Art's book and I must say that I am impressed with his choice of subjects, his clear style and his light humor throughout. Moreover, his examples teach in fact what his narative imaginatively presents. If you are buying this book to teach you about writing a certain kind of driver, say NDIS 4.0, then don't. It's very short on specifics of that kind. But, if you want to learn the "basics" about writing NT drivers, this will get it done. The gift that Art brings through his book is more than dry facts about NT driver lore and technique. Unlike the DDK Guide, Art has placed the process into perspective by presenting the material in a logically progressive manner. Further, we benefit from Art's long experience at teaching the subject as well as, I suspect, his hobknobbing with the MS development folks and having had many a question answered. I would not be surprised to learn that he has "even" seen NT's sources since he formerly worked for DEC with their intimate association with Uncle Bill's Belleview Works. I know that I have come to understand some things that, frankly, I thought were in the realm of accepting on faith.

To sum up, though experienced, I learned alot, and, I am convinced I will continue to do so every time I pick it up in the future. I do have one caution to offer. Art's book is no substitute for the DDK Guide -- I think after reading Art's book, you should then skim through the Guide. It's in the neighborhood of 1000 screen pages of if you want to print it out, it's much shorter. Since the Guide was written as an online reference, it tends to repeat a lot of information. This can be tedious reading but, hey, that's how I learned -- I only wish that I had had Art's book as a primer

I just finished this book. Great book. I almost did not buy the book based on the review by Tomer Cohen. He stated that the book was outdated. Actually, it is completely updated for Windows 2000. I took the advice of a coworker and bought it - I'm glad I did. The book has great examples and was easy for me to follow. I definitely give it a 5+.

The book is really useful if you are programming in a "conventional" single-processor environment and do not wish to experiment too much. As pointed out by others already, there are some GLARING errors. So if the thing does not work and the book says it should, check the DDK!

The concepts of this book are still useful for
Vista and XP.

It also explains the fundamentals well.

I read newer books first, and they were too confusing.

Oney's book gives too much information, and jumps from
topic to topic.

Orwick's book is too high level, and not enough detail.