J.J. Sakurai book on Q.Mech 
*

*I just want to ask how recommended is the book on quantum mechanics by J.J. Sakurai. Is it any good as an introductory text? 

*And are there better suggestions (substitutes)?
 A: IMHO it is an excellent book, particularly the first chapters. It goes immediately to the point and forces you to get used right away to bra's and ket's and so on. Many years have passed since I used the book (and now there is even a new edition which I have not looked at) but back when I was an undergraduate student it was my favorite book of choice.
On the other hand I don't think of Sakurai's as an introductory text. The first and second chapter could work a bit like that but only for the brightest students. For starters (studying Physics) I would rather suggest a book like Griffith's or from a very different perspective, the recent book by Schumacher "Quantum Processes, Systems and Information".
A: Sakurai is an odd book when it comes to this issue.  On the one hand, he develops the bra, ket, and operator formalism from zero in a way that anyone with some knowledge of linear algebra should have no difficulty following.  In fact my first semester of undergrad quantum was based entirely on Sakurai.  Solving problems with Pauli matrices, calculating expectations and dynamic equations is all covered more or less from zero.  But then he throws in statements like the one on p.112 (revised edition):

It is assumed that the reader of this book has some experience in
  solving the time-dependent and time-independent wave equations. He or
  she should be familiar with the time evolution of a Gaussian wave
  packet in a force-free region; should be able to solve one-dimensional
  reflection problems involving a rectangular potential barrier, and the
  like; should have seen derived some simple solutions of the
  time-independent wave equation—a particle in a box, a particle in a
  square well, the simple harmonic oscillator, the hydrogen atom, and so
  on—and should also be familiar with some general properties of the
  energy eigenfunctions and energy eigenvalues, such as A) the fact that
  the energy levels exhibit a discrete or continuous spectrum depending
  on whether or not (2.4.12) is satisfied and B) the property that the
  energy eigenfunction in one dimension is sinusoidal or damped
  depending on whether E - V(x') is positive or negative. In this book
  we will not cover these topics. A brief summary of elementary
  solutions to Schrodinger's equations is presented in Appendix A.

That's just after spending 10 pages solving the simple harmonic oscillator!  Anecdotal experience (my own and other grad sutdents) indicates that Sakurai is currently the reigning undergrad textbook for a serious first course in quantum (his ket/bra/operator-first approach is a great contribution, didactically) but invariably needs to be supplemented with other materials to cover basic potential well type problems and a lot of the background to how it fits in with electrodynamics and classical mechanics.
It's clear that Sakurai intended his book to be read by graduate students who had to suffer through a less modern wave-function style derivation as undergrads and can now really appreciate how neatly everything fits together when you use bras and kets almost exclusively.  The problem is that he did too good a job at proving his point--starting out with this formalism makes a lot of intuitive sense and as a result the book is almost readable by a complete beginner.  And that makes it really annoying.  IMHO, they should put out a new edition that fills in all the gaps and actually turn it in to an introductory-intermediate book.  I think 50-70 pages would effectively cover all the gaps.
My advice is to use it (all your colleagues will have used it) but to supplement it with some of the other suggested books: Messiah, L&L, etc.
A: I remember finding it totally dense and confusing and unreadable as a young undergrad. When I reread it a few years later, I found it totally clear, and I could not (and still cannot) figure out why I had found it confusing before. It's strange!
Regardless, I suggest reading Griffiths, then reading it again and again, it's wonderful. :-)
