Talk:Optimizing C++

Motivation
Here I would like to explain why I felt there was a need for a new book about C++ optimization.

I have read many books, articles, and blogs about C++ optimization, but most had one or more of the following shortcomings:
 * Too unhelpful. Many advices were like "avoid this feature", without specifying exactly when such a feature was avoidable, and, mainly, without specifying what could replace such a feature.
 * Too low level. Many advices were actually advices for a compiler writer, as they delved into compiler optimization technology, that are of little interest to a typical C++ programmer.
 * Too high level. Many advices were about computational complexity, just classifying algorithms according their asymptotic time, with big-O notation.
 * Too compiler dependent. Many advices were compiler specific.
 * Too operating system dependent. Many advices were operating system specific.
 * Too hardware dependent. Many advices were hardware specific.
 * Too little distinction between design-time and optimization-time. Few advices specified if they were to applied when the application was designed and coded, or at a specific optimization stage.

--Carlo.milanesi (talk) 19:34, 1 June 2008 (UTC)

How can an book be too low level and too high level at the same time?

--85.241.101.78 (discuss) 13:10, 18 September 2012 (UTC)

I think the OP's point is that all other books suffer from at least one of those problems, but not necessarily all of them at once. --Fishpi (discuss • contribs) 22:13, 19 September 2012 (UTC)