C++ Multithreading Cookbook in 2014 (publication year), that seems quite interesting, with all the new stuff from the current C++ standard. Is it what the book delivers?

Content and opinions

Unfortunately, when you read the table of contents, there are already orange flags. Chapter 6 is about threads in the .NET framework. What?? This is a book on C++ multithreading, not Windows specific things, right?

OK, let’s start with the beginning. Chapter 1 is a poor presentation of C++. The author says that he will be using Hungarian notation. Actually even Microsoft says do not use it. It predates modern C++, so stop it. Period. I won’t talk about the issues with misunderstanding of a .lib in Windows or what precompiled headers are.

The second chapter is actually interesting. Processes and threads are quite complex beasts, and it is not always properly explained.

The bad gets worse with the third chapter, which is supposed to be about thread. Don’t forget this is a C++ book. C++11 brought native thread management, locks, mutex… But nothing of the sort is here. Only Windows specific C threads (not even C++). Not talking about specific unexplained pragmas. Chapter 4 addresses processes, which is actually not in the C++11 standard. This is where the book could have shined, but no, it has to talk about Message Passing Interface, but this has nothing to do with Message PAssing Interface, which is an official C/Fortran/C++ standard.

By then, I’m fed up with the book, although chapter 7 does pose some good code practices and warnings about concurrency. But then, it goes bad again, even talking about OpenMP (although there is a C++ “version”, no real HPC code actually uses this unusable interface in any properly designed code).

Conclusion

In the end, the book title has nothing to do with the content. It may have been interesting 10 years ago with a title like “Windows C threads and processes”, but it is definitely not worth it with C++11.