
Nisha
Im a quite new boost user and I wonder, why C++ Program is so slow.
I wrote a program, and I have the same program in Matlab, I expected that my C++ version is much faster, but its not.
I use matrices, vectors, matrix- and vector proxies. Do some solving and compute calculations with prod, element_div, etc..
Is boost in general slower than matlab, should I use something else?
Id be real happy about any help.
This is unfortunately a very broad question, and the way it is worded will make it difficult to answer. In general, a compiled language such as C++ will be faster than an interpreted language such as Matlab. However, as with all computing languages, it is very possible to use C++ (and/or Boost) improperly or inefficiently. Perhaps, the best way to obtain help would be to post a small code example that exhibits the performance problem you are seeing. It would be good to see the Matlab and C++ equivalents that you are trying to compare. With C++, compilation settings (compiler flags and defines) make a significant difference. For instance, many additional checks are done on the running code when it is compiled in debug mode. The checks are great for making sure that the code works properly, but they should be disabled for release builds in order to obtain optimal runtime performance. It is generally a good idea to compile with NDEBUG defined when building in release mode. Additionally, it is always good to get familiar with profiling tools. These tools make it easier to spot performance bottlenecks in your code. Hope This Helps, Justin