Everything I have read recommends using stl vectors instead of c style arrays. But when a ran some tests of 100 million accesses of arrays and vectors, I found that the array part finished in 1 second and the vector part took 6 seconds. This was using the [] notation with the vector. It took even longer with the at() notation. Is the difference in performance this large? If this is the case, presumably vectors are not the way to go if the program requires many accesses to an array?
Thanks
Mark
Did you turn on optimizations?
I dont know which optimizations to turn on. I am using the g++ compiler on linux.
Last edited by markucd; 04-19-2006 at 10:04 AM.
Can you show your code?
Are you timing only accesses, or are you timing filling of the vector and array and accessing their elements? Are you comparing the vector to a local array, or a dynamically allocated one?
In most cases, any performance differences between the two won't be noticable if you are using vector properly.
Compile with the -O2 flag to turn on optimisations.
dwk
Seek and ye shall find. quaere et invenies.
"Simplicity does not precede complexity, but follows it." -- Alan Perlis
"Testing can only prove the presence of bugs, not their absence." -- Edsger Dijkstra
"The only real mistake is the one from which we learn nothing." -- John Powell
Other boards: DaniWeb, TPS
Unofficial Wiki FAQ: cpwiki.sf.net
My website: http://dwks.theprogrammingsite.com/
Projects: codeform, xuni, atlantis, nort, etc.
I am using statically allocated arrays, and I am only timing the array access part. The O2 optimization flag worked.
Thanks
Mark
The soon-to-be standard alternative to static arrays is std::tr1::array, which is available from boost right now (perhaps just as boost::array). It might be more appropriate than vector if all you need is a static array.
