Re: comparing algorithm perf.

On Mar 13, 6:42 am, Thierry MARTIN <thierry-mar...@xxxxxxxxxxx> wrote:

2. if you launch time twice with the same program and same data, you get
two different results, which lets me think it depends on the load of
your machin.

I'm afraid this is essentially unavoidable. You must make sure the
machine load is the same in both cases.

For example, if there are no other processes ready-to-run, this
process will always find the caches warm. This will result in a
different run time than when there is a lot of context switching going
on.

There are many other dependencies. You are entering an area that is
something of a black art.

DS
.