Re: [Hampshire] [Tech] The 'speed of a language'

>> That's one of the problems with performance optimisation. You cannot
>> optimise everything (for a large software project you never have time)
>> so
>> you have to work on the areas you think are important / used the most.
>> Of
>> course if you get the wrong areas, then it doesn't help ;).
>
> This is true but it's a completely different type of optimisation, one
> that cannot be done automatically. Compilers can fully perform
> instruction-level optimisation, but they won't rewrite an O(n²) algorithm
> to an O(n log n) algorithm, which is usually the kind of optimisation
> programmers usually look for when they are coding.

Indeed, I probably didn't make it clear in my original post that I was
going off on a tangent about needing to understand what you are optimising
;)

Of course there are arguments in both directions about this is faster than
that, and I've seen various people comparing speed both ways with tests
that show things are faster in C vs Java, and other test showing code in
Java is faster than C.

But I would agree with original post that the "speed of a language" is
largely irrelevant for all but a very small sub-set of the software
written today.

The biggest cost of a software project is service and maintenance of the
software that has been written, and mitigating the impact mistakes. The
language and environment that you choose to write in can massively impact
this and is probably the largest factor is choosing a language.

Personally if I was running a software house, I probably wouldn't start
any new C or C++ software projects because those skills are far to scarce
and far too expensive. (unless of course you had very very good reasons
why it had to be in one of those languages).

For most general use software the speed of using something like Java or
.Net or similar, which is now taught as the main language to graduates
makes little or no difference, but will greatly impact your staff costs,
and of course make your maintenance costs cheaper too.