Re: Ruby?

[David Jeske <jeske@...>]

On Tue, Sep 26, 2000 at 11:12:07AM +0100, John Batty wrote:
> I have seen quite a few performance comparisions of Lua vs other scripting
> languages, but has anyone done corresponding C/C++ benchmarks?

I did C# benchmarks of the ScriptPerf benchmarks you can find at:

http://www.chat.net/~jeske/Projects/ScriptPerf/

I used C# because it's garbage collection and easy access to complex
data-types made it easy to port ScriptPerf. It performs as you would
expect, the simple code tests such as looping, function calls,
returns, etc. are all incredibly fast (much faster than any of the
script languages), while hashtable dominated tests perform about the
same. I espect straight C/C++ would perform about the same.

I will put the results on my ScriptPerf page today or tommorow (they
are on my windows machine at home, which I can't get to remotely).

> The current (scripting language) benchmarks are useful for answering the
> question: "I need a scripting language - which one should I use?".
> 
> However, once you have chosen the scripting language(obviously Lua!), then
> you start to ask the question "How much is the perfomance going to change if
> I move this C++ function into Lua?" - so you can work out where to draw the
> line between your C++ and scripting code.  This is obviously going to depend
> on the actual code involved.  However, if anyone is going to run any new
> benchmarks, I'd be interested in some C/C++ timings too!

-- 
David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net