Re: Shootout (sum-file)

Bartc wrote:
> Suppose the C startup was 0ms, and that of Java was 50ms. 

You're missing the point: in the case shown, C was faster than Java.
It doesn't matter what aspect of Java caused that, it was simply slower. 
So for this specific test case of a 6kB file, Java is slower.

And supposing you

Supposing you ran the C version on an 8080 and the Java version on a Sun 
M8000? Supposing you excluded from your time-count all operations whose 
opcode has a vowel in? Supposing you ran the C version while running a 
huge Fortran numerical simulation with lots of disk IO in another thread?

> If someone is really interested in how much faster Java is than C, I
would 
> not trust your tests. Because with timings at these level, the figures
are 
> unreliable.

As Richard said, can anyone here spell the action of flattening 
garmentry with a hot metal object?

-- 
Mark McIntyre

CLC FAQ <http://c-faq.com/>
CLC readme: <http://www.ungerhu.com/jxh/clc.welcome.txt>