[Shootout-list] fannkuch (timer resolution; HZ=1000?)

[Isaac Gouy]

--- Jon Harrop <jon@ffconsultancy.com> wrote:
> On Friday 20 May 2005 22:08, Isaac Gouy wrote:
> > > As you can see, half of them are <0.1s and 82% are <1s. So half
> > > of the current benchmarks are actually just measuring startup
time
> > > for some languages (most notably Java).

> Which is precisely why the results I gave were for the largest values
> of N.

Sorry, I did misread your posting. I didn't notice that you stopped
saying "using only tests which ran for >0.1s or even >1s but there is
virtually no such data on the shootout", and started saying that 50% of
the tests ran for >0.1s and 20% ran for >1s. That's a lot for virtually
none.

***Note: ndata.csv includes both the current shootout and the old Doug
Bagley shootout.

For the current shootout there are 8 out of 28 tests <=0.1s

Of those, 4 are remnants from the old Doug Bagley.

heapsort, no-doubt will be replaced
object, has a replacement in progress
object-methods, no-doubt will be replaced
random, is used in so many tests that it will stay around

fannkuch has 3 programs <=0.1s (0.07, 0.08, 0.1)
mandelbrot has 5 programs <=0.1s (0.07, 0.08, 0.09, 0.1)
pidigits has 3 programs <=0.1s (0.03, 0.07)
threads has 5 programs <=0.1s (0.04, 0.05, 0.07, 0.08)

No doubt the max N for those 5 will be increased sometime.

Shockingly there are several things much more broken than this which we
need to fix.


		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Helps protect you from nasty viruses. 
http://promotions.yahoo.com/new_mail