I think the speed rather depends on the garbage collection and the Rebol
version
The new 2.2.0.3.1 is around 4 secs slower than 2.1.1.3.1.
Running on a Celeron 333 and Windows 2000 RCII

5 Consective runs in freshly started REBOL 2.1.1.3.1
start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:22
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:40
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:44
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:22
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:24

5 Consective runs in freshly started REBOL 2.2.0.3.1
 start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:26
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:48
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:50
>> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]] now/time -
start
== 0:00:26
>>

Cheers

Allen K


----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, November 10, 1999 3:43 PM
Subject: [REBOL] [REBOL] How fast is REBOL? Re:(2)


> 35 seconds on my AMDK6 2D 300MHz running RedHat 6.0
>
> Jim
>
> > > Just doing some tests of REBOL 2.2 floating point speed.  On a
> > 450 MHz P2
> > > running Win98, at the console I get:
> > >
> > > >> start: now/time a: 9.0 b: 0.5 loop 1000 [loop 10000 [c: a ** b]]
> > > now/time - start
> > > == 0:00:20
> > >
> > > The overhead for the empty loop is about 1 second.  I am interested in
> > what
> > > others get with different platforms and processors, or even functionally
> > > identical loop in other scripting languages.  On my machine, REBOL is
> > > slightly faster than VBScript for the floating point, but the VBScript
> > loop
> > > overhead is about 6 seconds.
> > >
> > > Cheers
> > >
> > > Larry
> > >
>
>

Reply via email to