1. loops & perfomace
This is a multi-part message in MIME format.
------=_NextPart_000_0005_01C0CFDA.F95FD000
charset="iso-8859-2"
Hi,
i've got an performace question. We've got a loop
for i=3D1 to length(s) do
...
end for
is length(s) evaluated every time loop is done? So would be in Euphoria =
faster
l=3D length(s)
for i=3D1 to l do
...
end for
as i read in one C book?
------=_NextPart_000_0005_01C0CFDA.F95FD000
Content-Type: text/html;
charset="iso-8859-2"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-2" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2614.3500" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3D"Arial CE" size=3D2>Hi,</FONT></DIV>
<DIV><FONT face=3D"Arial CE" size=3D2>i've got an performace question. =
We've got a=20
loop</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3DCourier size=3D2>for i=3D1 to length(s) =
do</FONT></DIV>
<DIV><FONT face=3DCourier=20
size=3D2> ...</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2>end for</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3D"Arial CE" size=3D2>is length(s) evaluated every time =
loop is=20
done? So would be in Euphoria faster</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3DCourier size=3D2>l=3D length(s)</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2>for i=3D1 to l do</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2> ...</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2>end for</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3D"Arial CE" size=3D2>as i read in one C =
book?</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3D"Arial CE"=20
------=_NextPart_000_0005_01C0CFDA.F95FD000--
2. Re: loops & perfomace
I don't think it makes a difference. Euphoria would appear to be executing
the for loop the 'smart' way. I did a quick test with this code--
sequence s
s=repeat(32,50000000)
atom a,t0,loop_overhead
integer lengthofs
t0=time()
lengthofs=length(s)
for c=1 to lengthofs do
a=power(2,20)
end for
loop_overhead=time()-t0
puts(1,sprintf("%d seconds",loop_overhead))
...and then tried it, replacing this line:
for c=1 to lengthofs do
with this one:
for c=1 to length(s) do
and it ran at the same speed.
This code was just lifted from the docs located in
C:\EUPHORIA\HTML\LIB_S_T.HTM, and slightly modified...
--Ted
--On Saturday, April 28, 2001 12:01 PM +0200 martin.stachon at worldonline.cz
wrote:
>
>
>
> Need a better decision-making tool to research, evaluate and
> select software for enterprise applications or manufacturing
> point solutions? Masg.com is where you'll find help.
>
>
> Hi,
> i've got an performace question. We've got a loop
>
> for i=1 to length(s) do
> ...
> end for
>
> is length(s) evaluated every time loop is done? So would be in Euphoria
> faster
>
> l= length(s)
> for i=1 to l do
> ...
> end for
>
> as i read in one C book?
>
>
>
3. Re: loops & perfomace
With the quick test I did below, the times seem to say that length() is
faster than integer! But only one tenth of a second over 67 million
iterations. So I would worry about it.
---------------
atom e
integer aa
sequence ss, x
ss = "jhsdkjahsh akshdkjahsdkjaskjdaskhjdakjsdhkjashdkjahskdjhaksjdhkasdh"
x={}
aa = length(ss)
e = time()
for j = 1 to 1000000 do
for i =1 to length(ss) do
x = append(x,ss)
x = {}
end for
end for
? time() - e
? aa
---------------
------
Derek Parnell
Melbourne, Australia
"To finish a job quickly, go slower."
----- Original Message -----
From: martin.stachon at worldonline.cz
To: EUforum
Sent: Saturday, April 28, 2001 8:01 PM
Subject: loops & perfomace
Hi,
i've got an performace question. We've got a loop
for i=1 to length(s) do
...
end for
is length(s) evaluated every time loop is done? So would be in Euphoria
faster
l= length(s)
for i=1 to l do
...
end for
as i read in one C book?
----------------------------------------
4. Re: loops & perfomace
Euphoria is smarter than C...
----- Original Message -----
From: Ted Fines <fines at macalester.edu>
Subject: Re: loops & perfomace
> >
> >
> I don't think it makes a difference. Euphoria would appear to be
executing
> the for loop the 'smart' way. I did a quick test with this code--
>
> sequence s
> s=repeat(32,50000000)
> atom a,t0,loop_overhead
> integer lengthofs
>
> t0=time()
> lengthofs=length(s)
> for c=1 to lengthofs do
> a=power(2,20)
> end for
> loop_overhead=time()-t0
> puts(1,sprintf("%d seconds",loop_overhead))
>
> ...and then tried it, replacing this line:
> for c=1 to lengthofs do
> with this one:
> for c=1 to length(s) do
>
> and it ran at the same speed.
>
> This code was just lifted from the docs located in
> C:\EUPHORIA\HTML\LIB_S_T.HTM, and slightly modified...
>
> --Ted
>
> --On Saturday, April 28, 2001 12:01 PM +0200 martin.stachon at worldonline.cz
> wrote:
>
> >
> >
> >
> > Need a better decision-making tool to research, evaluate and
> > select software for enterprise applications or manufacturing
> > point solutions? Masg.com is where you'll find help.
> >
> >
> > Hi,
> > i've got an performace question. We've got a loop
> >
> > for i=1 to length(s) do
> > ...
> > end for
> >
> > is length(s) evaluated every time loop is done? So would be in Euphoria
> > faster
> >
> > l= length(s)
> > for i=1 to l do
<snip>
> >
> >
> >
5. Re: loops & perfomace
- Posted by jbrown105 at HotPOP.com
Apr 28, 2001
Probally because Eu saves the data from length into an int, but
doesn't for the eu-integer type.
P.S. Off topic from message thread, anyone know any good guis for "Euphoria
for Linux"?
jbrown
On Sun, Apr 29, 2001 at 02:34:42AM +1000, Derek Parnell wrote:
>=20
>=20
> With the quick test I did below, the times seem to say that length() is
> faster than integer! But only one tenth of a second over 67 million
> iterations. So I would worry about it.
>=20
> ---------------
> atom e
>=20
> integer aa
> sequence ss, x
>=20
> ss =3D "jhsdkjahsh akshdkjahsdkjaskjdaskhjdakjsdhkjashdkjahskdjhaksjdhkas=
dh"
> x=3D{}
> aa =3D length(ss)
> e =3D time()
> for j =3D 1 to 1000000 do
> for i =3D1 to length(ss) do
> x =3D append(x,ss)
> x =3D {}
> end for
> end for
>=20
> ? time() - e
> ? aa
> ---------------
>=20
> ------
> Derek Parnell
> Melbourne, Australia
> "To finish a job quickly, go slower."
>=20
> ----- Original Message -----
> From: martin.stachon at worldonline.cz
> To: EUforum
> Sent: Saturday, April 28, 2001 8:01 PM
> Subject: loops & perfomace
>=20
>=20
>=20
> Hi,
> i've got an performace question. We've got a loop
>=20
> for i=3D1 to length(s) do
> ...
> end for
>=20
> is length(s) evaluated every time loop is done? So would be in Euphoria
> faster
>=20
> l=3D length(s)
> for i=3D1 to l do
> ...
> end for
>=20
> as i read in one C book?
>=20
> ----------------------------------------
>=20
>=20
>=20
>=20
>=20
> =3D=3D^=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
>=20
<snip>
--=20
Linux User:190064
Linux Machine:84163
http://jbrown105.1avenue.com
6. Re: loops & perfomace
This is a multi-part message in MIME format.
------=_NextPart_000_0024_01C0CFF0.02A2D3A0
charset="iso-8859-2"
It also states in the Euphoria performance docs to save values where you can
I did a test of length( ) and found that if I got the length of a sequence
OUTSIDE the loop first that this was the faster method...
Euman
----- Original Message -----
From: martin.stachon at worldonline.cz
To: EUforum
Sent: Saturday, April 28, 2001 05:01
Subject: loops & perfomace
Hi,
i've got an performace question. We've got a loop
for i=1 to length(s) do
...
end for
is length(s) evaluated every time loop is done? So would be in Euphoria
faster
l= length(s)
for i=1 to l do
...
end for
as i read in one C book?
----------------------------------------
------=_NextPart_000_0024_01C0CFF0.02A2D3A0
Content-Type: text/html;
charset="iso-8859-2"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-2">
<META content=3D"MSHTML 6.00.2462.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>It also states in the Euphoria =
performance docs to=20
save values where you can</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>I did a test of length( ) and found =
that if I got=20
the length of a sequence</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>OUTSIDE the loop first that this was =
the faster=20
method...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV>Euman</DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV=20
style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: =
black"><B>From:</B>=20
<A title=3Dmartin.stachon at worldonline.cz=20
=
href=3D"mailto:martin.stachon at worldonline.cz">martin.stachon at worldonline.=
cz</A>=20
</DIV>
<DIV style=3D"FONT: 10pt arial"><B>To:</B> <A =
title=3DEUforum at topica.com=20
href=3D"mailto:EUforum at topica.com">EUforum</A> </DIV>
<DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Saturday, April 28, 2001=20
05:01</DIV>
<DIV style=3D"FONT: 10pt arial"><B>Subject:</B> loops & =
perfomace</DIV>
</PRE><PRE>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Need a better decision-making tool to research, evaluate and
select software for enterprise applications or manufacturing
point solutions? Masg.com is where you'll find help.
<A =
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D</PRE>
<DIV><FONT face=3D"Arial CE" size=3D2>Hi,</FONT></DIV>
<DIV><FONT face=3D"Arial CE" size=3D2>i've got an performace question. =
We've got a=20
loop</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3DCourier size=3D2>for i=3D1 to length(s) =
do</FONT></DIV>
<DIV><FONT face=3DCourier=20
size=3D2> ...</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2>end for</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3D"Arial CE" size=3D2>is length(s) evaluated every =
time loop is=20
done? So would be in Euphoria faster</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3DCourier size=3D2>l=3D length(s)</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2>for i=3D1 to l do</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2> ...</FONT></DIV>
<DIV><FONT face=3DCourier size=3D2>end for</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3D"Arial CE" size=3D2>as i read in one C =
book?</FONT></DIV>
<DIV> </DIV>
Window? Not=20
yet. First, check our technology forum for help. <A=20
=
=
href=3D"http://topica.com/u/?b1dd66.b2lJ88">http://topica.com/u/?b1dd66.b=
2lJ88</A>
Or send an email To: EUforum-unsubscribe at topica.com
This email was sent to: euman at bellsouth.net
<A =
=3D=3D^=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
------=_NextPart_000_0024_01C0CFF0.02A2D3A0--
