Some results on OSX, Intel dualcore 2.8GHz, ParCS compiled from CVS
this afternoon, now correct

with ksmps = 100

csound -n sheppard.csd   =>  0.450s, CPU: 0.432s
csound -n sheppard.csd -j 1 => 0.436s, CPU: 0.421s
csound -n sheppard.csd -j 2 => 0.828s, CPU: 1.293s
csound -n sheppard.csd -j 4 =>  2.170s, CPU: 3.672s

with ksmps = 1000

csound -n sheppard.csd   =>   0.392s, CPU: 0.378s
csound -n sheppard.csd -j 1 => 0.393s, CPU: 0.377s
csound -n sheppard.csd -j 2 => 0.287s, CPU: 0.476s
csound -n sheppard.csd -j 4 =>  0.420s, CPU: 0.710s

So the ksmp sizes do make a difference. Interesting to see that the  
'real' timing is, with > 2 threads less than the CPU time.
Also, as you can see, there is a gain with -j 2, processing is 1.36  
times faster

Sorry for the earlier mistakes, I thought I was dreaming. Must the  
fever I have at the moment.

Victor


------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

What is -j?

Regards,
Mike

On Tue, Aug 17, 2010 at 11:04 AM, Victor Lazzarini
 wrote:
> Some results on OSX, Intel dualcore 2.8GHz, ParCS compiled from CVS
> this afternoon, now correct
>
> with ksmps = 100
>
> csound -n sheppard.csd   =>  0.450s, CPU: 0.432s
> csound -n sheppard.csd -j 1 => 0.436s, CPU: 0.421s
> csound -n sheppard.csd -j 2 => 0.828s, CPU: 1.293s
> csound -n sheppard.csd -j 4 =>  2.170s, CPU: 3.672s
>
> with ksmps = 1000
>
> csound -n sheppard.csd   =>   0.392s, CPU: 0.378s
> csound -n sheppard.csd -j 1 => 0.393s, CPU: 0.377s
> csound -n sheppard.csd -j 2 => 0.287s, CPU: 0.476s
> csound -n sheppard.csd -j 4 =>  0.420s, CPU: 0.710s
>
> So the ksmp sizes do make a difference. Interesting to see that the
> 'real' timing is, with > 2 threads less than the CPU time.
> Also, as you can see, there is a gain with -j 2, processing is 1.36
> times faster
>
> Sorry for the earlier mistakes, I thought I was dreaming. Must the
> fever I have at the moment.
>
> Victor
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Number of threads.
On 17 Aug 2010, at 16:51, Michael Gogins wrote:

> What is -j?
>
> Regards,
> Mike
>
> On Tue, Aug 17, 2010 at 11:04 AM, Victor Lazzarini
>  wrote:
>> Some results on OSX, Intel dualcore 2.8GHz, ParCS compiled from CVS
>> this afternoon, now correct
>>
>> with ksmps = 100
>>
>> csound -n sheppard.csd   =>  0.450s, CPU: 0.432s
>> csound -n sheppard.csd -j 1 => 0.436s, CPU: 0.421s
>> csound -n sheppard.csd -j 2 => 0.828s, CPU: 1.293s
>> csound -n sheppard.csd -j 4 =>  2.170s, CPU: 3.672s
>>
>> with ksmps = 1000
>>
>> csound -n sheppard.csd   =>   0.392s, CPU: 0.378s
>> csound -n sheppard.csd -j 1 => 0.393s, CPU: 0.377s
>> csound -n sheppard.csd -j 2 => 0.287s, CPU: 0.476s
>> csound -n sheppard.csd -j 4 =>  0.420s, CPU: 0.710s
>>
>> So the ksmp sizes do make a difference. Interesting to see that the
>> 'real' timing is, with > 2 threads less than the CPU time.
>> Also, as you can see, there is a gain with -j 2, processing is 1.36
>> times faster
>>
>> Sorry for the earlier mistakes, I thought I was dreaming. Must the
>> fever I have at the moment.
>>
>> Victor
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by
>>
>> Make an app they can't live without
>> Enter the BlackBerry Developer Challenge
>> http://p.sf.net/sfu/RIM-dev2dev
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>
>
>
> -- 
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

If you have only 2 physical cores, there's no gain from -j > 2. Hence
the decline in performance after that. Where can I find sheppard.csd?

Regards,
Mikie

On Tue, Aug 17, 2010 at 11:04 AM, Victor Lazzarini
 wrote:
> Some results on OSX, Intel dualcore 2.8GHz, ParCS compiled from CVS
> this afternoon, now correct
>
> with ksmps = 100
>
> csound -n sheppard.csd   =>  0.450s, CPU: 0.432s
> csound -n sheppard.csd -j 1 => 0.436s, CPU: 0.421s
> csound -n sheppard.csd -j 2 => 0.828s, CPU: 1.293s
> csound -n sheppard.csd -j 4 =>  2.170s, CPU: 3.672s
>
> with ksmps = 1000
>
> csound -n sheppard.csd   =>   0.392s, CPU: 0.378s
> csound -n sheppard.csd -j 1 => 0.393s, CPU: 0.377s
> csound -n sheppard.csd -j 2 => 0.287s, CPU: 0.476s
> csound -n sheppard.csd -j 4 =>  0.420s, CPU: 0.710s
>
> So the ksmp sizes do make a difference. Interesting to see that the
> 'real' timing is, with > 2 threads less than the CPU time.
> Also, as you can see, there is a gain with -j 2, processing is 1.36
> times faster
>
> Sorry for the earlier mistakes, I thought I was dreaming. Must the
> fever I have at the moment.
>
> Victor
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

-o dac -d


0dbfs = 1 ; MAX AMPLITUDE IS 1.0 !!!!
ksmps=100

instr 1
idur = 120
kf   oscili 1,1/idur,3,p4
ka   oscili 1,1/idur,2,p4
asig oscili ka,kf*3000,1
      out asig/10
endin



f1 0 16384 10 1
f2 0 16384 20 6 1
f3 0 16384  5 1 16384 [2^-10]

i1 0  100  0
i1 0  100  0.1
i1 0  100  0.2
i1 0  100  0.3
i1 0  100  0.4
i1 0  100  0.5
i1 0  100  0.6
i1 0  100  0.7
i1 0  100  0.8
i1 0  100  0.9




On 17 Aug 2010, at 21:42, Michael Gogins wrote:

> If you have only 2 physical cores, there's no gain from -j > 2. Hence
> the decline in performance after that. Where can I find sheppard.csd?
>
> Regards,
> Mikie
>
> On Tue, Aug 17, 2010 at 11:04 AM, Victor Lazzarini
>  wrote:
>> Some results on OSX, Intel dualcore 2.8GHz, ParCS compiled from CVS
>> this afternoon, now correct
>>
>> with ksmps = 100
>>
>> csound -n sheppard.csd   =>  0.450s, CPU: 0.432s
>> csound -n sheppard.csd -j 1 => 0.436s, CPU: 0.421s
>> csound -n sheppard.csd -j 2 => 0.828s, CPU: 1.293s
>> csound -n sheppard.csd -j 4 =>  2.170s, CPU: 3.672s
>>
>> with ksmps = 1000
>>
>> csound -n sheppard.csd   =>   0.392s, CPU: 0.378s
>> csound -n sheppard.csd -j 1 => 0.393s, CPU: 0.377s
>> csound -n sheppard.csd -j 2 => 0.287s, CPU: 0.476s
>> csound -n sheppard.csd -j 4 =>  0.420s, CPU: 0.710s
>>
>> So the ksmp sizes do make a difference. Interesting to see that the
>> 'real' timing is, with > 2 threads less than the CPU time.
>> Also, as you can see, there is a gain with -j 2, processing is 1.36
>> times faster
>>
>> Sorry for the earlier mistakes, I thought I was dreaming. Must the
>> fever I have at the moment.
>>
>> Victor
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by
>>
>> Make an app they can't live without
>> Enter the BlackBerry Developer Challenge
>> http://p.sf.net/sfu/RIM-dev2dev
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>
>
>
> -- 
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Bison and flex run on my PC now and so I am close to finishing the
ParCS build, I know what the remaining small problems are. If my build
actually runs as it probably will, I will do some tests and
comparisons this evening.

Regards,
Mike

------------------------------------------------------------------------------
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net