MathGroup Archive 2012

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Could someone help optimize this for Mathematica 9?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg128954] Re: Could someone help optimize this for Mathematica 9?
  • From: Christoph Lhotka <christoph.lhotka at fundp.ac.be>
  • Date: Wed, 5 Dec 2012 03:15:18 -0500 (EST)
  • Delivered-to: l-mathgroup@mail-archive0.wolfram.com
  • Delivered-to: l-mathgroup@wolfram.com
  • Delivered-to: mathgroup-newout@smc.vnet.net
  • Delivered-to: mathgroup-newsend@smc.vnet.net
  • References: <20121204091410.B2B8E68C4@smc.vnet.net>

Dear Paul,

  I did not investigate the issue further but I have the feeling that it 
has to do something with packed arrays:

I get a speed up of a factor of 6.5 if I change your code into (I just 
add a '.' after 100000 and 10):

mNumberSamples = 100000.;
mFrequency = 10.;
m = Table[{t/mNumberSamples,
     Sin[t/mNumberSamples mFrequency*2 \[Pi]]}*1.0, {t, 1,
    mNumberSamples}]; mStartTime = SessionTime[];
ListLinePlot[m]
mStopTime = SessionTime[];
StringForm["Elapsed time = ``", N[mStopTime - mStartTime]]

If you try your original code Developer`PackedArrayQ[m] returns False while
in my modified code Developer`PackedArrayQ[m] returns True. So I guess
that ListLinePlot works faster with packed arrays.

Best,

    Christoph




On 12/04/2012 10:14 AM, McHale, Paul wrote:
> We have a strong need to view large numbers of data points.  Mathematica 8 was excellent at this.  The following code executed under both 8 and 9 (Windows):
>
> Mathematica 8: Less than 0.5 seconds
> Mathematica 9: Greater than 4 seconds
>
> mNumberSamples=100000;
> mFrequency=10;
> m=Table[{t/mNumberSamples ,Sin[t/mNumberSamples mFrequency*2 \[Pi]]}*1.0,{t,1,mNumberSamples}];mStartTime=SessionTime[];
> ListLinePlot[m]
> mStopTime=SessionTime[];
> StringForm["Elapsed time = ``", N[mStopTime-mStartTime]]
>
> I have tried the code on a slower mac laptop (version 9 only) and had the same slower results.  Any input welcome as this functionality is extremely important to us.
>
>
> Paul McHale  |  Electrical Engineer, Energetics Systems  |  Excelitas Technologies Corp.
>                           
>                           
>                           
>                           
>             
> Phone:   +1 937.865.3004   |   Fax:  +1 937.865.5170   | Mobile:   +1 937.371.2828
> 1100 Vanguard Blvd, Miamisburg, Ohio 45342-0312 USA
> Paul.McHale at Excelitas.com
> www.excelitas.com
>
>
>
>




  • Prev by Date: What algorithm does DegreeGraphDistribution use?
  • Next by Date: Re: Integral that should converge, but Mathematica says it does not
  • Previous by thread: Re: Could someone help optimize this for Mathematica 9?
  • Next by thread: Re: Could someone help optimize this for Mathematica 9?