Thread: spectrogram: strange behavior

Started: 2012-08-03 16:34:22
Last activity: 2012-08-10 20:32:43
Topics: SAC Help
marco.olivieri@bo.ingv.it
spectrogram: strange behavior
2012-08-03 16:34:22

Dear SAC users,
I could notice unexpected differences in spectrogram output when
changing ymin and ymax ranges, and I did this synthetic test:

SAC> fg random 1 2 npts 1200 delta 0.5 ; w noise.sac
SAC> fg sine 1 0 npts 1200 delta 0.5 ; mul .5 ; w sine.sac
SAC> r sine.sac ; addf noise.sac ; w sac_noise.sac
SAC> r sac_noise.sac ; rtr ; spectrogram w 30 s 5 ymin .05 ymax 1.5
<-- case 1
SAC> r sac_noise.sac ; rtr ; spectrogram w 30 s 5 ymin .5 ymax 1.5
<-- case 2, (zoom around 1Hz)

in case 1, I see the orrect spectrogram with peak around 1
while in case 2, the peak is at about 0.8. (snapshots attached)

Am I doing something wrong or this is an anomaly of spectrogram?

Thanks
Marco



Attachments
  • Brian Savage
    Re: spectrogram: strange behavior
    2012-08-10 20:32:43
    Macro,

    That looks like a bug to me. I was able to reproduce the behavior and found that the spectrogram code does not work properly if a window on the frequency is set. If you change the second case ymin value, you can effectively move your spectrogram peak in the plot. This is not good.

    A fix for this bug will be included in an upcoming version of SAC (maybe sooner considering the severity)

    Thanks so much for submitting the bug report.

    Brian Savage
    savage<at>uri.edu

    On Aug 3, 2012, at 3:34 AM, marco.olivieri<at>bo.ingv.it wrote:


    Dear SAC users,
    I could notice unexpected differences in spectrogram output when changing ymin and ymax ranges, and I did this synthetic test:

    SAC> fg random 1 2 npts 1200 delta 0.5 ; w noise.sac
    SAC> fg sine 1 0 npts 1200 delta 0.5 ; mul .5 ; w sine.sac
    SAC> r sine.sac ; addf noise.sac ; w sac_noise.sac
    SAC> r sac_noise.sac ; rtr ; spectrogram w 30 s 5 ymin .05 ymax 1.5 <-- case 1
    SAC> r sac_noise.sac ; rtr ; spectrogram w 30 s 5 ymin .5 ymax 1.5 <-- case 2, (zoom around 1Hz)

    in case 1, I see the orrect spectrogram with peak around 1
    while in case 2, the peak is at about 0.8. (snapshots attached)

    Am I doing something wrong or this is an anomaly of spectrogram?

    Thanks
    Marco


    <case_1.png><case_2.png>_______________________________________________
    sac-help mailing list
    sac-help<at>iris.washington.edu
    http://www.iris.washington.edu/mailman/listinfo/sac-help



NSF SAGE Facility has been certified as a trustworthy data repository by the CoreTrustSeal Standards and Certification Board
20:45:53 v.b3198453