Experimental Physics and
Industrial Control System

"Allison, Stephanie" <[email protected]> · Tue, 17 Jun 2014 14:56:28 -0700

Hi Zen,

> Why is the widget showing 11 point when NORD is 10?

Actually, that 11th point is actually 990 points plotted on top of each other so that it looks like 1 point.

In my experience (and perhaps someone knows a way around this), EDM doesn't care about NORD.  It plots everything returned by channel access which is NELM (1000) points.  What I do to make the plot look right is to fill in the unused part of the waveform with the same last value so that the 10th point is 991 values plotted on top of each other.  What a kluge.

Hopefully, someone has a better way around this.  Hopefully, tools like MEDM and BOY plot only NORD of a waveform instead of NELM in their X-Y graph equivalents.

Stephanie Allison

> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf
> Of Zenon Szalata
> Sent: Tuesday, June 17, 2014 12:27 PM
> To: [email protected]
> Subject: EDM XY Graph
> 
> Hi,
> I am developing an IOC to fit data using Chi Square fitting routine.
> This is working well.  I don't understand why the EDM x-y graph widget
> is showing one extra point, see attachment.
> Here are the test data that is being fitted with a quadratic function:
> -1.7237128     23.095947     0.2
> -1.3703436     15.971818     0.2
> -0.96608055    10.204468     0.2
> -0.28394297     5.40507      0.2
> -0.14970151     4.9337711    0.2
>   0.042581975    4.7668524    0.2
>   0.82065094     8.7348375    0.2
>   1.3416969     15.787588     0.2
>   1.3757038     16.520903     0.2
>   1.8712276     26.449392     0.2
>   where the first column is x, second is y, and third is error on y.
> The EDM graph shows the y data points as circles (red) and the fitted
> curve (blue).
> I have three waveform records of type FLOAT and NELM 1000.  When these
> waveforms are filled with data, NORD is 10 in each record.
> 
> Why is the widget showing 11 point when NORD is 10?
> I don't know how to make it not show the 11th point at (0,0).
> Can it be done by changing some parameter in EDM?
> I see this behavior in an older version of EDM as well as in version
> 1-12-96.
> Any suggestions will be welcome.
> Thanks,
> Zen

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System