Hello,
The code currently will draw any bit of the spectrum out to the nearest
block of data, so if it spills over just a little it will get more than
just a little extra drawn. But I can see from your final comments that
this is not done with respect to the peaks, so that is indeed confusing,
we'll have to fix this.
There is a data model item, analysisProject.contourToUnaliased, that if
set to False prevents the aliasing ranges from changing even if a peak is
outside that range. So if, on the other hand, it is True then when a peak
falls outside the existing range then the range is extended. And by
default it is true. We used to have a GUI way of changing this value but
when User Profiles were introduced this somehow got lost in the shuffle.
From the console you can set it via:
>>> top.analysisProject.contourToUnaliased = False
(or True).
Wayne
On Tue, 16 Mar 2010, Mattias Brink wrote:
>
> Hi,
>
> I have noticed something strange:
> Without touching the referencing or minimum and maximum aliased frequencies
> I suddenly saw extension of the contours of my spectrum, showing aliased
> peaks outside the normal spectrum boundaries.
> We had a careful look at it, and it seems that changing the maximum aliased
> frequency with only 0.001 ppm make the difference between a normal
> contoured spectrum, and a spectrum with contours extended (duplicated) 4
> ppm in 13C. I am wondering if there may have been some rounding error
> somewhere that may have caused the maximum aliased frequencies to change a
> tiny bit, and that that caused the spectrum contours to extend. I have been
> adapting tolerances for spectra. One more thing: the extended contours
> contain a peak that was picked on its normal position, but the aliased peak
> box was not shown, only after the maximum aliased frequency was further
> increased it would show it. This added to the confusion.
>
> Regards,
>
> Mattias
|