Scaling of 3D spectrum intensities

Analysis Version 3
#1

I have the unusual problem that I cannot set an adequate contour level for a 3D NOESY spectrum that I moved over from 2.4 where I had no problems at all. In 3.2 I realised to my horror there seems to be a hard limit for contour levels at 1e12. Unfortunately that’s not enough.
I tried to change the scaling in Topspin but changing NC_proc with ft3d only got me an error message. Apparently it has to scale with NC_proc = 12 to work.
I also tried other maths in Topspin but while there are options for 1D and 2D spectra there was nothing I could find for 3Ds.
Is there any way that this hard limit of 1e12 can be changed? Or is there some scaling that one can done in Topspin?

Thank you very much,

mark

#2

Hi,

CCPN software currently uses NC_proc to put the Bruker stored values back to their original scale rather than values that fit in 32 bit int. This is done so that when comparing data that have been stored with different NC_proc, the relative intensities are faithfully maintained even if stored with different NC_proc scaling.

Your NC_proc value of 12 implies that your highest data points were in the region 2^(28+12) to 2^(29+12) (see the Topspin processing manual) so just creep over 10^12 but I guess it’s possible that Topspin processed data could have values up to 2^63 or so we should allow contours up to 10^19.

A workaround if you don’t need to compare the intensities in this spectrum with other datasets directly would be to edit your procs file in the Topspin dataset (with a text editor) and set NC_proc to a smaller number (e.g. 1 should be fine) before reloading to assign. Relative intensities within the spectrum will be maintained and a dynamic range of 5x10^8 should be enough for anyone!

Best wishes,
Brian Smith