home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.ms-windows.apps
- Path: sparky!uunet!munnari.oz.au!mel.dit.csiro.au!dmp.csiro.au!lachlan
- From: lachlan@dmp.csiro.au (Lachlan Cranswick)
- Subject: Re: Gridding in Sigma Plot
- Message-ID: <1993Jan26.013952.12039@dmp.csiro.au>
- Organization: CSIRO Division of Mineral Products, Melbourne, AUSTRALIA
- References: <25JAN199309174449@cc.utah.edu>
- Date: Tue, 26 Jan 1993 01:39:52 GMT
- Lines: 49
-
- msfuller@cc.utah.edu (MARC S. FULLER) writes:
-
- >Does anyone know what capabilities Sigma Plot has for gridding (turning
- >irregularly spaced 3-d data in regularly spaced 3-d data)? Do they offer
- >different algorithms, and which ones?
- >Thanks.
- > Marc Fuller
- > msfuller@cardio.med.utah.edu
-
-
- The only thing I have done similar to this
- on Sigma Plot was to get a surface of best fit -
- then plot the resulting mesh. I'm not sure
- what algorithm was used as it was not important.
-
- As SigmaPlot can used custom algorithms, I would ask
- this question of sci.stats or a similar newsgroup -
- or find one of your stats gurus.
-
- The reason for this is I once had data from
- different Powder X-ray Diffractometers using different
- radiations. While the data was collected using constant
- 2-theta steps, to be comparable it had to be converted to
- d-spacings which made the data of non-uniform steps
- (I don't know of any Powder X-ray diffractometer that will do
- constant d-spacing scans). For the type of matrix analysis
- that was contemplated, each data set had to have the same
- number of points of relatively similar values -
- which lead to the idea of interpolating/smoothing
- the in-between points etc. Sort of similar to what you want
- to done.
-
- The site statistician put a spanner in the works
- because he said interpolating points might have artificially
- made peak shapes artificially more Gaussian or Lorentzian.
- Because the analysis from this interpolated data could have
- been suspect, it was safer not to do it - wimp out!!
-
- I don't know what you are using this data for but
- your local stats person would probably know the best
- algorithm that can then be put into the "math"-"transform" part
- of Sigma Plot 5.0 to give you your modified data.
-
- Hope this helps? Lachlan.
- --
- Lachlan Cranswick - CSIRO _--_|\ lachlan@dmp.CSIRO.AU
- Division of Mineral Products / \ tel +61 3 647 0367
- PO Box 124, Port Melbourne 3207 \_.--._/ fax +61 3 646 3223
- AUSTRALIA v
-