Lloyd, Thanks for the reply.
The situation I described (using Compute to shift the Z position) was just a sample test case. I'm trying to use your ConvertToCellCentered macro on my data which is regular in x and y, but delta z varies. You had suggested in an earlier e-mail that I use Slice to create a set of horizontal slices as a series group. Then do a ForEachMember loop for each member of the series and use the macro and accumulate the results in a new series. Then, use Stack to recreate the volume. However, DX doesn't think the slices are regular so ConvertToCellCentered doesn't work. I have imported the data from a netCDF file and do not alter the data at all before I create the series of Slices. Am I just out of luck on this, or do you have any suggestions. Thanks...Betsy [EMAIL PROTECTED] wrote: > The short answer is because Mark->Compute->Unmark generically "warps" a > mesh (changes positions while connections, data, etc. don't change). Some > operations really do this (e.g., a map projection) others don't have to > (like your example). One workaround would be to have the positions be > generated by Construct with the displacement already. Another is to use > Translate to have the displacement applied at render time. > > Betsy Foote <[EMAIL PROTECTED]>@opendx.watson.ibm.com on 08/22/2000 > 02:54:08 PM > > Please respond to [email protected] > > Sent by: [EMAIL PROTECTED] > > To: opendx-users <[email protected]> > cc: > Subject: [opendx-users] why is data irregular? > > I do the following: > > 1) Use Construct to make a 3D regular field. > 2) Add 10 to the z component with > Mark->Compute(a)=[a.x,a.y,a.z+10]->Unmark > > Now if I use Inquire(is regular positions), I find that the field is no > longer regular. > > Why??? > > Thanks, > > Betsy > > P.S. This problem stems from the following situation: > > I have 3D data which is regular in x and y, but the delta in the z > direction varies. I want to take a series of slices in the z direction > and run them through a macro which requires a regular field and then > Stack the results together. I haven't been able to create a regular > slice, even though the data is regular in x and y.
