This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
> Russ, > > I got it past the nf90_enddef "HDF error" by first storing the > chunking and extend_increments arguments in array variables, and then > passing those as arguments. Of course, I have no idea why this should > matter, and it only gets me past nf90_enddef if I compile with -g. If > I do this, then the test code slows down severely on nf90_put_var -- > it is taking 9 seconds to write a single 32x64x64 array, which means > it takes about 3 minutes to do what netCDF-3 was doing in between a > fraction of a second and a few seconds. Of course, it might also be > writing out all NaN's for all I know. I'll take another look at this > tomorrow. > > David > Howdy David! It's been some time since your support question, but I have just put in an important optimization in the netcdf-4 read/write code. Perhaps this might impact your problems? Why don't you get a copy of the netcdf-4 snapshot and give it a whirl? ftp://ftp.unidata.ucar.edu/pub/netcdf/snapshot/netcdf-4-daily.tar.gz Please let me know if this helps. Thanks, Ed Ticket Details =================== Ticket ID: NNS-961385 Department: Support netCDF Priority: Normal Status: Closed