Dear Ferret users,
I seem to be missing something, but I can't even list a single point of my 4D variable:
yes? save/clobber/i=1:1/j=1:1/k=1:1/l=1:1/file=tmp.nc temp[i=1,j=1,k=1,l=1] . . . memory error . . .
yes? list/i=1/j=1/k=1/l=1 temp
. . . memory error . . .
(Some more details are at the end of this message).
Basically, I just want to convert a 4D variable in a raw-binary format to a netCDF file.
Is there any fundamental limit to the size of the underlying raw binary file?
The array is in the regular Fortran order of (i, j, k, l) and because i=1:500 and j=1:400, the x-y plane (500 x 400) isn't so large. It's just 1.5 MB in double precision. So, I thought saving by
repeat/L=1:73 ( \
repeat/K=1:67 ( \
save/append/klimits=1:67/file=output.nc temp ) )
would be manageable.
Regards,
Ryo
---------
NOAA/PMEL TMAP
PyFerret v7.5 (optimized)
Linux 4.15.0-1061-azure - 11/14/19
1-May-20 16:40
[. . . ]
yes? file/form=STREAM/type=r8/var=temp/grid=grid_xt_yt_zt_ctime\
temp-pntd.bin
yes? set memory/size=300
yes? save/clobber/i=1:1/j=1:1/k=1:1/l=1:1/file=
tmp.nc temp[i=1,j=1,k=1,l=1]
**ERROR: request exceeds memory setting
To fulfill this request would exceed the current SET MEMORY/SIZE= limit of 300 megawords
At the moment that the memory limit was reached
memory was committed as follows:
- to objects used in computation: : 0 (0%)
The size of the requested object was: : 978200000 (326%)
yes? show grid temp
GRID GRID_XT_YT_ZT_CTIME
name axis # pts start end subset
XT LONGITUDE 500mr 105.05E 154.95E(154.95) full
YT LATITUDE 400 r 44.95S 5.05S full
ZT Z (METERS) 67 i 2.5 767.22 full
CTIME TIME 73mr 06-JAN 00:00 01-JAN 00:00 full
yes?