Dear Chris,
The -D does NOT mean double, and the program should run in
single-precision if compiled correctly. The run will use whatever it says
in the beginning, so something else is still wrong. Are you sure the
LD_LIBRARY_PATH AND PATH point to this install now?
HTH,
Sjors
> Sjors
>
> I tried to compile 1.4 in single precision with INSTALL.sh script by
> making the suggested change:
>
> # Use single-precision instead of the default of double-precision?
> FLOAT_PRECISION=true (the default was false)
>
> It compiled fine and runs, but in the build log file it has all the
> commands with -DFLOAT_PRECISION
>
> and I am wondering if indeed this means double float precision?
>
> In the standard output file for my mpi run I also get repeated messages
> that the job is running as
> double precision!!!!
>
> With my cluster, I could really use the memory boost of single precision,
> Not sure what to try next?
>
> cheers,
>
> C Akey
>
--
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|