Long double

From diagham
Jump to navigation Jump to search

DiagHam uses double precision numbers. If for you think this is not enough, you can turn on long double precision. Beware that you would not have access anymore to the Lapack support. The proper way to do that is the following

  • get a fresh new copy of DiagHam

svn checkout https://www.nick-ux.org/diagham/svn/trunk DiagHamLong

  • run bootstrap.sh
  • from the top directory DiagHamLong, run the following script

./scripts/switch2longdouble.pl

  • from the build directory, go as usual with something like

../configure --enable-fqhe


In addition to that, it might helpful to use Media:LongDoubleToDouble.cc or Media:DoubleToLongDouble.cc to convert binary vector from or to long double accuracy.