1) Bad idea? Good idea? Why?
* If you use typedefs in a header file that are easy to switch over,
* so that groups of calculation are consistently done in one mode,
* and you don't do stuff that would pipeline on a vector machine,
then it is easy to benchmark the effect. The impact should be
Jon S. Berndt wrote:
I am thinking that some of our problems in JSBSim may be due to numerical
precision. To rule out this I am considering moving all floats to doubles. The
results of this are something I am not fully able to address, so I am asking
for opinions on this:
1) Bad idea?
Probably you could indentify different precition needs and typedef them
your own.
E.g. jsbsim_float jasbsim_double and jsbsim_log_double. Then you only
need to change one file (and there only 3 lines) to figure out what
precition you want or need.
for file in *.cpp, *.h, filtersjb/*.cpp,
