Vivian Meazza wrote:

Curt has stated the policy of only using the released version of plib etc.
This makes sense to me. That decision really only affects the 'crease' in
our models - the released version of the data cannot include them. And, as
you said, we miss out on the significant performance gain. Of course, you
can build FG with any version of plib you like. Perhaps all that is
necessary is that we let people know that they would benefit by using plib
cvs, while ensuring that the released version of FG is compatible with plib

I agree that the policy should be that the code will build against a released version of plib. The question was really if there's a compelling reason *not* to use the cvs version for a binary package, where those downloading it don't need to worry about these dependencies.

Jon Stockill

Flightgear-devel mailing list

Reply via email to