Sometimes, you need ODE in double precision. This makes it convenient. I adapted the code for the imagemagick formula. I hope this is the right way to do it.
Added --enable-double-precision option to ode formula
Could somebody could take a look at this pull request ? It's small, uncontroversial and useful.
+1 for this pull request. The current formula is causing really unnecessary documentation issues if one wants double precision.
+1 for this pull request also from me.
without it getting double precision in ode is a bit of a hack...
Is there a reason this should not always be enabled? What if we end up with things that depend on ODE being compiled both ways?
We have two choices :
1. The careful one : we just add the possibility of double precision through an option. It doesn't affect anybody (anymore than now) unless people that know what they are doing.
2. The optimist one : we activate double precision by default. I don't know enough about this issue to recommend or argue against such a decision.
ode: add --enable-double-precision