If you use Darrel’s original code you want to add 1 to the constant reload value you use (to the table at the top of this thread).
It appears if you are using Tabsoft’s modified version that calculates the value for you based on crystal frequency,
it should be correct already (and has been the whole time).
We are still talking, but also we must be across the world from each other, I only get his emails live if I stay up late.

I’m only waiting to see what he’d like me to do (if anything) about testing it.
The best I can tell for all of the crystal values Darrel provided, using the correct constant value should cause the software
to perfectly keep time indefinitely... of course crystal tolerance makes this not possible..
but the software should not introduce any of it’s own error.

Some testing will happen in any case, a six month long project of mine is a GPS equipped atomic clock!