On 2012/04/15 15:20:13, Graham Percival wrote:
wow, looks very nice and simple. I vote for pushing immediately, but
please
wait for at least one other such vote (or a countdown).
if max/min trigger, rint is applied to some rest_max_pos. It would appear that it should rather apply only to the calculated mean. The calculated mean uses the current rounding mode. If it is IEEE rounding, this means round to even: (1+2)/2.0 will be rounded to 2.0, and (2+3)/2.0 will be rounded to 2.0 as well. That looks suspicious to me, like every second staff line will get avoided for values right in between. http://codereview.appspot.com/6035053/ _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
