valve lift is so incredibly important that it always be identically set for each and every test. Even just a few thousandths of an inch in valve lift is going to make a noticeable difference in flow since the valves themselves are the biggest bottleneck - as shown in the comparison between my test and the one shown here as well as what you have already stated.
Flowtesting *just* the manifolds themselves eliminates the possibility of there being any difference in valve opening alltogether.
