My guess is the following. They've had to dedicate one, or...

My guess is the following.
They've had to dedicate one, or several, wafers from each of the first and second non platform lots, to make them identical to the additional wafer lot, to prove repeatability. I think the second platform lot has two parallel sets being run, with split conditions, one to continue on the path of repeatability from the additional wafer lot, and the second incorporating the learnings from the second non platform lot, in an attempt to prove repeatability of the best wafer from the second non platform lot.