These strips that you linked are not comprised of lm301h diodes. The series name is 'H' but that has nothing to do with lm301h diodes. Their efficiency is significantly less than lm301h, they state they are 159 lm/w or 2.7 umol/j, whereas the lm301h are 220 lm/w or 3.10 umol/j.
As has already been pointed out by PJ, they are 301H diodes. And is stated in the data sheet. ANd are the only 301H strips from samsung.
301B's come in Q series strips and H influx industrial strips from samsung. And are all, white no reds.
OK. I wonder why the efficiency is so much lower. Maybe their stated efficiency for the bare diode is potential or something?
301's don't hit your stated figure very often, and that stated figure is at 25c temp and 65ma drive current on the top bin of 2 cct's(originally just 5000K but now some 4000K 70cri's) and is not even available to order to most people oro even companies. Epi yields just don't support it...but sure "it can be done" in the lab and is great for marketing.
Thos strips are binned at a current that is not 65ma...as well as have red 660's mixed in too. those 660s make the LPW take a big hit as they are not very efficacious lumin wise.
If you compare apples to apples driver current per diode the strip would be more efficient with the addition of the reds and their photon weight. Even with small losses going from spec sheet to actually on a board.