both, even with fire counts (a windproof grill lighter, one of my utube vids shows fire burning red algae nicely)
if you did all three, then you must be very serious about winning in that exact spot ~ there is no amnt too much or too creative or too harsh, for that one test spot.
aware how crazy this sounds lol many would take a calmer approach. every algae ruined tank we ever saw had a phase like yours where tufts were easy to get out but were left in...it seems to me the most thorough response is the only justified use of time in hindsight.
after so many wrecked tank threads am now reduced to pure overpowering destruction of algae. others take a calmer mode.
the mechanism I found was this: its lawn mowing (which gets growback weekly) vs digging up the plant by the root then burning the spot with kerosene and roundup
the fact you gouge-remove the holdfast/growback structures is the real winning act, nature does this via parrotfish and hawksbills and the raspers of the reef.... the chem cheat we use on the cleaned areas is just for cleanup of missed holdfasts, so tech m works, peroxide 3%, peroxide 35% which is what I use (never playing around lol always overdone and an old reef results from that) or even fire burning the rock to kill the leftovers. of them, I find 35% peroxide to be the best but 3% is ok, Tech M about equal actually to 3% as a simple external spot treat.
the reason we found out that applying the Tech M works is because its simply amplifying whatever mechanism it has that makes it work at 1500 ppm as a tank water solution. imagine its power full on...
so, this method is heavy on external work, not practical for all setups. but its the clincher, anyone with a true risk invasion who wants the upper hand can test rock a noncommittal area, and we think that will work out so well it would be worth it to expand the effort to other accessible areas. the heart of our method is get mean, but actually don't do anything to the water table, don't get mean to the non targets.
just my take

B