There are no open paths where a glider could plausibly destroy that blinker -- it's really well protected by the catalysts around it. I hate to be one of those boring it-will-never-work people, because an incredibly lucky B-heptomino or other traveling pattern in exactly the right place could reach in there and kill off the blinker. Or some kind of edge-shooter, I suppose.Kazyan wrote:A quick 2-catalyst run from generations 90-130 using the catalyst searcher doesn't yield anything, so I guess it's up to manual searches or Bellman catalysts to get that working (and hopefully, delete that annoying blinker with a glider or something).
It's just not very likely that the reaction will be clean, so then there will be new and different junk to clean up... and the blinker will probably even have to happen to be in the right phase for it to work:
Code: Select all
x = 58, y = 20, rule = B3/S23
#C [[ THUMBNAIL AUTOSTART HEIGHT 240 LOOP 40 GPS 5 ]]
Run the reaction in LifeHistory and you'll notice that after T=108, the whole southwest corner of the reaction is hiding inside its previous reaction envelope -- there are just two cells at T=106 and T=107 that could possibly be affected by a new catalyst, to reshape that reaction.
The new cells at T=123 and up are clearly too late -- catalyzing those locations would amount to rebuilding the block again, and of course it's much less likely the second time around because you've lost a lot of potential places for catalysts.
So that southwest mess would probably all have to be cleaned up by catalysts in the northwest, in twenty ticks or less, let's say from T=95 to T=115 but it's probably a narrower window of opportunity. Effects of catalysis only travel at lightspeed, and it's a fair distance across the active reaction to the problem you want to control, so you have to go back in time in this weird way. But if you modify the reaction earlier than about T=95, you'll almost certainly wreck the toad->block restoration reaction, which defeats the whole purpose.
So... at least according to my best guess, Bellman or catgl time will be better spent working on other reactions. It's a shame to just throw away that nice-looking transparent block, but two simultaneous probably-insoluble problems are lurking there, and the search space seems to small to compensate for the low probabilities.
-- Apologies for the long-winded analysis. Maybe if I've done a good enough job of explaining, some of these tools will be useful in guessing the odds of success for new reactions. Seems like this business needs an instruction manual. Except that it would all be just rules of thumb, with more exceptions than rules: "Tips and Tricks for Conduit Hunters"...