If we go with distinguished points space goes down, expected time goes up. There's no free lunch.
Currently testing ~ 500 billion tame kangaroos footprints against wilds, I don't really expect a collision but who knows. I'd need like millions of times more storage.
I don't think #120 or #125 were solved by existing (public) software, it just seems unrealistic from a resources / cost perspective. There's something else going on there.
Lol, why?
Current speed for a sample of cards using Kangaroo:
RTX 2080Ti = 3,790 MKey/s
RTX 3070 = 3,085 MKey/s
RTX 4070 = 3,900 MKey/s
RTX 4080 = 5,260 MKey/s
RTX 4090 = 7,500 MKey/s
H100 SXM = 13,600 MKey/s
For #120, that is roughly 58 days with 64 RTX 4090s, to solve
For #125, with 128 RTX 4090s, that would be around 163 days, to solve.
The software and hardware is out there.
So you basically stored 500 billion DP 0, tames, basically just printing pubs and privs to a file, and now are offsetting 130s pub by random amounts, and looking for a collision?
For the traditional Kangaroo algo, for 130, with DP 32, I need to find only 9 billion tames and 9 billion wilds to solve. So it sounds like you just stored random pubs and privs, because 500 billion tames, with a decent DP, would take a loooooong time.
Also, you need to perform roughly 2^66.05 "steps" for #130, that would be the average.