Solving ECDLP with Kangaroos: Part 1 + 2 + RCKangaroo

[Bram24732]

Possible.
The basic idea for N gaps is to split jumps in the jump table to N groups, so every jump performs a small jump in its (random) gap. Also define start positions of kangs in the same way.
Visiting known bits are possible of course after many jumps (same as for single range) but won't happen too often if you choose jumps and start positions properly.

That was my initial intuition yes. Would be cool to build the client side for this, great thing is you can re use the server part as nothing else changes there.

[kTimesG]

Possible.
The basic idea for N gaps is to split jumps in the jump table to N groups, so every jump performs a small jump in its (random) gap.

Only if the gaps are large enough to end up with something faster than (N-1)*linear + sqrt ops. Complexities around visited ranges compound.

I guess this will be your mini-puzzle challenge after you grab 135. Shall everyone fire up their coding skills in advance?

[Bram24732]

Possible.
The basic idea for N gaps is to split jumps in the jump table to N groups, so every jump performs a small jump in its (random) gap.

Only if the gaps are large enough to end up with something faster than (N-1)*linear + sqrt ops. Complexities around visited ranges compound.

I guess this will be your mini-puzzle challenge after you grab 135. Shall everyone fire up their coding skills in advance?

I’ll do it for fun this week - let’s see if I encounter unexpected setbacks

[RetiredCoder]

I guess this will be your mini-puzzle challenge after you grab 135. Shall everyone fire up their coding skills in advance?

Why should we wait? Let's have fun right now!   
https://bitcointalk.org/index.php?topic=5577390