I don't see the point of the question. It depends on the specs of the CPU and the specs of the GPU. Kangaroo is about computing, not about memory, so usually it should be very clear what the winner would be.

It seems you want to hint that when you have an identical throughput (total jumps/s), having more kangaroos (that each jumps slower) is better than having less kangaroos (that each jumps faster). If this is so, care to explain what is the reason of why you believe it is better?

Why do I feel like these 120 125 130 puzzles were not solved recently? Who would waste tons of $ just to get some key no one really cares about anymore? Maybe it was fun for 120 but 125 and 130 are no longer having any fun, as much as someone's boredom toy.

This is too much to grasp in such a short period of time. I am giving up at this, it starts to become a bad joke, not a challenge.

I encourage everyone to approach this with maximum care, and if you are already good at something that pays for your life, stick at that as a priority. Sorry to everyone with whom I seemed to engage in a too provocative manner at times - professional defect on my side. Take care.

Are you going to start with that BS? Why as soon somebody has some success, everyone start talking about conspiration Theories? WHY?

There is no way to prove or disprove it. So there is no sense of keep talking in that way.


He mention that he use his own optimized kangaroo version, so what we can deduce from this is that the JeanLucPons kangaroo's version wasn't enough efficient as expected.

If this is possible, Puzzle 70 can be solved in a few seconds on a potato PC.

As evidence, provide us with an EXE file that can solve any private key at 70 bits in seconds.

@RetiredCoder how close are you to solve #135?

Some people here have made very strange conclusions. Again, to stop it:
1. I never said that I did not solve puzzles.
2. I never said that I used 20 GPUs for #125 and #130.
3. I have no idea how some people calculated time two months between solving #125 and #130   it was a year or so.

Reality is:
1. I used 20 GPUs for #120. I had to get access and use >100 GPUs for #125 and #130.
2. There is no magic in solving these puzzles, just pure math and skills. And some money.
3. I'm not puzzle creator, this is a ridiculous idea if you think about it.
4. I'm working on 135, but it will take a lot of time, I'm not in a hurry. If you think that you compete - welcome But 99% of people even cannot correctly calculate chances and cost of this game.
5. I want to share my experience in open source, step by step. But it won't be ready-to-use software so everyone can just run it and wait for money  

Have you ever considered creating a pool to search the 135 quicker and split profits based on the found DPs?

That is true, it was a more than a year

Puzzle 120 was redeem on February  27, 2023 (anyone can check it here)
Puzzle 125 was redeem on July 9, 2023 (check it here)
Puzzle 130 was redeem on September 23, 2024 (check it here)

That was 1 year and two months between puzzle 125 and 130, with >100 GPU

wow that is still a world record no?

All of them were Nvidia 4090 ?


Nice that is fair, any information about the changes on kangaroo is welcome.

Thanks  for sharing your experience, btw i have three questions.

Why did you wait to much to announce it?
Did you receive external funding?
What is your ROI?

I found the "bug" of my previous attempts to increase DP without increasing runtime cofactor.

31 bits, DP 6


39 bits, DP 8


47 bits, DP 10


No cycles. No "hey, let's create kangaroos and pretend they don't count as a group operation".

This only works for secp256k1. It is basically 100% faster than the usual approach.

I hope no one actually believes they can compete with RetiredCoder aka 3Emi. It looks like a billionaire's hobby, not a contest. It costs at least 500K to break 135, but he doesn't do it for the money, and I won't be a sheep.