If I was in charge I would modify MTP. Whatever you can do to lesson FPGA’s profitablity.
I would also fix these if they aren’t already.
Let monero stay on randomx.
Progpow - Could be a good option but recently a bug was discovered so there may be more bugs to be found.
MTP has been patched against those potential attacks before it was released to public
MTP will be always susceptible to FPGA with the advent of FPGAs with large memory banks.
Those attacks have been fixed and even others addressed. https://arxiv.org/pdf/1606.03588v2.pdf
ProgPOW bug is fixed.
Just reviving this thread that we are still considering this but only after Lelantus and our other roadmap items are completed. We want to make sure we focus on delivering on the really important stuff first but it’s good to keep this discussion going.
Hope to have a proposal up soon.
whatever GPU friendly algorithm would make me happy. it’s been 2 years i think, since the last time i mined XZC. i can’t afford crashing every few hours mining XZC. i dunno why, since XZC uses MTP, i can’t keep the miner stable, always crash every few hours
Thanks William for the feedback!
Everything has pros and cons in many ways:
A) Optical POW is not tested yet and still very much a theoretical. Pros power neutral Cons likely high dollar entry.
B) ASICS high power, high dollar constant arms race. Operation favours centralization noise dampening isn’t cheap. Pros high security barring “secret” new chip designs.
C) Multi Algo progpow and randomX have as shown by EPIC problems with adjustments given how differnet the two are. DGB (yes it’s a shit coin) have one decent feature namely how their multipow distributes work. However, the DGB model can not to the best of my knowledge accommodate more constraint based algorithms like RandomX and Progpow.
D) ProgPow on its own, in light of the seed bug and additional utilization of it in a birthday attack as outlined on github seems like a potentially mixed bag. There are to the extent of my currently limited knowledge only a handful of operations that could perform a birthday attack as outlined on github. Trusting that these operations not abuse that position is frankly lunacy. If it were to be tabled, higher complexity seeds and the advantage might give to one GPU manufacturer over the other would be the route to go down to keep it on reasonably solid footing. RVN’s implementation is currently working well but they have chosen to incorporate some changes that may render a birthday attack ineffective.
E) RandomX on its own was outlined very well by Reuben a couple of posts back. However, as more and more projects incorporate it, the dangers of botnets get significantly diluted, 5 or 6 smaller- medium sized projects I know of are currently looking at implementing it or have people working on implementations.
F) UGLY KLUGE is a thoroughly unpalatable solution to most it is essentially redneck anglegrinder and welder pow where you add enough bits to end up with something that will not store easily on an FPGA or cost an absurd amount to implement on sillicon. The problem with this is it also limits its own implementability on GPUs.
G) Options that suck for everyone there is an implementation of lyra2x330 out which has a worksize that floods out hardware pretty well. Making it mineable both by CPUs and GPUs at roughly dollar parity on efficacy. I have no idea how that would implement on FPGAs or ASICS apart from a few comments of “DAMN that worsize is the size of Texas” and “it needs HBM.”
With the rather underwhelming performance of reasonably high dollar FPGAs on ETH (yes I know that is only public streams), it might be an option for bigger heads than mine to comment on.
$ parity on entry is about as fair as it can get in terms of making it inclusive for as large an audience as possibly. Who cares if a 2000$ FPGA gets 2mh/s if a 200$ CPU gets 200kh/s and a GPU somewhere in between or something proportionally close.
I think the bugs u mentioned on ProgPoW have all been addressed afaik.
I don’t see anything addressing what Solardiz raised on Github but I haven’t dug into it at the commit level. It’s not so much an attack in the sense of breaking it as it is some really specialized setups could have extensive advantages on it.
It boils down to the next evolution needing to support efforts put into a say a rebrand and help drive the coin forward providing the best possible security that lines up with visions and aims for community.
All of the options I mentioned work except for the possibility of the last one implementing too well on hardware. but they are not for everyone.
For ZCoin mining, I am for the exclusive CPU mining : everyone has a CPU, then everyone must be able to mine XZC :-).
Why exclusive?
Would looking for something where 1$ of CPU is roughly equal to 1$ of GPU or FPGA?
Because everyone has a CPU while a GPU is specific to gamers or miners.
The goal is for XZC can be mining by the majority and profitable for them (decentralization), not a minority who specialize in cryptomining.
How would dollar parity not do that better than exclusively CPU mining?
Excluding GPU and FPGA miners from participating lowers the potential amount of participants. More importantly it lowers the number of vested participants who have have an interest in seeing it grow. No investment= “I don’t need to care too much.”
Meaning you would likely end up reducing the overall number of people willing to market and promote the coin. Zcoin isn’t a fledgling any longer but neither is it yet so strong that it can afford to start alienating larger percentages of the space.
No, it’s the opposite : exlusive CPU mining increase the numbers of miners/participant and the interest about ZCoin.
Every one has one or more CPU, not a GPU, a FPGA or an ASIC : everyone can mine with his computer CPU, without buying anything (and some can buy more powerfull CPUs if they wants).
And each owner of GPU, FPGA, ASIC has at least one CPU (and often several) : they are not at all excluded.
Exclusive CPU mining is a great democratic choice for equality between all, security, anonymity and decentralization.
Exclusive CPU…lol
Because everyone has a Computer…That’s your logic…but that’s not logical. It’s not democratic to isolate and eliminate GPUs…which have been the genesis of proof of work. Which are also the most Democratic.
Exclusively CPU does not really exist unless you are suggesting burgerhash but that is not stable yet.
RandomX is not exclusively CPU so how do you propose to accomplish this?
Whereas there are options currently that more or less equalize 10$ of CPU hashes like 10$ of GPU or 10$ of FPGA.
To TylerX : GPUs are not the genesis of PoW. The genesis of PoW is CPUs, cf. the begining of the BTC mining (the genesis bloc and the others during months were mined exclusively with CPUs).
The CPU is the spirit of real PoW mining.
Exclusivity was never part of that concept. How would you go about achieving it?
Yes, CPU mining exclusivity is not formally in the concept of Bitcoin. The concept of Bitcoin comes from cypherpunk and the spirit is equality between every person. I think exclusive CPU use is the best way to achieve this.
How do you get there? I am not an expert in algo mining, but it seems there are some ways of reflexions with some other coins (XMG, maybe).
There is a GPU miner for that but since they do not really have a market other than Moondex at the moment that doesn’t matter overly.