Topic: Pow change? (Read 331 times)

Monero's PoW change only success because majority of the community have same opinion and the ASIC for CryptoNight wasn't released for public when the core team decide to "tweak" their PoW to prevent ASIC usage.
So, while it's true that dev team have bigger power, that doesn't mean they can force change if majority have different opinion.

But PoW algorithm change on Bitcoin which kill ASIC wouldn't work (at least without conflict or/and chain-split) since Bitcoin only profitable if mined with ASIC and the developer know the risks of PoW change from security and politic sides.

Plus the whole history of CryptoNight and the sketchyness of the developers behind it is/was a big issue. From day one there's been constant drama between all the forks of Bytecoin.

What's amazing is that Bytecoin being the scam it is was even able to survive at all for a period.

Monero proved to me that a PoW change does work and the best dev team will survive.  People have an almost cultish devotion to a dev team and the software releases that that specific dev teams releases are considered gospel.  Couple this with the fact that the exchanges determine which fork is "real" by asking the "official" dev team which one the real fork is.  So you have a game completely rigged in favor of the accepted dev team.  And if this dev team determines a PoW change, there is nothing anyone can do to stop it even with a 51% attack.

The greatest threat to bitcoin is not ASICS, it's mining pool, the pool case centralization of bitcoin because one pool only need to maintain one full node(the blockchain).
Even if there is no Bitmain, you only have two choose: AMD or Nvidia, and still can not avoid mining pool, and only big pool can survival,A miner has no choose but join a big pool.
To change current POW algorithm to other algorithm can not solve ASICS problem unless the new algorithm can self evolution, other wise if still a fixed POW algorithm, new ASICS will be developed soon.

In regards to this part, they wouldn't survive. They would end up being collateral damage in the fight against Bitmain. Even good actors who haven't really done much wrong end up getting killed off in a PoW change. That's probably one of the major reasons that it hasn't really been close to being a viable option yet. If it were only Bitmain, then perhaps it would have happened already.

I used bitcoin cash only as an example of a coin with the same hashing algorithms as bitcoin.
The point of my post was:
if hardware used to profit securing bitcoin can be used to perform other tasks, like mining altcoins, the security of bitcoin is fucked, because an attack can be performed and hardware reused to mine altcoins, an attacker don't lose his hardware investment, his investment is safe because can be used to mine altcoin or playing videogame, or doing anything different from mine bitcoin.

I think pow security come from two root: hardware costs and power consumption. If one of those fails pow security gets reduced(not compromised, only reduced) , if both fails security is gone

powr consumption: is needed so a single block generation is expensive and costs grow exponentially for each block an attacker would like to change, securing immutability.

hardware costs: is needed so hardware is take as guarantee to ensure an attacker have no profit in harming bitcoin.

We all know energy is cheap, it is really cheap, some people steal energy mining for free, a government can choose to waste some natural resources to harm bitcoin,a  new fantascientific hardware generation can mine blocks in seconds without consuming power, and so on.
so hardware costs is what ensure a 51% is not and will not be profitable.

So if hardware can be used to perform other tasks hardware costs guarantee is lost.

Actually we are in a situation where hardware can be reused to mine an altcoin so bitcoin have lost hardware costs guarantee.(this is valid also from a bitcoincash point of view)

And from my point of view the actual situation is really risky.

surfing I have seen some people talking about the possibility of a pow change .

But my question, and I hope it is a BIG question, is:
Is a pow change realistically possible as last defense instrument? Which can be the best way of performing this kind of hardfork, without harming decentralization?


If you mine altcoin you are not securing bitcoin. Atomic swaps are nice, but having altcoins is part of the problem I've described before. Bitcoin is not only a coin.

That is a nice debate ...

Initially I would like to approach it more as an open question ..
From computer science point of view, I do understand why Satoshi picked SHA-256 ... it is an elegant algorithm simple and powerful enough that could be done by hand

Mining Bitcoin with pencil and paper: 0.67 hashes per day
http://www.righto.com/2014/09/mining-bitcoin-with-pencil-and-paper.html

So build a machine to solve it should also be simple, therefore cheap ...
From the Social Science point of view, then things can become complicate very easy because we going to deal with problems like monopoly, wast of energy, unfair competition, government subsidies, legislation, etc ...
If from Social Science point of view wasn't problems enough,then we can start thinking about ethics ...
Regards to that subject I would like to make an analogy to the ethical problem of build a general propose AI or a machine that we cannot switch off once it is switched on..
For my analogy I would like to borrow from Computerphile a topic called "General AI Won't Want You To Fix its Code" ( https://youtu.be/4l7Is6vOAOA )
The idea is the following ... A hypothetical machine called "stamp collector" was built and one of the main functions is to collect as much stamps as possible in the most efficient way ... but at some point that algorithm in order to reach it goal start acting against its creator interests ... then the question is ..

Can we press the STOP button ? Will the machine respond "i am sorry Dave I can't do that"
  
Regards to more specific questions to our community ...
I am thinking about Atomic Swaps, you can mine on whatever algorithm you think it is fair for your goals than exchange for bitcoin

https://en.bitcoin.it/wiki/Atomic_cross-chain_trading

Algorithm change is very risky for Bitcoin because it'll require a hard fork. In the event of a contentious fork, the attack will increase dramatically.

As @aantonop says: "If a fork happens, we get to learn what happens when a fork attack occurs. Anyone who thinks a fork will be unopposed is quickly going to discover that this will be a battle on all fronts. Inevitably the two sides are going to attack each other on the network, with denial-of-service attacks, with the hash rate; they're going to attack each other publicly, privately, anonymously and not. Every bug in the software will get poked and poked again, so they better fix them well. The battle becomes "who has the best software development team? How quickly can they maintain that code and keep uptime?" That race is only 7 blocks wide. I don't think the people who are threatening to do a hard fork have thought clearly about the implications. This will allow Bitcoin to test all the attack vectors (nodes, relays, hashing, replay transactions, etc)."

source: https://www.youtube.com/watch?v=rZi86_ovB3Y

I think the better way to think is to decentralize the ASIC, to make miner's access to ASIC become easier. So we can buy ASIC as easy as buying a hard drive.

I'm ok with asics, if there are a lot of producers and different miners asics is ok, better than gpu where you only have 2 or 3 producers.
start from scratch to produce a new sha256 miner is a lot simpler than start from scratch to produce a new gpu.
also asics work better securing the network because they have no other uses so you can resell them if you succesfully attack the network.
so with asics there is a bigger economic incentive to be good and not harm bitcoin.
Now we have b-cash so asics can be reused, and thats not so good.
So a pow change can effectively be a solution to stop this problem.

but the bigger problem for me is we have to not harm too much miners. because it can harm decentralization a lot. as you say bitfury will die for shure, canaan will die with high probably, halong miner will probably die, all not owned by bitmain mining farm will probably die. Bitmain will probably survive, poor but alive. and with a pow change they have no concurrency to fight with. So end of decentralization.

How a pow change can solve this situation in a good way?

I wish only a lot of people start posting on bcash social networks so blockchain will be spammed and bcash attack will end and no pow change will be needed.

a POW change with wide support would be extremely detrimental to any industrial miner because they've already made significant investments into SHA-256 mining. last year, bitfury george made open threats of lawsuits against those promoting a POW change: https://www.reddit.com/r/Bitcoin/comments/60ast5/bitfury_george_pledges_to_sue_all_involved_with/

so while a POW change may target bitmain, it would also probably decimate most of today's mining outfits.

i also think a POW change doesn't solve this problem. nothing stops this exact same scenario (monopoly in ASIC chip manufacturing and high concentration of mining power) from happening again.

hi all,
I'm reading some respectable people talking about a pow change to kill bitmain monopoly.
I have no opinion about bitmain, if they are good or bad now is not important for the scope of this post.
Bitmain proved a lot of time to fast and efficent producing asics for a lot of algorithms, so I hope a pow change will only kill the little concurrency they have.
I can't figure how  canaan, or bitfury, or others producers and all others miners can survive a pow change better than bitmain.
So I'm looking for a suggestion to understand how a pow change can be a solution and not an harm to mining.