How viable is PoW for new coin?

[ABCbits]

If you check top PoW coins[1], many of those has been around for long time. Many new coins these days use seems to avoid PoW since they claim it's trivial to perform 51% attack by renting CPU, GPU or ASIC. Meanwhile, some people claim different algorithm such as PoS make the rich (usually developer and early adopter) become richer with less risk involved. So how viable is PoW for new coin?

[1] https://www.coingecko.com/en/categories/proof-of-work-pow

[PX-Z]

Well, PoS has a similar and possible way to attack like the 51% attack on PoW. So, it doesn't make sense that it's their own reason to avoid implementing the PoW. Besides that, it's much easier to do "stake grinding" than a 51% attack on PoW. Hardwares such as ASIC costs thousands of dollars on a single device, or let's say it just uses GPU instead of ASIC. The maintenance, technical knowledge, electricity cost, and cons if the attack doesn't work are just the reasons why it needs many resources to do a 51% attack. In simple words, it's not worth it since there will be a hardfork that will happen after the attack that could make the attack irrelevant.

While in stake grinding, the bold part above are missing, and the attackers just need financial investments that can have more than 50% of the total staked token/coins for that to happen.

All I can say is the primary reason behind new coins using PoS is related to its cost. Low cost to become a validator, this includes the energy of "mining" (staking) which connects to it's security (more validator more secured on its own way), then the possibility of having more transactions included per block, in short scalability, which Bitcoin really needs. As of now it only relies on layered solutions to scale such as lightning and liquid, segwit are not much of a help when transaction numbers rose to hundreds of thousands, you could send a transaction with low fee but it doesn't mean it won't get stuck for days which is the main problem that faced by many.

[bitmover]

Well, PoS has a similar and possible way to attack like the 51% attack on PoW. So, it doesn't make sense that it's their own reason to avoid implementing the PoW. Besides that, it's much easier to do "stake grinding" than a 51% attack on PoW. Hardwares such as ASIC costs thousands of dollars on a single device, or let's say it just uses GPU instead of ASIC. The maintenance, technical knowledge, electricity cost, and cons if the attack doesn't work are just the reasons why it needs many resources to do a 51% attack. In simple words, it's not worth it since there will be a hardfork that will happen after the attack that could make the attack irrelevant.

It depends on the cost of the attack.

There were times in the past when bitcoin forks where being 51% attacked.
People send money to exchanges, withdrawal and then performed the attack.

The same group made this to several coins in 2019/20.

I can't remember exactly,  but this is one of the attacks. Later on, many exchanges dislisted those coins

https://cointelegraph.com/news/bitcoin-gold-blockchain-hit-by-51-attack-leading-to-70k-double-spend

This is another one of bsv on 2021
https://www.forbes.com/sites/billybambrough/2021/08/04/bitcoin-fork-suffers-massive-51-attack-in-attempt-to-destroy-the-cryptocurrency-sending-its-price-sharply-lower/?sh=781e60bf248f

[Stompix]

While in stake grinding, the bold part above are missing, and the attackers just need financial investments that can have more than 50% of the total staked token/coins for that to happen.

It's all about the $.
Asics, work, electricity, acquiring coins for staking it all translates into $.
PoW coins have been attacked for hundreds of times, bitcoin forks, Ethereum forks, it's all about the amount required to launch such a thing.
What is simple to do right now , acquire 4000 S21 at a price of 20 millions and put BSV out of its misery or get 51% on Ethereum, which even if the price doesn't go up as you buy you would till need around 20-30 billion at current staking's! And things have become even murkier, you want to launch a POW attack, well you don't have to go though all that trouble, you could simply buy directly farms, ready to go, not that much difference!
Look how Mara bought 2 mining farms from HUT8 like normal people trade pokemon cards.

It's all about the $ involved, $ can buy you eveything as long as you have printing press in the basement..

[TomPluz]

All I can say is the primary reason behind new coins using PoS is related to its cost. Low cost to become a validator, this includes the energy of "mining" (staking) which connects to it's security (more validator more secured on its own way), then the possibility of having more transactions included per block, in short scalability, which Bitcoin really needs. As of now it only relies on layered solutions to scale such as lightning and liquid, segwit are not much of a help when transaction numbers rose to hundreds of thousands, you could send a transaction with low fee but it doesn't mean it won't get stuck for days which is the main problem that faced by many.

Must be the reasons why Ethereum decided to convert from PoW to PoS. Scalability is always the big question that a popular coins like Bitcoin and Ethereum have to deal with into the future. As more and more users are getting in and as the coin is getting into the mainstream, there can be a big challenge on congestion and this is especially true with Bitcoin even up to now. I recently experienced this a week ago...BTC sent to me was already confirmed by the network but it never arrived yet on my wallet so I am still waiting for it days after. Aside from scalability, a popular coins has to deal with the question of energy use which we know PoW is really energy intensive and if your coin is one then expect the scrutiny of many especially those who are considered to be Earth warriors or working for climate change. We know how BTC came under fire on this concern as its mining really consumed huge amount of energy even comparable to what some countries are consuming. Now, another area that many can be concerned with PoS is the decentralization question and whether this can be valid or not we are witnessing things right now in Ethereum so let's wait for years to come to see whether we should fear this or just another pigment of our imagination. Overall, we can see that both PoW and PoW got their own advantages and disadvantages as nothing can really be perfect only the fittest to survive just like the jungle of life.

[ABCbits]

All I can say is the primary reason behind new coins using PoS is related to its cost. Low cost to become a validator, this includes the energy of "mining" (staking) which connects to it's security (more validator more secured on its own way), then the possibility of having more transactions included per block, in short scalability, which Bitcoin really needs. As of now it only relies on layered solutions to scale such as lightning and liquid, segwit are not much of a help when transaction numbers rose to hundreds of thousands, you could send a transaction with low fee but it doesn't mean it won't get stuck for days which is the main problem that faced by many.

Must be the reasons why Ethereum decided to convert from PoW to PoS. Scalability is always the big question that a popular coins like Bitcoin and Ethereum have to deal with into the future. As more and more users are getting in and as the coin is getting into the mainstream, there can be a big challenge on congestion and this is especially true with Bitcoin even up to now. I recently experienced this a week ago...BTC sent to me was already confirmed by the network but it never arrived yet on my wallet so I am still waiting for it days after. Aside from scalability,

--snip--

Sorry, but i don't see correlation between either PoW/PoS and scalability. If Bitcoin community wish for bigger block size and more actively use both LN and sidechains, there'd be less congestion problem.

[Stompix]

Must be the reasons why Ethereum decided to convert from PoW to PoS. Scalability is always the big question that a popular coins like Bitcoin and Ethereum have to deal with into the future. As more and more users are getting in and as the coin is getting into the mainstream, there can be a big challenge on congestion and this is especially true with Bitcoin even up to now.

There is no limit on scalability because only of PoW or PoS, Doge is Pow and is outperforming Ethereum continuously terms of tractions per second at a fraction of the costs, the algorithms have no say in the capacity of the network.
For sure you won't be able to solo mine and hope to propagate a block if it's 1GB in size but the same will happen even if you stake a coin and do the same solo node not on a cloud or sever node with some ISP, scalability affects in some aspects decentralization, the best example being Solana validators, but nothing more but a PoW vs PoS bitcoin will have the same issues at the same capacity.

[bitmover]

Must be the reasons why Ethereum decided to convert from PoW to PoS. Scalability is always the big question that a popular coins like Bitcoin and Ethereum have to deal with into the future. As more and more users are getting in and as the coin is getting into the mainstream, there can be a big challenge on congestion and this is especially true with Bitcoin even up to now.

There is no limit on scalability because only of PoW or PoS, Doge is Pow and is outperforming Ethereum continuously terms of tractions per second at a fraction of the costs, the algorithms have no say in the capacity of the network.
For sure you won't be able to solo mine and hope to propagate a block if it's 1GB in size but the same will happen even if you stake a coin and do the same solo node not on a cloud or sever node with some ISP, scalability affects in some aspects decentralization, the best example being Solana validators, but nothing more but a PoW vs PoS bitcoin will have the same issues at the same capacity.

There point is that there is something called Blockchain trillema (I think the term was created by vitalik buterin)



You can scale to infinite, but not for free. There is a downside to security and decentralization if you do so.

[garlonicon]

Many new coins these days use seems to avoid PoW since they claim it's trivial to perform 51% attack by renting CPU, GPU or ASIC.

To avoid that problem, you can always track the strongest chain of Proof of Work headers in existence in SPV way. You don't have to follow transactions from other coins. But if you follow their hashrate, then you can determine the real Proof of Work in the world, for a given mining algorithm, and then adjust your network parameters accordingly.

Which means, that if you want to create for example some new altcoin, which could be mined with double SHA-256, then you have to receive 80-byte block headers from other double SHA-256 chains, just to calculate the difficulty in the right way. And the same is true for every other mining algorithm: if it exists, and is used anywhere else, then you cannot pretend that "whatever, blocks from other chains are invalid in my chain, so I won't count them, when I calculate my difficulty".

Bitcoin was created in the way it was, because there were no altcoins, which used Proof of Work before. Hashcash was used only as a spam protection, and it was based on SHA-1, and not double SHA-256. It used different algorithm, and there was no danger of "being reorged by hashcash miners". Which means, that the initial design didn't consider any other chains, so copy-pasting it 1:1 doesn't work, because if you start from scratch, you compete with existing huge networks (and Bitcoin had no such competition, so that design decision was good enough then, but shouldn't be repeated now).

But today, the situation is different. Today, if you want to make a successful Proof of Work altcoin, then you should count Proof of Work from other chains, because if you don't, then you will have the same problems, as for example NameCoin. If you have Merged Mining, then it can help a bit, but still, the design of your altcoin should guarantee, that you can reorg it, only if you can 51% attack the strongest hashrate with your algorithm, and not just reach 51% on your chain (also, that mistake is the very reason, why NameCoin was 51% attacked: that kind of attack, should be as hard, as doing it on Bitcoin, and to achieve that, it is needed to receive every block header, no matter what content is behind it, and count it to calculate "the real total difficulty", and not just "the local difficulty").

As Satoshi said:

I think it would be possible for BitDNS to be a completely separate network and separate block chain, yet share CPU power with Bitcoin.  The only overlap is to make it so miners can search for proof-of-work for both networks simultaneously.

The networks wouldn't need any coordination.  Miners would subscribe to both networks in parallel.  They would scan SHA such that if they get a hit, they potentially solve both at once.  A solution may be for just one of the networks if one network has a lower difficulty.

I think an external miner could call getwork on both programs and combine the work.  Maybe call Bitcoin, get work from it, hand it to BitDNS getwork to combine into a combined work.

Instead of fragmentation, networks share and augment each other's total CPU power.  This would solve the problem that if there are multiple networks, they are a danger to each other if the available CPU power gangs up on one.  Instead, all networks in the world would share combined CPU power, increasing the total strength.  It would make it easier for small networks to get started by tapping into a ready base of miners.

So, to achieve a successful Proof of Work altcoin, it is needed to fullfill that sentence: "Instead of fragmentation, networks share and augment each other's total CPU power". And the reason, why many altcoins failed, is that nobody designed it in such way. And as people saw, that for example NameCoin can be 51% attacked, they gave up the idea of Merged Mining, without realizing exactly, where it failed, how it failed, and what exactly NameCoin did wrong (they simply ignored Bitcoin headers, which didn't contain "NameCoin commitment", but they should instead blindly accept it, and use those headers to properly increase their difficulty, decrease their coinbase transaction, or perform similar actions, to measure "amount of coins per difficulty" properly).