[WIP] POC: Enable an equihash parameter hard fork at a fixed block height

[jc23424]

Here are roughly the changes that would be required to move to (144,5) equihash

[G5FoundationsL]

Please proceed with the hardfork to increase ASIC resistance. Mitigating Bitmains control of the Blockchain is imperative. Developers cannot allow a singular Chinese Company (Bitmain), to provide a majority of the blockchain power. This is a Strategic position that Bitmain and China want. In years to come if Bitmain and China power the worlds, "distributed and decentralized blockchains" there will no more decentralization. Bitmain and China will have sucessfully taken over, and there will be no going back. If they are allowed to power the majority then all projects are lost and so will the future of the Blockchain. Do not allow this to happen. Stop them!

[h4x3rotab]

Good move :)

[kornface13]

Thumbs up on this!

[ch4ot1c]

I disagree with reparameterizing the chain under pressures of new hardware.

[G5FoundationsL]

Its not just because of the new hardware. Its much bigger than that. The Chinese have a strategic vision. Make everyone buy their Mining Machines, and keep them in an infinite loop, where we, (Bitmain China), can power grab the entire blockchain. There is no direct issue with a new technology, however when one solitary governmet is on a Strategic course of action to dominate the worlds blockchain, you should worry, its a imminent threat to the survival of a decentralized Distributed Network.

[ch4ot1c]

ASIC miners will become more and more prominent as time goes on; fighting against this is futile. Decentralization can happen/always is present nonetheless; ASIC simply becomes the new status-quo. New and more manufacturers will compete.

Because of this, I wouldn't be opposed to an eventual change to SHA256, the dominant fast+lite ASIC-friendly hash algo. These proposed hardforking changes are largely a guess into the scalability of Equihash's ASIC-resistance, which will almost be certainly verified to be 'unscalable' in the future.

[G5FoundationsL]

Your missing the politial aspect to the entire argument and are focusing specifically on just the technical aspect. China is a powerhouse. They do not execute fair deals and have one interest. Themselves. They dont care about a decentralized distributed network. They want their machines powering the blockchain only.

Think about it from a Supply Chain perspective. Is it possible for other world manufactures to begin competing for ASIC production?

Remember Bitmain and China think long term strategy. Also, remember how the blockchain works. It takes 51% to attack. They build new machines that double the power every 2 years, if nobody can compete, they own everything. The only thing they currently dont own are GPU cards.

The danger is long term. Just look at what bitmain does. They secretly build a machine that makes everything obsolete overnight. Then people run out to buy them. Then they release something new thats faster, and you have no choice, but to buy their machine again. Then they drop the price by 70% after investing into the blockchain. Then they start over again. They baited you, now your in and you cant get out. 10 years later, they own the blockchain.

Read up on China's Trade deals. You will learn to appreciate how cunning they really are.

[jc23424]

I personally think that ASICs are fine and don't support this change - but consensus seems to desire this quality in the chain and that should dictate the course of action and so I made the PR.

But anyway - let's please keep discussion here focused on code review related comments only. (there are plenty of other forums to debate whether ASIC resistance is a desirable quality - Reddit, Discord, etc...)

[sound58]

I like that we are looking at moving ASIC resistant. Any thoughts towards switching to an algo that would be more resistant long term, like X16S. Would require less forks in the future.

[bitcartel]

@jc23424 Hi, would be great if you could provide attribution in the commit log message, say using a co-authored-by tag, or a link back to the original patch which I describe here: zcash/zcash#1211 (comment) Thank you. (I'm assuming you adapted the patch, if not, never mind!)

[jc23424]

Hi - actually I developed this independently, but am glad that we've roughly taken the same approach!
Anyway though happy to add the co-authored tag, will amend the commit.
Hope we can work together if any further changes are required