Topic: Would smaller blocks reduce Bitcoin energy requirements? (Read 3836 times)

The current banks take several days to settle an international transfer. Bitcoin does it in minutes. Why is Bitcoin not fast enough for you?

I'd love to, but right at the moment I'm mega-stressed about Brexit, and the weak politicians who are trying to massage it. I'm trying to sort out some videos, and I've been given hosting for a radio station.

My vision is that Bitcoin becomes the "central bank" of crypto-currencies, and that minor transactions are moved onto sidechains or other cryptos. Bitcoin could then become a store of wealth, and it could be used for house purchases and other big ticket items.

You need a project to prove this, you're obviously very determined, you talk about this alot.

Set up a testnet, where you tweak the Bitcoin client's block interval and block reward rate. Invite people to join, and develop a way of benchmarking the performance based on orphan rates (including all the forms of orphaned blocks that you speak of in this thread).

I'd be interested to see the results.

My interest in this is that I feel that a 10 minute average block generation time is inherently flawed on the modern internet. At some stage in the future, Bitcoin will have to reduce this imho. I suspect it would be easier to do this now, than to try to change things when the network and use has expanded. I believe Bitcoin will continue to grow, and needs to consider this in more fundamental ways than creating larger slow trucks when faster vehicles are needed to speed up deliveries.

Nah. Mining hardware doesn't really care how fast blocks are, nor how big they are.

Ah! Thanks, I can see that.

So if the blocks were generated faster, but were not as full, would we be worse off in respect of the aborted block generations?

Reducing the difficulty does not reduce the number of hashes computed per minute.  It just reduces the average number of hashes per block.

If you reduced the difficulty enough for an average of 5 minutes between blocks, you would need half as many hashes per block, but you'd generate twice as many blocks per day.  The end result is just as many "aborted searches".

To reduce the time interval, you would have to reduce the difficulty as I understand it. Surely with a reduced difficulty, there would be fewer aborted searches.

You will not decrease the number of aborted block searches. If I understand correctly, you are referring to the miner changing the information in the merkle root of the block they are hashing after a block is found. Since more blocks are produced per time frame, they have to abandon their current hashing block more.

The mempool size would definitely be reduced. The size of the mempool can decrease significantly; just that merchants can start asking for more confirmations.

But would it decrease the number of aborted block searches, and the mempool processing overheads?

Lowering the time increases the wasted work, both with more overhead in blocks and in more orphaned blocks.

I'd be convinced to start CPU or GPU mining on a changed PoW, Jetcash. Others may just well do the same (mining was pretty popular amongst computing enthusiasts in the early days of CPU and GPU mining, and hence far more decentralised).

You're describing it as if miners are part of the Bitcoin Corporation, who cannot lose their position/role. No such concepts exist in Bitcoin, either you've got the right hardware and running costs, or you haven't. That's all it takes to be a Bitcoin miner.

This would seem to be a fairly radical change. If this happens, would it be possible to combine it with another radical change - reduce the blocksize to 500Kb, and the generatiion interval to one minute.

How can miners be convinced to switch to this if it makes their hardware worthless.

Probably not.

The "other proof of work formula" will still require that the vast majority of work be wasted, with just as many "orphaned and aborted" attemtps.  It will just require different hardware so that the current hardware can't be used to attack the network.

I see in some other threads that people are suggesting that core may switch to a new PoW formula - isn't this what I am asking for?

and mining becomes centralised
or users migrate to systems with cheaper rates and faster confirmation times

I'm not suggesting that we have a system that is less secure. With increasing interest from globalists and the banking elite, we need to have a more secure system if anything. I'm suggesting that it is worth considering the fact that it may be possible to develop a system which has fewer wasted hashes ( abortions as well as orphans). Hopefully this thread will spark some interest in the developers.

There is an incentive structure built into bitcoin so that mining is always exactly as efficient as it needs to be.

As more energy efficient hardware is created, miners have an incentive to use the more efficient equipment since it will reduce their electricity costs. Therefore, hash power will move to the more efficient hardware, so mining is more efficient.

If the exchange rate drops, then mining isn't as profitable. The most inefficient miners shut off their equipment and the difficulty is reduced.  Less hashes are needed, so mining is more efficient.

If the block reward is reduced, then mining isn't as profitable. The most inefficient miners shut off their equipment and the difficulty is reduced.  Less hashes are needed, so mining is more efficient.

If the exchange rate increases, then the network is more valuable and needs more security.  With the increased revenue less efficient miners can afford to run hashing equipment and difficulty is increased. Mining becomes a bit more inefficient to increase security.

If transactions pay higher fees, then the users value the network more. With the increased revenue less efficient miners can afford to run hashing equipment and difficulty is increased. Mining becomes a bit more inefficient to increase security.

This incentive structure is balanced so that the exchange rate determines how much hashing is needed, and the amount of hashing determines how expensive it is to attack the network, and the cost of attacking establishes the security of the blockchain, and the security of the blockchain contributes to its perceived value, and the perceived value drives demand, and demand determines the exchange rate and with that we are back to the beginning.

As such, more efficient mining overall (less "wasted" hashes") means a less secure blockchain.  So, before you can decide how many "wasted" hashes are acceptable, you first need to decide how much security is desired.

How many were there in the first year of Bitcoin, and how many will there be in 5 years time? You don't have to eliminate all of them, just make block creation more efficient.

I don't have any solutions. I just put it forward for better minds than mine to consider the problem. If it is not addressed, then this may prevent the future development of Bitcoin.

Using this definition there are, on average, 237,345,536,820,000,000,000 aborted blocks every 10 minutes at the current difficulty.  Eliminating these would eliminate the concept of proof-of-work and would eliminate the consensus mechanism built into bitcoin.  How would you make that work?


That "thrashing about" that you are complaining about is the reason that the consensus mechanism of bitcoin works.  What would you replace it with?


Such as?

Let's assume that I do understand the mis-named "proof of work". It's actually a proof of luck - it's a proof that you got lucky and found an eligible hash. You could also consider it as a proof of processing power, or a proof of cheaper electricity than any body else. It's like a lottery, the more tickets you buy, the more chances you have of winning, but of course, having a warehouse full of tickets doesn't guarantee a win.

With regards to the way mining has developed. It seems it has moved away from the original distributed concepts. and now it seems to depend on the size of your processing capability, and your proximity and connectioins with other miners. The advances in computing for Bitcoin mining have forced many people out of mining, and have increased the investment required to find blocks. This has increased the amount of "work" that has to be done to find a lucky ticket. Any "work" that does not result in a usable product is wasted in my opinion. I am aware that this non-productive work was intended to keep the system decentralised, but we are moving to a position where this is no longer the case.

All I am asking is that people recognise that aborted blocks are as much of a waste of Bitcoin resources as orphaned blocks. Just to clarify. by aborted blocks I meant the abandoning of an attempt to find a block creating hash before the completed block can be submitted for peer acceptance.

Whilst miners are thrashing about wasting their computing power, transactions are stagnating in a pool, and users are having to wait for extended periods to gain confirmation of their transactions. Lets try to make more profitable use of that discarded computer power.