Topic: Blocksize increase (or not) may ultimately be irrelevant (Read 624 times)
lol i'm the most ill experienced one in this thread, can some noob post something silly just to take a share of the shame here?
Others have pointed out that there may be an equilibrium where the marginal income from fees is less than the marginal cost of a potentially orphaned block (where the size is greater than 0).
I can think of a few issues related to that suggestion.
1. The subsidy has a big effect on this equilibrium.
2. Without increasing the max size, it may not be possible to find the equilibrium.
3. The equilibrium only exists if the probability of being orphaned increases faster than the size of the block.
I can think of a few issues related to that suggestion.
1. The subsidy has a big effect on this equilibrium.
2. Without increasing the max size, it may not be possible to find the equilibrium.
3. The equilibrium only exists if the probability of being orphaned increases faster than the size of the block.
I personally think the practical limit to block size is not the one in code, but the one derived from mining and propagation constraints.
Larger blocks are slower to propagate, take more time to prepare, and have a greater chance of being orphaned. As an illustration, SPV mining and 1 transactions blocks are already common.
IMHO that's where there is the greater risk in terms of bitcoin scalability: it is already risky (in terms of profit) to pack too many transactions in a block.
As block rewards go down, all low-fee tx should be filtered out by profit-oriented miners, and they should strive to make small blocks packed with high fee tx.
It just does not make direct economical sense (from a miner PoV) to have large blocks (higher risk of orphans) unless fees are sufficiently higher. Including low fee or zero fee transactions is basically a donation from the miners, a kind of modern faucet, it will likely go away as block rewards go down.
So block size may ultimately be irrelevant in terms of fees, propagation speed may be the name of the game (at least for profit-oriented pools).
Larger blocks are slower to propagate, take more time to prepare, and have a greater chance of being orphaned. As an illustration, SPV mining and 1 transactions blocks are already common.
IMHO that's where there is the greater risk in terms of bitcoin scalability: it is already risky (in terms of profit) to pack too many transactions in a block.
As block rewards go down, all low-fee tx should be filtered out by profit-oriented miners, and they should strive to make small blocks packed with high fee tx.
It just does not make direct economical sense (from a miner PoV) to have large blocks (higher risk of orphans) unless fees are sufficiently higher. Including low fee or zero fee transactions is basically a donation from the miners, a kind of modern faucet, it will likely go away as block rewards go down.
So block size may ultimately be irrelevant in terms of fees, propagation speed may be the name of the game (at least for profit-oriented pools).
I had the same thoughts about bigger blocks being orphaned faster. I mean we already see enough 1 transaction blocks. But it is not an issue at all. There is matt's relay network which makes it possible to propagate a full 1mb block in, i believe, 2 seconds so that it gets >50%. Risk of orphaning banned.
In fact it should happen very seldom anyway when you think about that one block is found every 10 minutes only.
At the end you can't restrict the amount of allowed restrictions and await that bitcoin can survive with it. It is simple as that. There must be a solution with bigger blocksize or no restriction at all.
While i agree that this reasoning is sound, we have to first define what Bitcoin is, is it a high value settlements instrument or do we wish to use it as actual money.
Yes, this is the area where intents (what we want to do with it) and game theory (how people will take selfish advantage of it) meet.
I see the current reward/fee structure as being only sustainable if fees rise significantly, which will limit adoption long before technological limits are reached. That's without bringing any kind of anonymity, privacy, fungibility or other *y terms into the debate.
Initially I thought block size increase could alleviate that, but block size is already limited by propagation and the risk of orphans, so larger block size, even if allow, would likely remain oddities.
In terms of raw technological scaling, a single blockchain is just not scalable for widespread micro-transactions: it would already take 10 MB-sized blocks to handle Starbucks sales f.i. (2 billion $ annual revenue, $4 per sell, that's 10k tx every 10 minutes if I did the math right), and that's just Starbucks...
That also gives another angle for the "how to reward full node operators?" debate, as nodes (or lack of, as in centralization) have a role wrt propagation times.
I personally think the practical limit to block size is not the one in code, but the one derived from mining and propagation constraints.
Larger blocks are slower to propagate, take more time to prepare, and have a greater chance of being orphaned. As an illustration, SPV mining and 1 transactions blocks are already common.
IMHO that's where there is the greater risk in terms of bitcoin scalability: it is already risky (in terms of profit) to pack too many transactions in a block.
As block rewards go down, all low-fee tx should be filtered out by profit-oriented miners, and they should strive to make small blocks packed with high fee tx.
It just does not make direct economical sense (from a miner PoV) to have large blocks (higher risk of orphans) unless fees are sufficiently higher. Including low fee or zero fee transactions is basically a donation from the miners, a kind of modern faucet, it will likely go away as block rewards go down.
So block size may ultimately be irrelevant in terms of fees, propagation speed may be the name of the game (at least for profit-oriented pools).
Larger blocks are slower to propagate, take more time to prepare, and have a greater chance of being orphaned. As an illustration, SPV mining and 1 transactions blocks are already common.
IMHO that's where there is the greater risk in terms of bitcoin scalability: it is already risky (in terms of profit) to pack too many transactions in a block.
As block rewards go down, all low-fee tx should be filtered out by profit-oriented miners, and they should strive to make small blocks packed with high fee tx.
It just does not make direct economical sense (from a miner PoV) to have large blocks (higher risk of orphans) unless fees are sufficiently higher. Including low fee or zero fee transactions is basically a donation from the miners, a kind of modern faucet, it will likely go away as block rewards go down.
So block size may ultimately be irrelevant in terms of fees, propagation speed may be the name of the game (at least for profit-oriented pools).
While i agree that this reasoning is sound, we have to first define what Bitcoin is, is it a high value settlements instrument or do we wish to use it as actual money. In the case of the later, then your logic would fail as high fees on micro transactions, or even a regular transaction would make the system prohibitive and not any better than ViSA and cohorts. The decision on block size limit must weigh in all these factors and more when the community decides the direction to support.
One would assume from reading what it used to be aimed at, that it was meant to be a monetary system capable of all things money is...without cutting out any particular section or class of society. The cost of maintaining the status quo is the millions of potential users and the massive adoption as a result.
Please note i do not advocate unreasonable numbers (such as the continuous doubling), i do however feel that a static limit of 20-40 MB would serve us well for at least fifty years and by then technology would have advanced far enough that 50Mb then will feel much like 10Kb does now.
I personally think the practical limit to block size is not the one in code, but the one derived from mining and propagation constraints.
Larger blocks are slower to propagate, take more time to prepare, and have a greater chance of being orphaned. As an illustration, SPV mining and 1 transactions blocks are already common.
IMHO that's where there is the greater risk in terms of bitcoin scalability: it is already risky (in terms of profit) to pack too many transactions in a block.
As block rewards go down, all low-fee tx should be filtered out by profit-oriented miners, and they should strive to make small blocks packed with high fee tx.
It just does not make direct economical sense (from a miner PoV) to have large blocks (higher risk of orphans) unless fees are sufficiently higher. Including low fee or zero fee transactions is basically a donation from the miners, a kind of modern faucet, it will likely go away as block rewards go down.
So block size may ultimately be irrelevant in terms of fees, propagation speed may be the name of the game (at least for profit-oriented pools).
Larger blocks are slower to propagate, take more time to prepare, and have a greater chance of being orphaned. As an illustration, SPV mining and 1 transactions blocks are already common.
IMHO that's where there is the greater risk in terms of bitcoin scalability: it is already risky (in terms of profit) to pack too many transactions in a block.
As block rewards go down, all low-fee tx should be filtered out by profit-oriented miners, and they should strive to make small blocks packed with high fee tx.
It just does not make direct economical sense (from a miner PoV) to have large blocks (higher risk of orphans) unless fees are sufficiently higher. Including low fee or zero fee transactions is basically a donation from the miners, a kind of modern faucet, it will likely go away as block rewards go down.
So block size may ultimately be irrelevant in terms of fees, propagation speed may be the name of the game (at least for profit-oriented pools).
Jump to: