...

[turtlehurricane]

....

[1714040029]

1714040029

[rebuilder]

re: average block size - wouldn't it make sense to look at the growth rate as well? Zoom out a bit on the chart and see how it looks:
https://blockchain.info/charts/avg-block-size?timespan=all&showDataPoints=false&daysAverageString=1&show_header=true&scale=0&address=

[Amph]

If fees ever do get too crazy I'd support a block size limit increase too, but I don't expect that to happen for over a decade if ever.

i think the point is that we cannot wait that it become necessary to increase the blocksize, to actually increase it, we must forecast this, before it happen

if it happen next year what you will do, proceed to fork anyway and built even a bigger mess than what we are facing right now?

if it must be done at some point, it's better to do it early, it's always like that for every thing, not only bitcoin

[rebuilder]

Note that alot of the spike recently was a stress test, it just dropped back down to where it should be under 0.4 mb. Still growing at a good clip but will be years if you extrapolate it not including the spam attack/stress test.

Sure. Disregarding the spike, even in the chart you posted, you can see the average block size has nearly doubled in one year. If that continues, we'll hit 0.8mb per block in a year or so, and that's assuming a steady growth rate. A year or so to start hitting the 1 mb ceiling seems like a safe estimate if you want to figure out what to do about it before you find out experimentally, like it or not.

[lottery248]

Note that alot of the spike recently was a stress test, it just dropped back down to where it should be under 0.4 mb. Still growing at a good clip but will be years if you extrapolate it not including the spam attack/stress test.

Sure. Disregarding the spike, even in the chart you posted, you can see the average block size has nearly doubled in one year. If that continues, we'll hit 0.8mb per block in a year or so, and that's assuming a steady growth rate. A year or so to start hitting the 1 mb ceiling seems like a safe estimate if you want to figure out what to do about it before you find out experimentally, like it or not.

but it doesn't mean it would double every year, unless the number active wallet grew explicitly fast, it would be impossible to double the block size per year. plus, i am tired to see most of the junk transactions in the blockchain.

[rebuilder]

but it doesn't mean it would double every year, unless the number active wallet grew explicitly fast, it would be impossible to double the block size per year. plus, i am tired to see most of the junk transactions in the blockchain.

Not every year, no. Next year, though? I don't think you can rely on that not happening.

[lottery248]

but it doesn't mean it would double every year, unless the number active wallet grew explicitly fast, it would be impossible to double the block size per year. plus, i am tired to see most of the junk transactions in the blockchain.

Not every year, no. Next year, though? I don't think you can rely on that not happening.

i mean, even though it would happen, i still would not let the blockchain to increase the size which is unneeded. that would harm newbies to use the full node wallet. doubling the block size limit would ever encourage the spammers to spam in the blockchain network if the active users did not increase. that's shit-awful with those junk transactions.
we need a sustainable currency to help bitcoin mingle with the world , just like a currency joining to the IMF.

[rebuilder]

i mean, even though it would happen, i still would not let the blockchain to increase the size which is unneeded. that would harm newbies to use the full node wallet. doubling the block size limit would ever encourage the spammers to spam in the blockchain network if the active users did not increase. that's shit-awful with those junk transactions.
we need a sustainable currency to help bitcoin mingle with the world , just like a currency joining to the IMF.

Honestly, I think Joe and Jill Average running full nodes is a pipe dream, whatever we do with the block size:

-If you increase the block size indefinitely, the storage and bandwidth requirements will become high enough to deter casual users.

-If you don't increase the block size but increase fees instead, the fees will become high enough to push casual users off the blockchain, to use payment processors etc. instead. This leaves them with no reason to run a full node even if they technically could.

-either way, I don't see many apart from a hard core of enthusiasts wanting to run a node.

[Moonpig]

If fees ever do get too crazy I'd support a block size limit increase too, but I don't expect that to happen for over a decade if ever.

i think the point is that we cannot wait that it become necessary to increase the blocksize, to actually increase it, we must forecast this, before it happen

I agree. It will become a problem at one point and it obviously needs to be raised, it's just to what and and when that people can't seem to agree on. I don't think it needs raising it 8 though.

[galbros]

3) Once we hit the 1 mb limit free and nearly no fee transactions will be forced off chain. There is TONS of dust, spam, and gambling that can be done off chain. Over half of all Bitcoin transactions right now are basically dusty junk. This will make plenty of room for legitimate transactions.

4) Once almost all the junk is forced off chain the rise in fees will be very gradual, and it will take a long time for the fees to become an issue for people. In any case the blockchain will function perfectly fine regardless of transaction data volume, fees will simply rise.

If fees ever do get too crazy I'd support a block size limit increase too, but I don't expect that to happen for over a decade if ever.

5) Hard fork of Bitcoin is dangerous and can result in mass confusion, double spends, and loss of Bitcoins. It will damage the value of Bitcoin at least temporarily.

6) We need to maintain a group of scientists which reach a consensus, not just centralize Bitcoin under a couple of developers.

Feel free to add if I missed any.

Thanks for this.  I appreciated your summary as I have not seen a good one elsewhere.  (If someone can point me to another one I'd appreciate it.)

I don't think increasing fees is a very attractive selling point for bitcoin adoption.  The block rewards becoming very small is very far out in the future.  As even your chart points out, the blocksize is much more of an immediate issue.

I'm afraid you can't really force your #4, moving people off chain, though I agree there is a lot of dust transactions, not sure why though, it is all satoshi dice, faucets?  I fear we are close to having #5 already.

Totally agree on #6.

Thanks again.

[DooMAD]

After 5 pages in your last thread, which is basically a repeat of the same flawed premise, I can't help but notice you still haven't answered my question, so I'll pose it again here:

Also this is probably something you haven't considered: As miner rewards approach zero transaction fees will be the only incentive to mine and keep the network secure. We need fees to rise in the longterm to keep Bitcoin secure. Increasing block size totally ruins this and could be catastrophic for the network in the long run.

We've considered that just fine, thanks.  What you're not considering is the following:

Assume Network 'A' can process a maximum of 2000 transactions per block and the average fee is .0001

Assume Network 'B' can process a maximum of 10000 transactions per block and the average fee is .00005

Which network can potentially generate a higher amount in fees?  That's right, the one that can support more transactions.  Raising fees is not the most efficient way to increase mining revenue.

we need fees to increase up to 0.50 $ , 1 $ - the poor people on the planet can use something else!

 
/s

As miner rewards approach zero....in 2140  

And also this.  There's a moral argument to go along with the one where small-blockians aren't very good as simple math (more transactions equals more fees, not less, it isn't rocket science).  The more people who are able to use Bitcoin, the stronger it becomes.  What some people think of as a "lesser" transaction could mean a not-insignificant amount of money for those in other parts of the world.

Please use your eyes and read this: Bitcoin transaction fees go up when transaction data volume goes up, the blockchain functions perfectly normal regardless of transaction # or size. This is what the Bitcoin core developers already agree on.

Yes, we will clearly continue to be able to fit 10 minutes worth of transactions from a global populace into two-thirds of a floppy disk for the rest of forever.  

Fees alone will not solve scalability.  If you honestly believe that then you don't understand this system as well as you claim to.  

Here's an honest question to those who advocate keeping a 1MB block:  how long would the backlog of transactions have to get before increasing the blocksize would become a viable option in your opinion?  I genuinely want to know just how far you're prepared to cripple the network to justify your agenda.

Under organic circumstances the Bitcoin transactions per day will increase slowly, and a slight increase in transaction fees will keep things running smoothly. Currently we are nowhere needing a blocksize increase, and Bitcoin transactions per day are growing steadily but definitely not exponentially.

I love Bitcoin so of course if there was a problem I'd support the fix. There simply is no problem though. Spam attacks helped prove that, transactions increased by insane amounts that would never happen naturally and network still ran fine.

Perhaps there was some ambiguity in my post.  That's not what I asked.  Those recent transactions may well have been spam, but if the network did grow to a point where it was filled by legitimate transactions, what figure in MB would you be willing to see the backlog reach before supporting larger blocks?

Still waiting for an answer on this one.  Be honest.

[DooMAD]

If blocks aren't filled then free transactions get confirmed, I already posted that last time you asked. Literally everyone can pay nothing and get in a block anyways, usually within 5 or so blocks.  

It's supply and demand. If we flood the transaction market with too much availability then the miners get screwed. It doesn't matter right now but in the future it will be a big deal.

4 years per halving so this isn't all that far into the future. We need transaction fees to be alot more than they are now within 5 years.

So what you're telling me is you're going to keep repeating the same thing over and over because you don't have an actual reply to a question I'm starting to think you haven't understood?

Try working under the assumption you're wrong about "decades away" (because I'm pretty sure you are).  If all the transactions you (wrongly) call spam were already forced off-chain and we frequently have enough "legitimate" transactions to fill a 1MB block, how large would the backlog have to be in megabytes before you would support an increase in blocksize?  

If you are incapable of answering this simple question after numerous attempts, I have to assume your brain isn't functioning correctly.  Please give me a number.

[agath]

1) Bitcoin block rewards will eventually become infinitesimal, at that point transaction fees will be the only rewards miners get. Right now transaction fees are around 0.1-0.5 BTC per block, which is nowhere near enough funding to secure the network by itself. We need transaction fees to go up, and the only thing that will force them up is the blocksize limit. Increasing block size might cause fundamental damage to Bitcoin due to this.

When miners will need funds to secure the network, they won't include too cheap transactions. Who wants their transaction to be included in a block, will need to increase the fee. Competition between miners will allow to discover what is the right fee. If now fees are low, it's just because the new generated coins cover happily all the costs. Miners are not stupid, let them do their work. If you think you can do better, start your own pool.

2) Currently average block size is less than 0.4 mb now that the stress test bullshit is over. Gonna be a long time until we even hit the 1 mb limit.

https://blockexplorer.com/blocks-date/2015-08-16

It seems that in some periods of the day we are already approachin the limit. When would you wait to increase the block size? Your reasoning is like closing the fence door after all the pigs are already escaped. I remind you that the white paper didn't have a block size limit and it has been introduced only for testing purposes, with the intention to remove it already in mind.

3) Once we hit the 1 mb limit free and nearly no fee transactions will be forced off chain. There is TONS of dust, spam, and gambling that can be done off chain. Over half of all Bitcoin transactions right now are basically dusty junk. This will make plenty of room for legitimate transactions.

When other off-chain technologies will be able to offer instant payments, you will not pay for your coffee with the blockchain, even if the fee is 0. And at the same time, if the block size is large, legitimate transaction will always have theri place.

4) Once almost all the junk is forced off chain the rise in fees will be very gradual, and it will take a long time for the fees to become an issue for people. In any case the blockchain will function perfectly fine regardless of transaction data volume, fees will simply rise.

I don't see anything useful on the rise of the fees. What makes bitcoin special is the adoption. Small blocks means smaller adoption. Small blocks are the best way to lead bitcoin to failure.

If fees ever do get too crazy I'd support a block size limit increase too, but I don't expect that to happen for over a decade if ever.

Just remove the limit and let the miner decide.

5) Hard fork of Bitcoin is dangerous and can result in mass confusion, double spends, and loss of Bitcoins. It will damage the value of Bitcoin at least temporarily.

The hardfork happens when the largest part of the network agrees on the new version. At that point there is no confusion.

6) We need to maintain a group of scientists which reach a consensus, not just centralize Bitcoin under a couple of developers.

If you don't want to centralize bitcoin under a couple of developers, allow for its extension.

Please stop the fud.


People: if you love bitcoin, install XT. I already did.

[Zz]

I agree all 6 reasons here. We should get rid of that dust transactions, so we won't need any larger blocksize for a long time. Transaction fees are already low, just double it and we don't see half of the dust.

[DooMAD]

If blocks aren't filled then free transactions get confirmed, I already posted that last time you asked. Literally everyone can pay nothing and get in a block anyways, usually within 5 or so blocks.  

It's supply and demand. If we flood the transaction market with too much availability then the miners get screwed. It doesn't matter right now but in the future it will be a big deal.

4 years per halving so this isn't all that far into the future. We need transaction fees to be alot more than they are now within 5 years.

So what you're telling me is you're going to keep repeating the same thing over and over because you don't have an actual reply to a question I'm starting to think you haven't understood?

Try working under the assumption you're wrong about "decades away" (because I'm pretty sure you are).  If all the transactions you (wrongly) call spam were already forced off-chain and we frequently have enough "legitimate" transactions to fill a 1MB block, how large would the backlog have to be in megabytes before you would support an increase in blocksize?  

If you are incapable of answering this simple question after numerous attempts, I have to assume your brain isn't functioning correctly.  Please give me a number.

You keep changing the question, but fine here is an answer to this new question.

There will be little backlog ever. Transaction fees increase so transactions decrease, just about equal and opposite forces so no backlog develops.

The question remained identical.  And you still haven't given me a number.  If you honestly believe a global populace can be contained by two-thirds of a floppy disk for another decade, I'm sorry, but there's no hope for what little credibility you have left.  I can only assume you'll happily sit back and watch while the user experience degrades to a point where the forums are packed with complaints about Bitcoin being so slow and unreliable that it's no longer of any use.  Your precious chart shows what's happening now, not what will happen in future.  You cannot predict how large the userbase will become, or how rapidly it might grow.  If Bitcoin sees a surge in adoption and the network can't support those transactions, your words here will come back to haunt you.

[balu2]

Great summary OP.

The next thing is:
Increasing blocksize will also make BTC inaccessible to half of the world - especially the unbanked in developing countries (all of africa, most of south america, most of asia, most of pacific region) due to bandwidth.

... and bigger blocks lead to more centralisation... not even talking about the risks of a meltdown in marketcap in case of a controversial hardfork ...

It's really all not even rational... and i tell you right now: it's not about bigger blocks; it's about a hostile takeover

This whole blocksize-increase-debate is irrational at so many levels and in so many points. Why is bloating the chain ad infinitum even an option for longterm growth? The chain is already so large most people don't download it and downloading the chain is a major hurdle for people to enter the btc ecosystem. Why even raise the bar for people for adoption ahead of demand like you point out?
All this is not rational and because it's not rational we know it's not about the tps or blocksize, it's about getting people to accept their alternative xt client to take over bitcoin entirely. Basically the blocksize drama is an attack on bitcoin by Gavin and Mike.

They thought people were stupid and would go along but their populism campaign in social media failed (though they are still trying hard). It's politics, not technics.

[pedrog]

"We need transaction fees to go up, and the only thing that will force them up is the blocksize limit."

We don't need fees to go up, cheap transactions is one of the major points of using bitcoin.

This is an argument for increasing the block size.

"Once we hit the 1 mb limit free and nearly no fee transactions will be forced off chain. There is TONS of dust, spam, and gambling that can be done off chain. Over half of all Bitcoin transactions right now are basically dusty junk. This will make plenty of room for legitimate transactions."

Decentralization is one of the major points of using bitcoin, off-chain transactions are centralized, again, this is an argument for increasing block size.

Don't know what a "non legitimate transaction" is, as far as know, if it follows the network rules it's "legitimate".

[balu2]

"We need transaction fees to go up, and the only thing that will force them up is the blocksize limit."

We don't need fees to go up, cheap transactions is one of the major points of using bitcoin.

This is an argument for increasing the block size.

Tradeoff with accessibility issue ... would you rather let everyone pay 5 cent for a tx or would you rather exclude half the planet from using it?

And who is mining for free? Raising max txs ahead of demand will weaken the network in the future aswell as exclude billions of users and push centralisation.

Arguing for a contentious hardfork, hostile takeover of btc by Gavin/Mike and all that is really a lost cause...

[pedrog]

"We need transaction fees to go up, and the only thing that will force them up is the blocksize limit."

We don't need fees to go up, cheap transactions is one of the major points of using bitcoin.

This is an argument for increasing the block size.

Tradeoff with accessibility issue ... would you rather let everyone pay 5 cent for a tx or would you rather exclude half the planet from using it?

And who is mining for free? Raising max txs ahead of demand will weaken the network in the future aswell as exclude billions of users and push centralisation.

Arguing for a contentious hardfork, hostile takeover of btc by Gavin/Mike and all that is really a lost cause...

Then we need to increase transactions on-chain to generate more fee revenue, not the other way around, off-chain transactions are the centralized ones, unless you have a real proposal for decentralized off-chain transactions...

If users choose XT, XT it is!

And because of bitcoin network nature changes need to be planed way ahead.

[balu2]

"We need transaction fees to go up, and the only thing that will force them up is the blocksize limit."

We don't need fees to go up, cheap transactions is one of the major points of using bitcoin.

This is an argument for increasing the block size.

Tradeoff with accessibility issue ... would you rather let everyone pay 5 cent for a tx or would you rather exclude half the planet from using it?

And who is mining for free? Raising max txs ahead of demand will weaken the network in the future aswell as exclude billions of users and push centralisation.

Arguing for a contentious hardfork, hostile takeover of btc by Gavin/Mike and all that is really a lost cause...

Then we need to increase transactions on-chain to generate more fee revenue, not the other way around, off-chain transactions are the centralized ones, unless you have a real proposal for decentralized off-chain transactions...

If users choose XT, XT it is!

And because of bitcoin network nature changes need to be planed way ahead.

YOU don't CREATE transactions. They occure or they don't. Bloating the chain won't make more txs magically appear - well, maybe spam and dust-junk increases. But when you want a bloated chain full of junk, why would you want to tarnish bitcoin for that?  

Increased spam and junk is not genuine growth.