The Barry Silbert segwit2x agreement with >80% miner support. - page 56.

d5000

legendary

Activity: 3906

Merit: 6249

Decentralization Maximalist

Quote from: dinofelis on June 03, 2017, 11:44:49 PM

I wouldn't touch LN with a stick on a hard-limited block chain where you have no guarantee of settlement. The LN is an entirely different way of doing transactions that has never been experimented, nobody knows if *in practice* it will take off.

I also think it is a little bit overrated (for me it will never replace on-chain transactions entirely), but in principle it can be a great micropayments tool. That's also why my answer to ComputerGenie (who preferred to omit a response to my answer) was cautious, I don't say on-chain volume will go down after a successful LN implementation, but on-chain volume would grow a bit slower because LN would be the main tool for transactions of e.g. $5 or $20 and so a part of the pressure is taken from the "transactions market".

dinofelis

hero member

Activity: 770

Merit: 629

Quote from: d5000 on June 03, 2017, 04:28:29 PM

On blockchain.info you can also find charts that indicate that the transaction rate grows less than 50% per year. That would mean that with Segwit alone blocks will be full at the end of 2018, and with Segwit2x you have time until 2020/21. But you have to take into account that Lightning Network and other approaches will very probably mean that the on-chain volume will grow slower.

I wouldn't touch LN with a stick on a hard-limited block chain where you have no guarantee of settlement. The LN is an entirely different way of doing transactions that has never been experimented, nobody knows if *in practice* it will take off. The step from on-chain transactions to LN is about as big as the change from, say, using a credit card to using a crypto. Maybe it will be a success, maybe not at all. It's something entirely different than what crypto was about until now. The LN would be a great way to connect to one's exchange, that's true. It would keep the exchanges honest, and at the same time, it would turn the exchanges into the banks of tomorrow. But that would entirely change how people see crypto, namely essentially "back to banking".
If an exchange has a serious problem, all its customers would need to settle immediately on chain ; if they can't, because the chain is too small, this is a way for the exchange to run with the money as usual.

dinofelis

hero member

Activity: 770

Merit: 629

Quote from: ComputerGenie on June 03, 2017, 07:03:54 PM

If dreams, wishes, and delusions earned an income, some of you folks would be rich. Undecided

I thought that was the basic principle of crypto

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

If dreams, wishes, and delusions earned an income, some of you folks would be rich. Undecided

d5000

legendary

Activity: 3906

Merit: 6249

Decentralization Maximalist

Quote from: ComputerGenie on June 03, 2017, 11:07:17 AM

Even if you make usable size 25M, there's still more tx than space and people will "fight" for that space through higher fees offered (thus increasing the higher average). Because, math.

Ah OK, if you part from the assumption that 2+x MB won't be enough for the current transaction volume, then I can agree.

But I don't think so. The present enormous mempool sizes are created because of the backlogs of previous hours and days.

This chart (transactions added to the mempool per seconds) confirms that Segwit for now will be enough even without the 2MB expansion. Average transaction rate (mempool, not blocks!) is about 3,5 per second now - if we think of an "effective" capacity increase to 2,2 MB by Segwit alone that will mean that the blocks - giving today's transaction rate - will be only full to 55-60%, given the current average transaction size of about 500 bytes.

On blockchain.info you can also find charts that indicate that the transaction rate grows less than 50% per year. That would mean that with Segwit alone blocks will be full at the end of 2018, and with Segwit2x you have time until 2020/21. But you have to take into account that Lightning Network and other approaches will very probably mean that the on-chain volume will grow slower.

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

Quote from: dinofelis on June 03, 2017, 12:18:32 PM

...

I could be wrong, "rejected" might mean something other than "not accepted". Roll Eyes

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

Quote from: The One on June 03, 2017, 11:38:04 AM

Quote from: The One on June 03, 2017, 08:27:00 AM

Capacity will only go up if all users use segwit. If they don't then capacity will remain the same and the fees will not go down.

I never said the fees will drop.

Pretty sure that "If [not x] then ...will not go down", implies that fees will go down if x. Because, English.

dinofelis

hero member

Activity: 770

Merit: 629

Quote from: European Central Bank on June 03, 2017, 12:20:09 PM

Quote from: jbreher on June 03, 2017, 11:12:02 AM

You User Accidented Segwit Failers are just too precious. You do realize that it is trivial to make such reorgs impossible, right?

yep, but it also makes an official and irreversible chain split. there ain't no going back.

This chain split is acted when the first few bip148 blocks are making a prong from the legacy chain, in fact. And these bit148 blocks with minority hash rate need to get listed as a separate coin on exchanges BEFORE users get in any way their (economic) say, by selling one type of coin, and buying the other type of coin.

Now, it would seem logical that if two coins are listed on exchanges, that both chains become bilaterally protected so that everything that happens on one chain is not going to be gone in a puf of logic whenever the legacy chain would be overtaken by the forking chain, so as to keep the holdings that people have. After all, for those *selling* legacy coins, to buy bip148 coins, they couldn't care much ; but for those buying legacy coins, they may want these transactions to be made immutable and would not like to be reorganized away. So most probably, if the two coins are listed on exchanges, the legacy chain will propose a hard fork to protect itself from re-organisation (the simplest hard fork would be a check point after the split).

European Central Bank

legendary

Activity: 1288

Merit: 1087

Quote from: jbreher on June 03, 2017, 11:12:02 AM

You User Accidented Segwit Failers are just too precious. You do realize that it is trivial to make such reorgs impossible, right?

yep, but it also makes an official and irreversible chain split. there ain't no going back.

dinofelis

hero member

Activity: 770

Merit: 629

Quote from: ComputerGenie on June 03, 2017, 09:52:23 AM

Quote from: dinofelis on June 03, 2017, 09:07:58 AM

...Well, there remains something unclear about this, and when I ask, I get alternating answers ...

I don't know which one corresponds to what is BIP148.

When in doubt, read the BIP.

Quote

While this BIP is active, all blocks must set the nVersion header top 3 bits to 001 together with bit field (1<<1) (according to the existing segwit deployment). Blocks that do not signal as required will be rejected.

Yes, but what does that mean, the block gets rejected ? Does that mean that the node will wait for another block, that does satisfy the criteria, *and is a direct successor* of the last valid block ? Or does it mean that these block headers are still accepted, but not "counted" or something ? Because someone told me that even if there continued to be a single chain with signalling and non-signalling blocks, it would only pick the signalling ones from the chain.

However, if a non-signalling block is truly rejected in full, then the node will wait for a DIRECT SUCCESSOR to the last valid block, and hence, will stop unless there is a true fork in the bitcoin chain, with two prongs: the "legacy" one, and a "pure signalling" one. Someone told me that that was NOT the case.

The One

legendary

Activity: 924

Merit: 1000

Quote from: ComputerGenie on June 03, 2017, 11:01:53 AM

Quote from: The One on June 03, 2017, 09:59:18 AM

Quote from: ComputerGenie on June 03, 2017, 09:56:28 AM

Quote from: The One on June 03, 2017, 09:25:08 AM

Quote from: ComputerGenie on June 03, 2017, 08:28:47 AM

Can you support the claim that fees are proportionally related to size with an equation?

It's simply supply and demand. Supply of capacity lower than the demand of capacity = higher fees. The concept of fee market is retarded, a concept based on devs not understanding economics and/or accountancy. Segwit capacity to accept more transactions can only happen is users use it. If users stick to the current means of transaction, the capacity will be the same.

So, in other words, "no" you can't support your claim, you're just assuming what will happen (and ignoring the possibility that it will have exactly zero impact on fees).

Already have.
No such a thing. Nothing is ever static in economic.

No, you haven't supported your claim, you have asserted assumptions. You are assuming that users will do x if size is y, with no factually mathematical correlation between x and y.
You're conflating "static" and "impact" (as well as conflating "proof" and "belief"). I didn't say that fees would never change, I said that you cannot mathematical show that fees will, in fact, drop as a direct result of segwit.

Quote from: ComputerGenie on June 03, 2017, 08:28:47 AM

Quote from: The One on June 03, 2017, 08:27:00 AM

Capacity will only go up if all users use segwit. If they don't then capacity will remain the same and the fees will not go down.

Can you support the claim that fees are proportionally related to size with an equation?

I never said the fees will drop.

jbreher

legendary

Activity: 3080

Merit: 1688

lose: unfind ... loose: untight

Quote from: BillyBobZorton on June 03, 2017, 06:42:53 AM

UASF wins, because all the jihancoins will go to 0 due reorg

You User Accidented Segwit Failers are just too precious. You do realize that it is trivial to make such reorgs impossible, right?

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

Quote from: d5000 on June 03, 2017, 10:13:35 AM

Quote from: ComputerGenie on June 03, 2017, 08:22:55 AM

Will segwit reduce the fees someone is willing to pay for a given transaction? Maybe.
Will segwit reduce the average fee when people manually set "high" fees to be at the "top of the list"? No.

And why should the users not reduce fees when block capacity gets higher and blocks are not longer full? ....

For the same reason that fees rose in the 1st place, the mentally that higher fees = faster confirmation. Even if you make usable size 25M, there's still more tx than space and people will "fight" for that space through higher fees offered (thus increasing the higher average). Because, math.

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

Quote from: The One on June 03, 2017, 09:59:18 AM

Quote from: ComputerGenie on June 03, 2017, 09:56:28 AM

Quote from: The One on June 03, 2017, 09:25:08 AM

Quote from: ComputerGenie on June 03, 2017, 08:28:47 AM

Can you support the claim that fees are proportionally related to size with an equation?

It's simply supply and demand. Supply of capacity lower than the demand of capacity = higher fees. The concept of fee market is retarded, a concept based on devs not understanding economics and/or accountancy. Segwit capacity to accept more transactions can only happen is users use it. If users stick to the current means of transaction, the capacity will be the same.

So, in other words, "no" you can't support your claim, you're just assuming what will happen (and ignoring the possibility that it will have exactly zero impact on fees).

Already have.
No such a thing. Nothing is ever static in economic.

No, you haven't supported your claim, you have asserted assumptions. You are assuming that users will do x if size is y, with no factually mathematical correlation between x and y.
You're conflating "static" and "impact" (as well as conflating "proof" and "belief"). I didn't say that fees would never change, I said that you cannot mathematical show that fees will, in fact, drop as a direct result of segwit.

d5000

legendary

Activity: 3906

Merit: 6249

Decentralization Maximalist

Quote from: ComputerGenie on June 03, 2017, 08:22:55 AM

Will segwit reduce the fees someone is willing to pay for a given transaction? Maybe.
Will segwit reduce the average fee when people manually set "high" fees to be at the "top of the list"? No.

And why should the users not reduce fees when block capacity gets higher and blocks are not longer full? The actual user behaviour to (almost panicky) set very high fees is a consequence of the full blocks. You can see that at weekends, average fees are lower then because demand is lower.

Quote from: dinofelis on June 03, 2017, 09:12:09 AM

You can't have a soft fork and 2 MB.

According to a (veteran) user of the German local forum there seems to be a way, although obviously only "new-style" (P2WPKH/P2WPSH) transactions would benefit from the 2MB "base" size.

The One

legendary

Activity: 924

Merit: 1000

Quote from: ComputerGenie on June 03, 2017, 09:56:28 AM

Quote from: The One on June 03, 2017, 09:25:08 AM

Quote from: ComputerGenie on June 03, 2017, 08:28:47 AM

Can you support the claim that fees are proportionally related to size with an equation?

It's simply supply and demand. Supply of capacity lower than the demand of capacity = higher fees. The concept of fee market is retarded, a concept based on devs not understanding economics and/or accountancy. Segwit capacity to accept more transactions can only happen is users use it. If users stick to the current means of transaction, the capacity will be the same.

So, in other words, "no" you can't support your claim, you're just assuming what will happen (and ignoring the possibility that it will have exactly zero impact on fees).

Already have.
No such a thing. Nothing is ever static in economic.

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

Quote from: The One on June 03, 2017, 09:25:08 AM

Quote from: ComputerGenie on June 03, 2017, 08:28:47 AM

Can you support the claim that fees are proportionally related to size with an equation?

It's simply supply and demand. Supply of capacity lower than the demand of capacity = higher fees. The concept of fee market is retarded, a concept based on devs not understanding economics and/or accountancy. Segwit capacity to accept more transactions can only happen is users use it. If users stick to the current means of transaction, the capacity will be the same.

So, in other words, "no" you can't support your claim, you're just assuming what will happen (and ignoring the possibility that it will have exactly zero impact on fees).

ComputerGenie

hero member

Activity: 1092

Merit: 552

Retired IRCX God

Quote from: dinofelis on June 03, 2017, 09:07:58 AM

...Well, there remains something unclear about this, and when I ask, I get alternating answers ...

I don't know which one corresponds to what is BIP148.

When in doubt, read the BIP.

Quote

While this BIP is active, all blocks must set the nVersion header top 3 bits to 001 together with bit field (1<<1) (according to the existing segwit deployment). Blocks that do not signal as required will be rejected.

BillyBobZorton

legendary

Activity: 1204

Merit: 1028

Quote from: d5000 on June 03, 2017, 08:14:56 AM

Quote from: ComputerGenie on June 03, 2017, 07:43:16 AM

Quote from: d5000 on June 03, 2017, 07:40:38 AM

...Most support Segwit because it would reduce transaction fees...

Can we, please, stop peddling that myth?

Why is that a myth? In the short term, Segwit would almost surely reduce fees, with >2x the capacity of the old 1MB chain. Or what do you mean?

Quote from: mindrust on June 03, 2017, 07:47:38 AM

Quote from: d5000 on June 03, 2017, 07:40:38 AM

Do you really think most users are caring about the details about Segwit? They want Bitcoin to work, period.

If you call min 5$ rip-off fee per transaction or from 20 hours to infinite hours of delay as "working", then yes it is working. (not as intended™)

You didn't get my point.

Segwit+2MB solves the transaction fee problem the same way than SegwitUASF does, at least in the short term. Transaction fees with Segwit2x would even be smaller in the mid-term if after an hypothetical UASF there is no 2MB adjustment in one or another way.

That's why I don't think your assumption that the majority users are insisting on "Segwit alone" and are against "Segwit2x", is true. The majority support Segwit, that's clear for me, and not the BU approach. But the details - bit 1 or bit 4 and so on - are really insignificant.

Segwit2x, for me, is not ideal because it involves a hard fork and that should be avoided if possible. But with a large support by economy & miners I think the hard fork would go smooth.

There is an attempt to properly hard fork to 2MB with some Core devs involved. I would be ok with that as long as we get everyone on board. A HF without Core-backed software is dead on arrival, you also need the entire community to get ready, or else we risk an actual ETH-ETC scenario.

The NYC is bullshit and already failed, so there are attempts now to create something better.

Of course Bitmain bribed agents will not want to get rid of covert ASICBOOST so we'll end up with UASF anyway.

The One

legendary

Activity: 924

Merit: 1000

Quote from: ComputerGenie on June 03, 2017, 08:28:47 AM

Quote from: The One on June 03, 2017, 08:27:00 AM

Capacity will only go up if all users use segwit. If they don't then capacity will remain the same and the fees will not go down.

Can you support the claim that fees are proportionally related to size with an equation?

It's simply supply and demand. Supply of capacity lower than the demand of capacity = higher fees. The concept of fee market is retarded, a concept based on devs not understanding economics and/or accountancy. Segwit capacity to accept more transactions can only happen is users use it. If users stick to the current means of transaction, the capacity will be the same.

Topic: The Barry Silbert segwit2x agreement with >80% miner support. - page 56. (Read 120030 times)