Fee

[byt411]

Pretty much this right here. 10 cents though would come with a fee that (right now) is 5 cents, or 50%. $10 would still be 5 cents. So really it's still a pretty small fee.

Just out of reference, with the newest implementation fees should drop to 0.5c so it basically means that txs are pretty damn cheap.

In theory, you're correct. The problem is that a lot of the miners and pools are still using older versions of the Bitcoin client. There was a thread about this, saying that like 70% of them still haven't upgraded in like a year or so and so are far behind as it is. On top of this, the miners/pools can apparently set their own requirements for fees and decline transactions that don't match what they want.

You are correct, we need some way to force nodes to update, since they play a huge part in the bitcoin network. I guess you could set up your own node, but that's just a bit difficult and unnecessary.

[ranlo]

Pretty much this right here. 10 cents though would come with a fee that (right now) is 5 cents, or 50%. $10 would still be 5 cents. So really it's still a pretty small fee.

Just out of reference, with the newest implementation fees should drop to 0.5c so it basically means that txs are pretty damn cheap.

In theory, you're correct. The problem is that a lot of the miners and pools are still using older versions of the Bitcoin client. There was a thread about this, saying that like 70% of them still haven't upgraded in like a year or so and so are far behind as it is. On top of this, the miners/pools can apparently set their own requirements for fees and decline transactions that don't match what they want.

You are correct, we need some way to force nodes to update, since they play a huge part in the bitcoin network. I guess you could set up your own node, but that's just a bit difficult and unnecessary.

Setting up your own node wouldn't do any good, either, unless you happened to mine a block. The only people whose daemons really matter (as far as I understand it) are the ones that are minting. And since a couple people are minting the majority of blocks, it's them that need to change (or not, depending on how they feel).

[byt411]

Pretty much this right here. 10 cents though would come with a fee that (right now) is 5 cents, or 50%. $10 would still be 5 cents. So really it's still a pretty small fee.

Just out of reference, with the newest implementation fees should drop to 0.5c so it basically means that txs are pretty damn cheap.

In theory, you're correct. The problem is that a lot of the miners and pools are still using older versions of the Bitcoin client. There was a thread about this, saying that like 70% of them still haven't upgraded in like a year or so and so are far behind as it is. On top of this, the miners/pools can apparently set their own requirements for fees and decline transactions that don't match what they want.

You are correct, we need some way to force nodes to update, since they play a huge part in the bitcoin network. I guess you could set up your own node, but that's just a bit difficult and unnecessary.

Setting up your own node wouldn't do any good, either, unless you happened to mine a block. The only people whose daemons really matter (as far as I understand it) are the ones that are minting. And since a couple people are minting the majority of blocks, it's them that need to change (or not, depending on how they feel).

But if your node is directly connected to a pool's one, wouldn't it accept your transaction and broadcast it to the specific pool?

[ranlo]

Pretty much this right here. 10 cents though would come with a fee that (right now) is 5 cents, or 50%. $10 would still be 5 cents. So really it's still a pretty small fee.

Just out of reference, with the newest implementation fees should drop to 0.5c so it basically means that txs are pretty damn cheap.

In theory, you're correct. The problem is that a lot of the miners and pools are still using older versions of the Bitcoin client. There was a thread about this, saying that like 70% of them still haven't upgraded in like a year or so and so are far behind as it is. On top of this, the miners/pools can apparently set their own requirements for fees and decline transactions that don't match what they want.

You are correct, we need some way to force nodes to update, since they play a huge part in the bitcoin network. I guess you could set up your own node, but that's just a bit difficult and unnecessary.

Setting up your own node wouldn't do any good, either, unless you happened to mine a block. The only people whose daemons really matter (as far as I understand it) are the ones that are minting. And since a couple people are minting the majority of blocks, it's them that need to change (or not, depending on how they feel).

But if your node is directly connected to a pool's one, wouldn't it accept your transaction and broadcast it to the specific pool?

It broadcasts it, yes. As far as I'm aware, there are no issues with the broadcasting part. It doesn't ensure it'd get into a block, though. You can test this by sending a really small transaction without a fee. It will be broadcast to all the nodes (and even show up on Blockchain.info within seconds). But it will take days to get in a block, if it ever does. Basically it works like this:

1) You send a transaction from your client or a web wallet
2) Transaction is broadcast to all the nodes to share
3) When a pool is minting, it chooses what transactions it wants to include
4) If a pool mints a block, the transactions it was including are now "confirmed"

The problem isn't with the broadcasting, it's with the pools/miners not packaging them into a block due to not being valuable enough.

[byt411]

Pretty much this right here. 10 cents though would come with a fee that (right now) is 5 cents, or 50%. $10 would still be 5 cents. So really it's still a pretty small fee.

Just out of reference, with the newest implementation fees should drop to 0.5c so it basically means that txs are pretty damn cheap.

In theory, you're correct. The problem is that a lot of the miners and pools are still using older versions of the Bitcoin client. There was a thread about this, saying that like 70% of them still haven't upgraded in like a year or so and so are far behind as it is. On top of this, the miners/pools can apparently set their own requirements for fees and decline transactions that don't match what they want.

You are correct, we need some way to force nodes to update, since they play a huge part in the bitcoin network. I guess you could set up your own node, but that's just a bit difficult and unnecessary.

Setting up your own node wouldn't do any good, either, unless you happened to mine a block. The only people whose daemons really matter (as far as I understand it) are the ones that are minting. And since a couple people are minting the majority of blocks, it's them that need to change (or not, depending on how they feel).

But if your node is directly connected to a pool's one, wouldn't it accept your transaction and broadcast it to the specific pool?

It broadcasts it, yes. As far as I'm aware, there are no issues with the broadcasting part. It doesn't ensure it'd get into a block, though. You can test this by sending a really small transaction without a fee. It will be broadcast to all the nodes (and even show up on Blockchain.info within seconds). But it will take days to get in a block, if it ever does. Basically it works like this:

1) You send a transaction from your client or a web wallet
2) Transaction is broadcast to all the nodes to share
3) When a pool is minting, it chooses what transactions it wants to include
4) If a pool mints a block, the transactions it was including are now "confirmed"

The problem isn't with the broadcasting, it's with the pools/miners not packaging them into a block due to not being valuable enough.

That's because the miners choose not to accept it, that has nothing to do with the Bitcoin Core version. Also, don't use "mint", use "mine".

[ranlo]

Pretty much this right here. 10 cents though would come with a fee that (right now) is 5 cents, or 50%. $10 would still be 5 cents. So really it's still a pretty small fee.

Just out of reference, with the newest implementation fees should drop to 0.5c so it basically means that txs are pretty damn cheap.

In theory, you're correct. The problem is that a lot of the miners and pools are still using older versions of the Bitcoin client. There was a thread about this, saying that like 70% of them still haven't upgraded in like a year or so and so are far behind as it is. On top of this, the miners/pools can apparently set their own requirements for fees and decline transactions that don't match what they want.

You are correct, we need some way to force nodes to update, since they play a huge part in the bitcoin network. I guess you could set up your own node, but that's just a bit difficult and unnecessary.

Setting up your own node wouldn't do any good, either, unless you happened to mine a block. The only people whose daemons really matter (as far as I understand it) are the ones that are minting. And since a couple people are minting the majority of blocks, it's them that need to change (or not, depending on how they feel).

But if your node is directly connected to a pool's one, wouldn't it accept your transaction and broadcast it to the specific pool?

It broadcasts it, yes. As far as I'm aware, there are no issues with the broadcasting part. It doesn't ensure it'd get into a block, though. You can test this by sending a really small transaction without a fee. It will be broadcast to all the nodes (and even show up on Blockchain.info within seconds). But it will take days to get in a block, if it ever does. Basically it works like this:

1) You send a transaction from your client or a web wallet
2) Transaction is broadcast to all the nodes to share
3) When a pool is minting, it chooses what transactions it wants to include
4) If a pool mints a block, the transactions it was including are now "confirmed"

The problem isn't with the broadcasting, it's with the pools/miners not packaging them into a block due to not being valuable enough.

That's because the miners choose not to accept it, that has nothing to do with the Bitcoin Core version. Also, don't use "mint", use "mine".

Mining, minting, either way it's the creation of more coins, :p.

I was under the impression that the client is what determines the minimum fees for mining blocks?

[PerrythePlatypus]

It is actually risky, not with the app but with the transaction fees. Transactions that does not include a fee may never confirm depending on the factors. There are a list of factors: https://en.bitcoin.it/wiki/Transaction_fees#Sending

[1986]

You should always send a fee. It's not very much and helps the mining community.

[ranlo]

It is actually risky, not with the app but with the transaction fees. Transactions that does not include a fee may never confirm depending on the factors. There are a list of factors: https://en.bitcoin.it/wiki/Transaction_fees#Sending

If it doesn't confirm, though, you should be able to reverse the transaction (or, rather, delete it locally) and re-send with the fee. I think some clients (Electrum and MultiBit?) automatically do this after like 48h.

[byt411]

It is actually risky, not with the app but with the transaction fees. Transactions that does not include a fee may never confirm depending on the factors. There are a list of factors: https://en.bitcoin.it/wiki/Transaction_fees#Sending

If it doesn't confirm, though, you should be able to reverse the transaction (or, rather, delete it locally) and re-send with the fee. I think some clients (Electrum and MultiBit?) automatically do this after like 48h.

That is technically incorrect, you cannot "reverse" a transaction. It just gets dropped out of the memory pool (a database of unconfirmed transactions), and, if it doesn't get rebroadcasted, then you get your coins back.

[ranlo]

Quote from: ranlo on Today at 02:36:31 PM

Quote from: PerrythePlatypus on Today at 07:21:32 AM

It is actually risky, not with the app but with the transaction fees. Transactions that does not include a fee may never confirm depending on the factors. There are a list of factors: https://en.bitcoin.it/wiki/Transaction_fees#Sending


If it doesn't confirm, though, you should be able to reverse the transaction (or, rather, delete it locally) and re-send with the fee. I think some clients (Electrum and MultiBit?) automatically do this after like 48h.


That is technically incorrect, you cannot "reverse" a transaction. It just gets dropped out of the memory pool (a database of unconfirmed transactions), and, if it doesn't get rebroadcasted, then you get your coins back.

It may not be called a reversal but that's technically what it is. You broadcasted a transaction, then reversed it. It just wasn't official yet.

[DeathAndTaxes]

Does it matter if I define a fee of 0.00005 or 0.00001? Would both transactions be treated as "no fee" transactions? 

The fee (I forgot what the lowest is, but I know it was changed) would be present, so no, it wouldn't be a no-fee one. It would just be less of a fee.

No fee would literally be 0. For example,  you are sending 1 BTC, that's all. Not 1.0000001, but 1.0.

Fees below the min for relaying are treated by default nodes as no different than zero.  For v0.8x that is 0.1 mBTC and for v0.9x that is 0.01 mBTC.  Paying a fee greater than zero but less than the min to relay is pointless.

[ranlo]

Quote from: ranlo on May 21, 2014, 12:54:29 AM

Quote from: roeylee on May 21, 2014, 12:46:58 AM

Does it matter if I define a fee of 0.00005 or 0.00001? Would both transactions be treated as "no fee" transactions? 


The fee (I forgot what the lowest is, but I know it was changed) would be present, so no, it wouldn't be a no-fee one. It would just be less of a fee.

No fee would literally be 0. For example,  you are sending 1 BTC, that's all. Not 1.0000001, but 1.0.


Fees below the min for relaying are treated by default nodes as no different than zero.  For v0.8x that is 0.1 mBTC and for v0.9x that is 0.01 mBTC.  Paying a fee greater than zero but less than the min to relay is pointless.

 I was completely unaware of that. Wouldn't it be more efficient to just organize based on fee paid, such that 5 satoshi would be above 1 satoshi, rather than treating them as equals?

[DeathAndTaxes]

I was completely unaware of that. Wouldn't it be more efficient to just organize based on fee paid, such that 5 satoshi would be above 1 satoshi, rather than treating them as equals?

It could however then it would almost guarantee that free tx would never be included in blocks.  Instead miners devote a set amount of space to high priority txs (relayable without a fee) and a set amount of space for paying txs and a min fee.   

This might be 50KB, 350KB, and 0.1 mBTC as an example.
So transactions they are ranked by priority (fees if any are ignored) and the top 50KB worth are included in the fee portion of the block.
The transactions are then ranked by fee (per KB) and the top 350KB that are over 0.1 mBTC are included in the block.

You might say "what prevents a miner from just setting the min fee to 1 satoshi per KB?" the answer is nothing however miners rely on other nodes to relay transactions and a low priority tx with a fee below 0.1mBTC (0.01 mBTC for v0.9+) will just be dropped.  The miner will never even learn about it.  Without modification the tx would be dropped by the miners own node as well.   So there is no reason for a miner to set their fee requirement below the min required to relay.

[ranlo]

I was completely unaware of that. Wouldn't it be more efficient to just organize based on fee paid, such that 5 satoshi would be above 1 satoshi, rather than treating them as equals?

It could however then it would almost guarantee that free tx would never be included in blocks.  Instead miners devote a set amount of space to high priority txs (relayable without a fee) and a set amount of space for paying txs and a min fee.   

This might be 50KB, 350KB, and 0.1 mBTC as an example.
So transactions they are ranked by priority (fees if any are ignored) and the top 50KB worth are included in the fee portion of the block.
The transactions are then ranked by fee (per KB) and the top 350KB that are over 0.1 mBTC are included in the block.

You might say "what prevents a miner from just setting the min fee to 1 satoshi per KB?" the answer is nothing however miners rely on other nodes to relay transactions and a low priority tx with a fee below 0.1mBTC (0.01 mBTC for v0.9+) will just be dropped.  The miner will never even learn about it.  Without modification the tx would be dropped by the miners own node as well.   So there is no reason for a miner to set their fee requirement below the min required to relay.

That makes sense. I didn't think about people doing 1 satoshi to kick out the freebies (and essentially get free transactions still).

Is it possible to see what nodes are accepting x minimum payments? Like can we somehow see (based on the blockchain and mined blocks) what the transactions are that each major pool/miner is going after?

[PerrythePlatypus]

It is actually risky, not with the app but with the transaction fees. Transactions that does not include a fee may never confirm depending on the factors. There are a list of factors: https://en.bitcoin.it/wiki/Transaction_fees#Sending

If it doesn't confirm, though, you should be able to reverse the transaction (or, rather, delete it locally) and re-send with the fee. I think some clients (Electrum and MultiBit?) automatically do this after like 48h.

That is technically incorrect, you cannot "reverse" a transaction. It just gets dropped out of the memory pool (a database of unconfirmed transactions), and, if it doesn't get rebroadcasted, then you get your coins back.

Mine was automatically rebroadcasted for 11 days already, using multibit's wallet. It is more effective to double spend it rather than wait.