Bitcoin Forum
December 12, 2017, 08:47:05 AM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: [1]
  Print  
Author Topic: Question regarding minimum fees for small transactions  (Read 967 times)
SebastianJu
Legendary
*
Offline Offline

Activity: 1974


Legendary Escrow Service - Tip Jar in Profile


View Profile WWW
March 21, 2013, 03:40:45 PM
 #1

Hello,

i read at https://en.bitcoin.it/wiki/Transaction_fee#Sending the rules for not having to include a fee. Now i wonder whats the sense to it. I mean 0.01BTC means you only can send $0.64 bitcoins valued in USD without having to pay a fee. And one have to pay a fee of 0.0005BTC for that. That means the smaller a transaction the more unuseable is bitcoin for it. If you would want to send someone a testsatoshi you would have to add 5000 satoshis to process that transaction. That means a big part of the digits can only be used to narrow down a big value more exact but not to send such a small amount because the fee would eat the sense of sending that much.

On top... once the demand for bitcoins is high enough the digits can be extended to 9 digits for example. But its not even possible yet to use them for small values to transfer because of the forced fee. If you dont want to pay a fee anyway the transaction will never go into blockchain it looks to me.

So it looks a bit strange to me. High transactions can go without fee but small transactions have to pay a relatively high fee. While you wouldnt notice a small fee on big transactions its the opposite for small ones. And its not even narrowed down if you only want to send one satoshi. I mean if it would be a spamprotection then one satoshi fee for sending one satoshi would prevent ddos-attacks good enough.

So why is this minimum fee so high and why the fee is handled this way? Its somehow like rich persons dont have to pay tax but the poor persons have to pay the more. Dont misunderstand me. I make big transactions too, but how the fees are constructed isnt quite logic to me.

Thanks!
Sebastian
1513068425
Hero Member
*
Offline Offline

Posts: 1513068425

View Profile Personal Message (Offline)

Ignore
1513068425
Reply with quote  #2

1513068425
Report to moderator
1513068425
Hero Member
*
Offline Offline

Posts: 1513068425

View Profile Personal Message (Offline)

Ignore
1513068425
Reply with quote  #2

1513068425
Report to moderator
1513068425
Hero Member
*
Offline Offline

Posts: 1513068425

View Profile Personal Message (Offline)

Ignore
1513068425
Reply with quote  #2

1513068425
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1513068425
Hero Member
*
Offline Offline

Posts: 1513068425

View Profile Personal Message (Offline)

Ignore
1513068425
Reply with quote  #2

1513068425
Report to moderator
1513068425
Hero Member
*
Offline Offline

Posts: 1513068425

View Profile Personal Message (Offline)

Ignore
1513068425
Reply with quote  #2

1513068425
Report to moderator
pc
Sr. Member
****
Offline Offline

Activity: 253


View Profile
March 21, 2013, 04:49:55 PM
 #2

For "traditional" currencies, costs of moving the money are to some extent proportional to the amount moved. While there's certainly a fixed cost (of moving the bits in the computers usually), there are things like risk of chargebacks and insurance against loss where it makes some sense to charge more when there's more to insure.

But for bitcoin, the costs are mainly measured in bytes. If I have 1 BTC, breaking it in half (sending 0.5 to you and keeping 0.5 "change") means that there is now twice as much data for the system to keep track of. It needs to be able to handle whether or not you spent your 0.5 and whether I spent my 0.5. The nodes in the system have to do the same amount of validation (and store it more-or-less-forever in the global transaction history, at least until it gets spent again) whether somebody is spending 100,000 BTC or 1 satoshi.

Because of this, the fee structures are based much more on bytes to represent the transaction instead of on value transferred. I expect that one day, the positive things that consumers do get from the "traditional" system (having a reversible transaction in the case of merchant fraud, being able to insure value easily) will be done in layers "on top" of bitcoin, and possibly those services will charge a fee in proportion to the value. Bitcoin was never designed to be a micropayment system (it does, after all, need to store all the transactions globally) so it's likely that micropayment-friendly services will be built on top of bitcoin too one day. (And if the value of bitcoin somehow keeps on skyrocketing so that 0.01 BTC isn't a micropayment anymore, then the default fee schedule used by nodes and miners will be updated accordingly.)

Or, compare it to the traditional system: How much would a merchant need to pay in credit card fees, wire fees, or armored truck/courier fees in order to receive a $0.64 payment today from anywhere in the world? It's probably more than 0.0005 BTC worth.
SebastianJu
Legendary
*
Offline Offline

Activity: 1974


Legendary Escrow Service - Tip Jar in Profile


View Profile WWW
March 21, 2013, 05:57:45 PM
 #3

Ok, i see the point. You claim that with rising btc-exchange price the fee will be updated accordingly. But a bitcoin is already way way more worth now then it was some years ago. The same goes for the minimum fee. Couldnt this be taken into account now? I mean it was worth 50Cent once, now its 50USD, thats 100 time more or 10000%. And the fee is still the same. That means earlier you could transfer way smaller parts of value and paid a less fee. Now the price is way higher and it meets a wall where you cant pay without fee anymore. And that wall means a way bigger value than at the time when the fee-structure was set. I wonder if the exchange price shouldnt taken into account now because of course bitcoins remain bitcoin but bitcoin is a real world currency too and one should buy realworld things with it. So the worth of bitcoins is important too.

I know the blockchain is really big now and it will grow even bigger with more transactions but transactions are the blood of a currency. So restricting transactions (which happens when you cant transfer a small value because the fee is way higher or a too big portion) doesnt sound like the best idea, even though i dont know how to solve the size of the blockchain problem. (Is it possible to take only the last 100 blocks and calculate from there if you can trust these blocks? Probably not isnt it?)

I think having to pay at least 1 satoshi for each transaction would be a good enough ddos-protection. So maybe one time it can be done this way that a transaction below 0.01BTC that havent a fee, or with a minimum-fee of one satoshi, will only wait longer instead not being executed at all? I mean its not that all the time of the day there are many transactions. Maybe if its more or less idle those transactions can be taken into the blockchain. The user only has to wait longer then. Its not ideal but at least it would be possible to move small money without a high percent fee.

The credit card fee is a point. But if the price is rising higher bitcoins couldnt be used for example buying a lollipop for 20Cent in a kiosk by school kids that pay with the smartphone. The fee now wouldnt bee too much but in some months it will be another thing. Of course bitcoins doesnt have to be used for this. But it meets a wall now that wasnt there in the beginning when the fee was created the way it is now.
pc
Sr. Member
****
Offline Offline

Activity: 253


View Profile
March 21, 2013, 06:22:08 PM
 #4

The default fee schedule in the reference client is just the default. Miners can (and do) change it. If you can convince one of them that it's worth 1 satoshi to process your transaction, the network deals with that just fine. The standard client tries to have you add a fee mainly because it's hard to present to the user that they might be signing up for an indefinitely long wait if they are going to pay less than the defaults. Having a better market for the fees so that it's clear what you're paying for one service is definitely one of the priorities of the project.
DeathAndTaxes
Donator
Legendary
*
Offline Offline

Activity: 1218


Gerald Davis


View Profile
March 21, 2013, 06:26:13 PM
 #5

Ok, i see the point. You claim that with rising btc-exchange price the fee will be updated accordingly. But a bitcoin is already way way more worth now then it was some years ago. The same goes for the minimum fee. Couldnt this be taken into account now?

The min mandatory fee has been reduced once already.  It was 0.01 BTC at one time.  As he said it can be reduced.  The last time it was reduced is when BTC spiked to $32 in 2011.  So the fee isn't much higher then it was then.  If BTC continues to increase in price the min mandatory fee will be reduced.

Quote
I think having to pay at least 1 satoshi for each transaction would be a good enough ddos-protection.

Obviously you haven't thought about this much.
SebastianJu
Legendary
*
Offline Offline

Activity: 1974


Legendary Escrow Service - Tip Jar in Profile


View Profile WWW
March 21, 2013, 11:43:13 PM
 #6

Quote
I think having to pay at least 1 satoshi for each transaction would be a good enough ddos-protection.

Obviously you haven't thought about this much.

Why do you think that? I mean ddos works with thousands and thousands of transactions. So if you have to pay 1 satoshi for sending one satoshi then you would have to pay 1BTC to make an attack of 50000000 transactions. Thats a bit costly attack. $0.69 for 500000 transactions. I might be wrong but i thought ddos only works in very high amounts of connections. But if someone has the money and wants to spend it youre right. It would be a possible risk then.

I already tried to send 2 transactions below the border and without a fee. After days nothing happened. No node did accept the transaction so that even blockchain.info didnt know the t-id and no miner could decide that it will be implemented because it wasnt spread in the net. Today i tried to send one satoshi with 1 satoshi as fee but after 10 hours no conf is there and blockchain.info doesnt know the tid either. So it seems to not being spread because of the low fee and so doesnt get the chance to find a miner that would include the transaction.
I used electrum to test this because it doesnt enforce the minimum fee. The developer says its the nodes then that doesnt accept the transaction and the electrum server tries to propagate the transaction without success. So it doesnt get the chance to find a miner that will include it.
DeathAndTaxes
Donator
Legendary
*
Offline Offline

Activity: 1218


Gerald Davis


View Profile
March 22, 2013, 12:32:49 AM
 #7

So 5,000,0000 tx at a cost of 1 BTC.  Average tx size is 400 bytes so thats ~2GB added to the blockchain for 1 BTC.  At a cost of 1,000 BTC a malicous user could permanently bloat the blockchain by 1TB.  Imagine having to bootstrap as a new node with a 1TB blockchain full of spam which needs to be carried by every node forever for a token cost of only 1,000 BTC.
Syke
Legendary
*
Offline Offline

Activity: 2464


View Profile
March 22, 2013, 03:32:44 AM
 #8

Why do you think that? I mean ddos works with thousands and thousands of transactions. So if you have to pay 1 satoshi for sending one satoshi then you would have to pay 1BTC to make an attack of 50000000 transactions. Thats a bit costly attack.

Costly? Each block can hold about 2500 transactions. If you split a single bitcoin into 1 satoshi transactions, it would fill up every block for the next 4.6 months. So no, 1 satoshi as a minimum fee wouldn't work.

Buy & Hold
SebastianJu
Legendary
*
Offline Offline

Activity: 1974


Legendary Escrow Service - Tip Jar in Profile


View Profile WWW
March 22, 2013, 04:48:07 PM
 #9

Ok, youre right... i didnt think about that. I only thought about the standard-ddos that means many many connections to block a system. But bitcoin can be blocked with too many transactions easy.
Pages: [1]
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!