Why a fixed transaction fee? | Bitcointalksearch.org

cp1

hero member

Activity: 616

Merit: 500

Stop using branwallets

Quote from: whtchocla7e on April 03, 2014, 12:08:54 PM

I gave my friend a nickel in person and it cost me absolutely nothing in postage.

We could exchange the same nickel between us 100 times per minute forever and still have the whole nickel at the end.

Try that with Bitcoin.

You could hand the private key back and forth and do the same.

whtchocla7e

full member

Activity: 392

Merit: 116

Worlds Simplest Cryptocurrency Wallet

Quote from: cp1 on April 03, 2014, 08:54:27 AM

I mailed my friend a nickel and it cost 49 cents in postage!! The postal service should change their fee based on if you have a nickel inside or not.

I gave my friend a nickel in person and it cost me absolutely nothing in postage.

We could exchange the same nickel between us 100 times per minute forever and still have the whole nickel at the end.

Try that with Bitcoin.

instagibbs

member

Activity: 114

Merit: 12

Quote from: Gavin Andresen on April 03, 2014, 10:30:34 AM

Quote from: instagibbs on April 03, 2014, 09:01:37 AM

Not quite right. Miners can and will mine things with a lower fee. Many nodes will simply not forward the transactions however.

For low-priority transactions, it doesn't look to me like many miners are accepting lower fees.

Ah. I don't have data, I was just trying to tease apart the idea that a certain fee is required and the txn will be rejected by the network otherwise. It became important recently as the min txn fee for propogation dropped a lot, while miners haven't changed their behavior much wrt what fees they want.

I should probably play around with the new rpc calls; in the end it's an auction, and auctions are fascinating.

Massimo80

full member

Activity: 168

Merit: 100

Quote from: cp1 on April 03, 2014, 08:54:27 AM

I mailed my friend a nickel and it cost 49 cents in postage!! The postal service should change their fee based on if you have a nickel inside or not.

Best answer so far Grin

Anyway, I didn't want to start a flame... just asking.

Relax, guys...

Gavin Andresen

legendary

Activity: 1652

Merit: 2301

Chief Scientist

Quote from: instagibbs on April 03, 2014, 09:01:37 AM

Not quite right. Miners can and will mine things with a lower fee. Many nodes will simply not forward the transactions however.

For low-priority transactions, it doesn't look to me like many miners are accepting lower fees.

Right now, a 999-byte transaction paying the reference-implementation-default fee of 0.0001 BTC will wait 2-3 hours to get into a block.

A typical 250-byte transaction paying the default 0.0001 BTC fee will see its first confirmation in 5 or 6 blocks (about an hour on average). If you want your transaction to confirm quickly, then right now you need to pay about double the default fee.

Some results running https://github.com/bitcoin/bitcoin/pull/3959 (more review/testing welcome) :

Code:

$ for i in {1..25}; do ./bitcoin-cli estimatefee $i; done
0.00077821
0.00044643
0.00044444
0.00044248
0.00044248
0.00038911
0.00038760
0.00026810
0.00024752
0.00022831
0.00020040
0.00017513
0.00016155
0.00014706
0.00013802
0.00012531
0.00011779
0.00011013
0.00010363
0.00010111
0.00008905
0.00008636
0.00007474
0.00006743
0.00004444

instagibbs

member

Activity: 114

Merit: 12

Quote from: DannyHamilton on March 31, 2014, 11:41:40 AM

If it is crazy to send 0.00005 BTC with a 0.0001 BTC fee, then perhaps, bitcoin is not useful for sending such small payments. I suggest using your local currency (or a credit card, or bank transfer) when you want to send amounts equivalent to 0.00005 BTC (At today's exchange rate, that's $0.028).

Not quite right. Miners can and will mine things with a lower fee. Many nodes will simply not forward the transactions however.

cp1

hero member

Activity: 616

Merit: 500

Stop using branwallets

I mailed my friend a nickel and it cost 49 cents in postage!! The postal service should change their fee based on if you have a nickel inside or not.

roslinpl

legendary

Activity: 2212

Merit: 1199

Quote from: Massimo80 on March 31, 2014, 08:02:48 AM

When making a Bitcoin payment, you can include a transaction fee to speed up processing of your transaction by rewarding miners for processing it. Ok.

But more often than not the software (at least the reference client) requires you to send a fee, which can range from 0.0001 to 0.0002 BTC, or even more.

The problem is, this becomes more and more expensive as the BTC price increases; when it reached $1000, a transaction could cost $1-2, making it effectively impossibile to send a $1 payment, and highly unpractical to send a $10 one. If the price should ever go up to $10K, or even more (which it technically could, if Bitcoin was ever to become really widespread, since it's a deflactionary currency), transaction fees could become so high to make you actually wish you were still using a credit card, and low-value payments would become unfeasible at all.

Thus my question: why a fixed transaction fee? Shouldn't it at least scale up with the actual sum you're sending? Asking for a 0.0001 fee when sending 0.00005 BTC is completely crazy...

IMO "sending 0.00005 BTC is completely crazy..."

And by using bitcoin-qt you can do not add any fee if you really want to wait forever for confirmations x)

softron

full member

Activity: 210

Merit: 100

It not fixed. It the recommended fee and it has been changed before and wiil be changed again if price sky rockets.

Chemistry1988

legendary

Activity: 1120

Merit: 1000

Quote from: DannyHamilton on March 31, 2014, 11:41:40 AM

The fee is not fixed. It is a fee per kilobyte. The space in the blockchain is the valuable thing that is being bid for by offering the fee.

Exactly.
But I have been wondering why the fee is per KB (round up) instead of per byte.

Chemistry1988

legendary

Activity: 1120

Merit: 1000

Quote from: Massimo80 on March 31, 2014, 08:02:48 AM

But more often than not the software (at least the reference client) requires you to send a fee, which can range from 0.0001 to 0.0002 BTC, or even more.

You can send bitcoin without any fee with raw transaction, but don't do it unless you really understand how it works and you don't mind waiting for a while for your transaction to confirm.

Quote from: Massimo80 on March 31, 2014, 08:02:48 AM

Asking for a 0.0001 fee when sending 0.00005 BTC is completely crazy...

If you really want to send 0.00005 btc (=$0.02), you shouldn't use bitcoin, or at least you should do it off-chain.

cr1776

legendary

Activity: 4228

Merit: 1313

Quote from: MarcoTC on April 02, 2014, 08:05:42 PM

In my opinion, the fees being based on KB is makes multi-signature transactions more expensive while I think multisig should be encouraged somehow.
It would be even better if multisig transactions were actually cheaper. Just to push it's use.

Maybe a better solution would be a % of the transaction amount. It's also easier marketing if Bitcoin only charges 0.000x% per transaction while banks, VISA etc. charge x%.

If you want to avoid spam, a better solution would be hardcoding the minimum transaction amount instead of hardcoding the minimum fees.

So if you start with 50 btc, spend 0.1 and 49.9 go to a change address, you'll pay a % of 50 btw in order to spend 0.1 since the miners can't tell what the "transaction amount" is?

You can always (so far) send a no-fee transaction.

MarcoTC

newbie

Activity: 22

Merit: 0

In my opinion, the fees being based on KB is makes multi-signature transactions more expensive while I think multisig should be encouraged somehow.
It would be even better if multisig transactions were actually cheaper. Just to push it's use.

Maybe a better solution would be a % of the transaction amount. It's also easier marketing if Bitcoin only charges 0.000x% per transaction while banks, VISA etc. charge x%.

If you want to avoid spam, a better solution would be hardcoding the minimum transaction amount instead of hardcoding the minimum fees.

grue

legendary

Activity: 2058

Merit: 1452

Quote from: Velkro on April 02, 2014, 04:49:49 PM

bitcoin transtaction fee should be based on current price
just like that, always small, not need to wait for updates then of clients

if only there was an unbiased way to get bitcoin prices.

Velkro

legendary

Activity: 2296

Merit: 1014

bitcoin transtaction fee should be based on current price
just like that, always small, not need to wait for updates then of clients

ScripterRon

full member

Activity: 136

Merit: 120

The 0.9.0 release reduced the minimum transaction relay fee/kB to 1000 satoshis (0.00001 BTC) and the dust limit is now 546 satoshis. That is 1/10 of the 0.8.6 amount.

Of course, until the various wallets update their algorithms, you won't be able to use the new minimums. And the delay is wise because the majority of the network is still running 0.8.x clients which would reject the lower fee. Just looking at the current connections to my server, I see 0.8.1, 0.8.3, 0.8.5, 0.8.6 and 0.9.0 with 0.9.0 still being in the minority (although it is improving)

Bitram

newbie

Activity: 28

Merit: 0

I expect fee transactions will reduced so it won't be that expensive.

DeathAndTaxes

donator

Activity: 1218

Merit: 1079

Gerald Davis

The critical resources being "purchased" is space in the blockchain. Charging based on anything other than the size of the transaction would be as misalignment of cost and price. It would be like a gas station which offered to sell you gas per mile. It might be attractive with those driving Hummers but ultimately the gas station would go out of business.

cr1776

legendary

Activity: 4228

Merit: 1313

Quote from: Massimo80 on March 31, 2014, 08:02:48 AM

...deflactionary currency...

Others have answered the fee questions and math issues, but some would argue as to whether bitcoin is deflationary (assuming you did mean deflationary) - it is certainly not deflationary now and will keep inflating for quite a while.

:-)

cp1

hero member

Activity: 616

Merit: 500

Stop using branwallets

Each kb in a block costs the same to mine, so the fee is the same.

Topic: Why a fixed transaction fee? (Read 1689 times)