Bitcoin Forum

Why not just raise the transaction fee to .0005 per transaction?? That will make the attack way more costly and the cost is still negligible to typical bitcoiners.

You are free to raise the minimum transaction fee that you pay. Just as I am free to raise the minimum transaction fee that I pay. Or not. If you do, your transactions will confirm promptly. If not, your transactions may languish waiting on some free block space.

Market mechanisms work. There is no problem here.

^This^.

It should be easy to set the transaction fee the user wants to pay.  If the wallet doesn't have this capability, hopefully the recent events will cause users to demand it.

This guy just payed $5 to get his coins into the chain.
https://blockchain.info/tx/4e4f96c5ba416961be347ffc496e8ce12046191ab7fb252e88966ce365d2bc5f

Eligius eventually saw the offer and took him up on it when they won a block.

transaction fees are voluntary and only sender should decide about setting the fee and I really don't like the idea, that fee will be somehow fixed as "minimal one". If somebody wants to prioritize TX, he may set higher fee, if he don't care how long it takes, he may set the lowest possible one.

this is just how market works, I hope, that devs will leave this part as is. what is problem these days is simply size of block..

The transaction fee set for blockchain can be adjusted while I pay a fixed fee of 0.0001 BTC. Now since there is an issue with blockchain, I need to raise it to 0.0005. I, however, don't want that too be a fixed fee for all my transactions and I want to decide the minimal fee every time.

You want to get your transaction confirmed faster than anyone else's? Raise your fee, you are free to do so. Raising the fee isn't necessary because the minimum fee is still feasible for the miners as incentives. Well, I can pay $1 for fees but I have the patience to wait for it to confirm, so I'll pay the minimum, which is $0.03 at current rates, Freedom to pay what you like for fees, that's what bitcoin offers that isn't available to any other payment processors out there.

It would be a problem for someone who just want send small amount of Bitcoin, lets say he want to send 0.0004. But the transaction fee is too high for him. So, transaction failed. 8)

I hope I don't tweak anyone's melon with this, but I personally am not too concerned that the guy who wants to transfer eleven cents needs to fork over more than that if he wants to get the transaction to confirm quickly.

I agree with most of the above, the cost system works pretty well and I don't think that changing it is the solution to the spam problem.   There are some good solutions but imo this isn't one of them.

Maybe it is what the attackers want, make miners rise their fees so transfer bitcoins will be less advantageous for the non spammers.

I wish there are other ways

I don't mind adding bigger fee but low fee is one of the good things about bitcoin that I boast to others whilst telling them about BTC and I guess if miners start delaying TXs with 10-20k Satoshi fee a lot than I would have to cross it from my list.

.0005 is still a tiny amount.

If the trans fee isn't an issue then why is everybody complaining about the spam issue? clearly something needs to be done and this is a simple way to make these attacks economically unfeasible.

As I have stated in two other threads already.

A large amount of the transactions on the network are spam in my humble opinion. If you cut out all the transactions under 5 cents that's like 90% of the traffic. (even before the attack)

As Bitcoin grows in popularity the network will naturally over time adjust the cutoff limits (minimum fees) so in times of heavy load micro transactions (less then 1 cent) will be "put on the back burner" till higher priority transactions with higher fees use the network and then when things calm down the smaller transactions will be validated.

I say let the market miners decide the minimum transaction fee rate.  If it gos above 50 cents then will have consensus for a small 2MB block increase before the sun sets.

The only risk right now is decoupling of the network (mico forks) because of propagation lag; this would only get worse right now if we increased the block size.

I, for one, welcome our new mining cartel overloads.

A possible fix would be to add optional code into our network nodes to change the minimum fee required per transaction size for propagation based on the current size of the mempool. The best part about this idea is its completely optional and dose not require a hard fork. But at the same time allows all users to vote on what the minium fee should be under heavy load.  As the mempool size increases the minimum fees necessary for propagation threw the network would rise accordingly and fall accordingly.  A simple sliding scale GUI could be added to all Full node clients for easy adjustments or be hard coded but configurable by the more tech savvy. This would prevent spam from ever reaching the mempool where upon it can't be deleted.

I think that this solution is not the best, because with this arrangement you give a worst condition to "basic user" or to some one that use BTC for the lower transaction fee, or for send smaller quantities.

.0005 is 15 cents dude.

I agree with this, for the moment I have also been increases fees from 0.0002 - 0.0005 to avoid transaction so long.

but what happens when .0005 bitcoin is worth $5 or more?  I don't like the current general impressions of what transaction fees should be.  I think they should cap out at .00000100 satoshi or be percentage rate based but an extremely low percentage... maybe .001 percent or lower.  High transaction fees at this stage are extremely self-defeating and do not do justice to the value of bitcoin.  

You can have your own choice about transaction fees, you can either fix it to the default value or increase according to your security and safety measures, if you fixed to default an attacker might know every time that the transactions going on intrinsic value and if you not the other way round.

they you cut down it to 100 satoshi or something, and unless 1 satoshi will be worth $5(which mean that everybody will be rich and will forget completely about this spam fest) i don't think there will be any problem, you can always customize it

Agree, the transaction cost tends to reflect the dust level of sending and is adjusted appropriately to circumstances
If a transaction wants to be put at a higher priority all that is required is a manual settings change to add on additional fees.

http://s2.postimg.org/r8qglgtah/btc_fail.jpg
why this is still doing? It seems like somebody really want to force people to use much higher fees than it should be! Somebody is hunting fees for his pool! It was common to use fee 0.0001 and tx was confirmed in several tens minutes. This is incredible, BTC is getting very bad reputation this way.
this tx is marked as high priority but it is waiting for confirmation 14 hours now!
http://s8.postimg.org/50keh94th/btc_fail2.jpg

I am reading blockchain twitter now and there is a tweet about using BTC at tavern in Greece on holiday. Do you really believe that they were able to pay for their lunch? now where network is rejecting thousands of txs? This tweet is only a part of propagation that there is nothing bad and in fact everything is fine, but no! there is nothing fine and something is going wrong.

The last stress test was mabe only the window for somebody - how to do it, how to break BTC network. Why did not you use test network for the last stress test??? Nobody would not be able to learn this kind of "attack" if you did so!
You are telling us that network is rejecting only dust/spam txs, but as you can see network is rejecting proper txs too. And not only several, but several thousands of them. Hurry up and do something! Why people run their BTCCore for supporting network if you are breaking it?

Miners are getting greedy and though we pat ourselves that BTC is decentralized and is not dependent on others like Banking systems are but without confirmations no one would accept any transaction and we ultimately rely on them to get our transactions confirmed.

And today we are thinking of raising the fee to 0.0005, tomorrow even that might not be enough for hungry mining farms and then we raise it 0.001 .. 0.002 .. and so on. And when the reward halves, what will be the fee then??

Tracing back to the original output (https://blockchain.info/tx-index/93739062/1), it is unconfirmed with a fee of 0.00000671BTC. Your transactions won't be confirmed till all the transactions that the input originated from confirms. No matter how much fees you pay, the initial transaction has to confirm first. This isn't a problem with the Bitcoin network.
Miners are raising their TX fees to reduce the possibility of including spam transactions in their blocks. This would only make attacks much more costly while allowing non-spam transactions to pay a reasonable fee. If the smaller pools decide to raise to a higher fee, it wouldn't make much impact if they don't generate blocks often. If the bigger pools does this, people would simply be leaving the pool for another.

Not entirely true.  Some miners (Eligis) will allow dependencies and sweep transactions together (CPFP).  So linking a high fee and low fee transaction can help, but only by about 4% (Eligis share).

sorry but it is a problem of BTC network. It should handle this txs original output with so low fee as it did many times and you can see it is doing right now too. There are txs with 0 fees and they are getting confirmations easy. And that is the problem of BTC that it is able to spend unconfirmed coins. And I did not do this tx, it is tx from the exchange where I bought this small amount via fiat. So the exchange website takes a part in this situation, because they have sent me BTCs which were not confirmed before. It is whole day and still propagation via 0 node, but high prio  :-[

what the hell I should do ? I am really sorry, but maybe LTC or other alts are better for such txs as mine. Going to try the litebit eu for direct purchase alts via fiat. Forget to buy BTC and then exchange it for alts!

I want the network to work! So I give full support to the idea of a raise in the minimum transaction fee, and to the minimal amount which could make a transaction. I suggest 0.005 BTC as the smallest allowed transaction. I understand many will be against the idea, but we need to find a solution to the current problem.

The solution is to voluntarily include a high enough fee to get your transaction included in one of the next few blocks to be mined. Nothing need be done about the 'minimum transaction fee' (what, pray tell, is that exactly?).

Bitcoin isn't designed to handle every transaction that pays a low fee. Your tranasaction can be considered as a free transaction as your fee is too low so we will look at the 50kb of free transaction. Your transaction has a priority of 6,116,086 which is very low. It would be ranked at the bottom of the free transaction anyway.

If you consider using alts, you might have faster confirmation but the adoption rate would be low and the price would be highly violatile. Bitcoin fees is only $0.03 and you would have a confirmation is a second or two.

that is the problem, it should be so but it is not. $0.03 is not enough these days, and TX is not confirmed with so small fee.

Nah. Yesterday I sent a transaction (https://blockchain.info/tx/34c7471118ca83f7da5abbc7e24421cac5df770538ee0aa059e7006179785d14) and it as confirmed at the first block thereafter. Note that transactions are ranked by TX fee so I have reasons to believe that it is easy for a 0.0001BTC/KB fee transaction to be confirmed at a few blocks after. If it is urgent, go for $0.06, it is significantly faster.

lol man with faucet i cant get sweet amount a day ,yeahhh people working faucet for couple weeks just to pay fee, it's bad man

I wouldn't raise the minimum transaction fee just because of transactions spammers, people should be well aware of manual fees and how the network confirmation works.

One advantage of Bitcoins are micropayments. With the Commission 0.0005 they simply disappear.
Everyone has to decide how much to pay commission.

It looks the fee's are already well overthought. Any change in current transaction fee has a massive influence in the future. Satoshi sure must have conciddered this part well. Basically he says that one day the transaction fee is surpassing the value of a mined block... That's how it is designed.

https://en.bitcoin.it/wiki/FAQ
"If no more coins are going to be generated, will more blocks be created?
Absolutely! Even before the creation of coins ends, the use of transaction fees will likely make creating new blocks more valuable from the fees than the new coins being created. When coin generation ends, these fees will sustain the ability to use bitcoins and the Bitcoin network. There is no practical limit on the number of blocks that will be mined in the future."

Bad idea, if the price going up, even current minimum transaction fee is too high for people.
                                                       

there is a lot of micro transactions being done using bitcoin, if the fee were to be risen they would simply not worth sending anymore.
besides you can increase it for your transactions but it should not be forced, and i am sure a lot of bitcoin users are against increasing amount of fees .

in fact, the transation fees is forced to be increase now,
if you giving a low transation fee, you will take ages for your confirmation.

That's very precisely what I want. I want all faucets to disappear, I've already written about that subject in this discussion:

https://bitcointalk.org/index.php?topic=980573.0

BTC hasn't been created for faucets, cam girls nor online gambling. BTC is a new currency for people to make business with.

What? Camgirls should get discount. People can use it for whatever they want.

Hopefully by then the price will be high enough, otherwise BTC would start losing one of its main charms which is the ridiculously low fees.

There is a problem if it's hard to know what transaction fee will guarantee inclusion in the next block.

It's not much of a market if none of the items on sale have a price tag.

Price isn't much of a factor. Even if you're paying the same amount of transaction few as now, the price would be much higher which wouldn't have any difference as it being low fee. The main concern now is to increase block size, add more transaction and allow miners to earn more fees.
It isn't about miner fees. Different miner accept different minimum fees and they mine block at different times. This would be impossible to predict. However, you can create a website to craft the average fee and set a higher than average fee for your transaction. Your transaction would at least confirm fine within the next few blocks.

This is why you should pick your wallet app wisely.  Some wallets will do statistical analysis of the blockchain to do just that... predict proper acceptance fee.  Bitcoin Core does this.  The relevant commands are estimatefee and estimatepriority.  It even has a slick little slider bar you can use to adjust it.

And when  0.0005 BTC becomes worth $1.00?

0.0001 BTC for extremely big amounts can be considered low...

yea, most likley paying 2% of what would sending would be a good fee even 1% some people just want to pay dirty cheap fees and the current state of transaction volume it takes a while to get in line, we at lease have a choice, either pay a bit more to get it faster or auto fees and wait a bit for it.

I don't see the solution in increasing the fees in a compulsory manner or the entire network. One of the advantages of bitcoins that is repeated over and over again is the low transaction fees compared to the traditional banking system. If we start implying artificial fees to fend off attacks, it can lead down a road where we might not want to go.

A problem with a fixed btc fee is the price of bitcoin. If the price would drop an attack could happen as it would be cheaper. If the price rises then legitimate users would be affected negatively. Percentage fees lead to a similar problem with bitcoin losing its edge over the banking system.

After all, wouldn't this problem be solved - at a cost of storage space - with increasing the block size?

Oh, well, that's pretty good then if it actually works ;)

Is there any list of wallets that do similar? Electrum afaik seems to have a fixed amount per Kb.

Yep, I've been watching the estimates, and it floats as blocks come in.

Most wallets have a fixed fee of 10,000 satoshi

Bitcoin-Qt will float the fee based on network traffic and the size of your transaction.

So on most wallets, a 500 byte transaction will cost 10,000 satoshi, on bitcon-qt, same 500 byte transaction would (as of this post) cost 166,612 satoshi if you ask for high priority, or 1442 satoshi if you ask for low priority.

I think fee prediction / suggestion is a MAJOR problem in most wallets out there today.  Fees should not be constant, they should have an economy of there own.  Cheap when the network is idle, expensive when the network is congested.  Know that will anger most of the bitcoin community, but it's free market economics... don't like it, use paypal.

Ah so qt actually does this? Gives an advice on how much fee is advised given the current network situation? I should go back to using QT...

If your current wallet allows to set the TxFee in any way, you can approximate this behavior without having to switch wallets.  Bit of a pain though... but here goes.

• Go to Statoshi.info Transaction (http://statoshi.info/dashboard/db/transactions) page
• Pick a time frame (top right) to average fees over ( 1 day; 12 hours; 6 hours ; 1 hour)
• Scroll down to the Fees / TX and the Satoshi / KB and write the Avg for each displayed at the bottom
• If your wallet lets you set the Fee / KB, then set it at 2x the Avg from (3), note, it's in bits
• If your wallet only lets you set a one time TX fee, set it to 2x the Fee / TX from (3), note its in satoshis
• Remember the goal here is to set your fee ABOVE average

This will require math and the ability to convert from BTC to mBTC to uBTC to bits to satoshis.  It will also consume about 3 minutes of your time, but worse of all, this could convince you to pay upwards of $0.20 in TxFees ;).

Hopeful one of the SPV wallet makers will get on the ball and stop making fees static.

Better: http://bitcoinexchangerate.org/fees

This is sweet, cheers!

Ahm, I use multibit at the moment and haven't found any option so far to change the amount of fees. I work something out. The site and the option in QT are both really nice, thanks again for the info.

If we do this we will 1. Kill off the incentive to use Bitcoin for micro transactions and 2. Play into the hands of the people doing this. If they see this is working to increase the miners income, they will just do it again. In the end it's the users of Bitcoin, who would be taking the punches. One of the advantages of using Bitcoin over other payment methods was cheaper fee's. Now you are saying, let's increase the fee's and lose this advantage?

the raise of the transaction fee to .0005 per transaction looks normal now because it will be around just 5 cents in USD but we all know bitcoin is not gonna stay at 300$ per bitcoins it ment to be rise if one day it is 5000$ per bitcoins and we are paying 0.0005 fee per transaction as you suggested then we are paying 5$ fee over per transaction which is i think to much

Yes, but as times and the price of Bitcoin changes, so will the transaction fees so long as they remain in balanced levels for both, the users and the miners.
So, I am guessing that when Bitcoin hits $5K the transaction fees will drop accordingly.

the fact that you can send a transaction with very small amount 0.0001BTC is one of the best qualities of bitcoin IMO. and i think it should not be changed.

you say rising it to 0.0005BTC looks normal now but you are forgetting about where it goes, to miners, and since the block reward is currently high enough and also the price of bitcoin is rising i say it is nothing normal about rising it.

But he means to raise the minimum fee so that the spamming attack will cost more. When he raises his own fee then the attack isn't changed.

jbreher In case the attackers are a combination of big miners then they will earn from spamming because what they invest will come back and more on top. When the minimum fee would be raised then they would get what they want without having to pay for it. Though that only when really miners are spamming.

I think your making the "Golden Ratio" argument, which was debunked...
https://i.imgur.com/snQEr4C.png

When spammers spam, all miners REVENUE go up, but only the spammers have the added EXPENSE of the spam.  Therefore they get less PROFIT.

i agree with you bitcoin were made to take out those killing fees that banks takes for any thing we wanna pay or tranfer to others,if we raise it we will come to a point that bitcoin would loose all interest since use it or bank will cost the same,on those time bitcoin would be dead.

I'm not worrying about this as long as the fee remains very low. We have free SEPA transfers within the EU, so it looks better than BTC but a .10 cents transaction fee is not a deal breaker, considering the other advantages of BTC.

That bank fee per transaction is low. Considering we pay $1 per transaction on our bank fees here, once you go over your bank plans allowable transactions in a month depending on the plan you are on.

SEPA doesn't just say that the bank cannot charge you more than it would for transferring to a bank in the country where you are sending from? I haven't really used it yet, not doing cross-border transfers.

About banking and fees: in e.g. Hungary I still pay transaction fees on bank transfer even if it within the same bank, what more even between two accounts in the same office/branch. On the other hand in Scotland I have never paid for any bank transfer. Nor for using ATM, no monthly fees, no withdrawal nothing at all.

Yes, I understand all the implications. However, history hath shewn that central planning is never as effective as the invisible hand of the marketplace. Block space _is_ a limited resource. Let us compete freely for it. If the miners clog all the block space sending big transaction fee transfers to themselves, they may crowd out regular users' lower fee transactions. Net effect: no new fees will enter the system. They will not make money. They'll eventually realize it, and roll back to a more market-appropriate level.

There is no problem here.

sepa may be free but then they charge you fees when you exchange with money outside the euro zone or when you withdraw or for you credit card(monthly) and other random crap

bitcoin from 10k to 100k satoshi can still work well and be very less expensive than fiat system

I think this will get faucet users down. Imagine you do captchas and close popups the whole day for 0.001 BTC and now there's a 0.0005 BTC fee.

I don't see that calculation debunking the theory. I mean when the biggest chinese pools stay together then they can fund the operation cheaply.

For sure they will raise profits for all other miners too but they most probably will have the ability to create more cheap miners for a lower price than small miners and pools. The profits from spamming would be miniscule, the same goes for the profits the other pools get. But an advantage in cost of creation of miners can be rather big. So even when others earn more, it could mean that the big pools still get more back in new mining hardware.

Though it looks like it weren't the miners, or they dropped their plan since the spamming stopped now.

The proof is that "Spammer Bob" makes 1821 BTC / day when spamming is off, but only makes 1810 BTC / day when spamming is on.

You may think that 11 BTC is no big deal for a miner, but the question remains... why would a miner engage in a behavior that reduces thier profit margin?

Nope, im not want to increase my fee transfer, because im not have more bitcoin to transfer it, om still use 10ksatosi for all my transaction

You can choose what fee you like to use. You can even send with 0 fee, but it's not recommended.

Yeahh i know it, but 0 fee transaction is like suicide, need more than 1day for get 1confirmation

n

That's the biggest minus of Bitcoin. Fucking long confirmation time.

The problem is, you don't know how much money they have. There is no guarantee that they couldn't keep it going. Also, it could be their goal for the standard tx fee to be increased by default.

This is my take on the situation as well. While I do see the need for transaction fees, I also see people complaining why bitcoin does not go more mainstream. While the .0005 fee may currently be 15 cents as someone pointed out, if I am going to use bitcoin for my everyday purchases that 15 cents will add up. If I make 30 small bitcoin transactions a day that would be an extra $4.50 just to use bitcoin. While I can see someone willing to pay $5.00 to move a $1,000+ transaction through a bit quicker, I cannot see a lot of people wanting to paying an extra $5 a day just to use bitcoin for their day-to-day activities.

This is not a bitcoin problem it is a math problem.  Bitcoin (for good or bad) relies on math.  The hope was that wallets would use math to calculate TX fees.  Many do... apparently... yours does not.  You might want to reconsider your choice of wallet programs, or do the math yourself.  If you don't want to do the math, I would set your TX fee commiserate to your disdain for the math required.  Again... switching wallets, or complaining to the wallet maker may be the easiest answer.

The transaction fee is a statistical question... As long as the entire mempool can fit in one block (which it can as of right now), you can go all the way down to 1 sat/byte.

The minimum TX fee to stay on the network (as of today) is 1000 sat / 1000 bytes (or 1 sat/byte).

If your TX has old coins, it is possible that the priority is high enough (as of today 57,600,000) to stay on the network with no TX fee

If the mempool is larger than one block (1,000,000 bytes) then your TX will need to compete with other TX to get in a block.

When transactions have to compete for blockspace the following sorting is used.

• The transactions are sorted by priority
• The highest priority (coin age) TX are selected until the block is 5% filled
• The remaining transactions are sorted by satoshi/byte
• Those payint the highest satoshi/byte are selected for the remaining 95% of the block space
• The rest of the TX are left into pool for the next sorting

If you studied statistics, this is straight forward problem to solve, but as others have stated about 76 times already... hard coding a fee value will not solve a statistical equation.  Best to get a wallet app that is mempool aware with sliding fees, or do the stats yourself if the mempool ever fills up again.

The sample data to solve this simple statistical problem by hand is conveniently provided here (http://bitcoinexchangerate.org/fees) if you scroll down and select "Raw Data (http://bitcoinexchangerate.org/images/bitcoin-fee-distribution.csv)"

PS: Admins... forgive my double post... just get weary of explaining it.

well bitcoin has a great potencial and the speed of confirmations just make it be more secure harder to hack currently a hacker can get into bank accounts ,can cloned credit or debit cards ,bitcoin as i know cant be  .

If history has shown something then that the invisible hand of the marketplace is very dangerous to the whole. I agree that central planning works even worse. But the marketplace brings out results like banks that trade with enourmous bubbles. They proceed because the others do so too. At the end they crash and bring misfortune to the whole market.

The thing is, when miners find a way to earn more, then they will do. SPV mining for example. So when it's possible to get an advantage and you even harm bitcoin, then they might do it. When they made their nice stash and bitcoin is dead then they have their money already. And they can be sure that most others won't do it, in order to protect bitcoin. So they can use it and know that they don't hurt bitcoin big enough.

I think that graphic includes some errors. First, with spam the fees are higher. Second, a potential mining corporation build out of all big chinese miners, could be more than 60%. So i think it's possible to increase mining profit with that. The most profitable it would be for the miners not spamming at all while the others spam, thats true.

I still challenge anyone to put any data (not formulas) on paper (real or imaginary) that shows how a spammer can receive more in fees than he spends.  Laws of entropy and conservation of energy dictate that a system can not produce more energy (money) then it consumes.