Bitcoin Forum

Hi. I need a wallet that support sending bitcoin with 0 fee. I tried electrum but i get that server can send transaction with 0 fee, i tried blockchain info but say is needed minimum 1sat/byte fee.

Anyone know any wallet that still support 0 fee?

Hi,

Even with 1sat/B, your transaction would not even be relayed by most of the nodes, what do you try to achieve ?
https://p2sh.info/dashboard/db/fee-estimation?orgId=1&from=now-7d&to=now

I think i’ve read somewhere that you can use Bitcoin core with and set to 0? (Or 0.00000001)That should allow the broadcasting of a transaction with little to no fees at all.

https://bitcoin.stackexchange.com/questions/48235/what-is-the-minrelaytxfee

EDIT: This probably is no longer possible as LoyceV suggests?

Older wallets can do this, but the current Bitcoin network doesn't accept it.

Transactions with 1 sat/byte fee confirm fast most of the time (https://bitcointalk.org/index.php?topic=2848987.0).

If you want to pay less fee, this may help you in the future:

Can you explain what exactly you're trying to accomplish?

Not exactly sure what you are trying to do but a miner will have no incentive to include your 0 fee transaction to the block. As long as there are transactions with higher fees (and there always will be) your transaction will unlikely be included or will take forever to do so.
Unless you are the miner and you are trying to include your own transaction without a fee, is that what you are trying to do? I think that is possible but not sure how to do it.

There still appear to be zero-fee transactions if you check even past 24 hours, there were 2 transactions but as Pmalek says above, probably just miners including their own transactions in their blocks.

Surprised myself to be honest, if I recall last I looked, still scores of 0-fees a day even if the total has been going down ever since I first started checking in late 2016.

Myself liked to do 0-fee txs, just for fun. I think I even have an Electrum installation from 2.x that allowed it, but really, there is hardly a need when 1 sat/byte confirms quickly and hardly costs anything especially with Segwit. Once Lightning gets underway, we won't even be thinking of fees.

Do transactions not expire from the mempool then if 0 sat per byre fees aren't accepted. I'd have fun sending with 0 sat per byte fee although 200 SATs isn't much...

If my wallet gets made, I'll be the irresponsible one and put 0 sat per byte fee limit on the software config so I can have cheap transactions (then do one of my strategies when it doesn't confirm to unlock the funds) - it'll probably get blocked by normal Bitcoin nodes anyway if I do too much in adjusting the protocols ;).

Miners mining 0 sat per byte fees is scamming if they're from a pool. I don't know why I didnt make one ;D.

I remember the days when an input had to have "medium priority" to be able to add other inputs up to 1 kB in a zero-fee transaction. It's too bad the concept of "coin days destroyed" was bandoned in favour of higher fees. It was a good feature, as it rewarded using older funds, while discouraging rapid (spam) transactions.
I gave up on it before it was abandoned, after some of my transactions got stuck for 2 weeks and I had to double spend them to get them confirmed.

If you really want to save on transaction costs you can try to use hybrid wallets that have an off-chain transfer within their system such as Coinbase, there is no transaction cost or any network fees involve and the BTC transfer are always instant. The only catch is your recipient must also have the same wallet as yours which in this case should also be a Coinbase wallet address. These off-chain transfers come in handy especially when you don't want any network fees involve.

you can still create a transaction with 0 fee with all the wallets as long as you are not using their user interfaces because that is where the blocking takes place. for example in Electrum the wallet doesn't allow you to set the fee to zero in the interface but you can still make an unsigned transaction with 0 fee using any additional method and then sign it using Electrum. then you again will be stuck in broadcasting it because the "servers" reject your tx.

one way of doing it is to preview your tx before signing it, then copying the hex and editing the amount field to include the fee in it (increase it +fee) and then giving it back to electrum to sign. of course you mustn't do it without knowing what you are doing or you may end up messing things up.

Do not manually create or manipulate Bitcoin transactions unless you absolutely know what you are doing (and think twice even then).

You don't want to end up as the guy spending 10 BTC in mining fees on a zero BTC transaction...

Mining fees are not stated on transactions.
There's the input data and the output data that is necessary and only the output says the specific amount afaik.

P.s it was 15btc ;D
If you import a transaction it goes into the preview window, you should be running transactions that way instead to make it more secure and know exactly what you're doing a bit better.

I entered Bitcoin just at the tail end of that... still remember priority values on blockchain.info. I too thought the idea of older coins having some weight in priority was a good thing, but was never able to enjoy it myself. All coins acquired were ever fairly new. Suppose scaling upgrades mean spammers have to get ever more creative now.

I believe the antpool "acceleration" submitter still works even for 0-fee tx now, so in the unlikely event someone inadvertently managed to make such a tx, there's still that one avenue for confirmation.

To be more precise, the transaction fee is implicitly stated -- it's the difference between the sum of all inputs and the sum of all outputs :) Still easy to mess up though

Good point about using Electrum's preview window!




Is Antpool's accelerator free though? The only free transaction accelerator I'm aware of that actually is working is the one from ViaBTC and that one requires a minimum transaction fee.

No and it also looked broken yesterday. It wouldn't work for me when I wanted to see what the fee would he to push something through.

It's also worth noting that once you sign a zero fee transaction in electrum. No one else will broadcast it unless you do a nonstandard transaction on Bitcoin core which isn't something to do lightly.

You can double spend a no fee transaction also. You guys are forgetting that a node won't broadcast a transaction that is a double spend but if it's a zero fee transaction most nodes don't accept it so it can be double spent in the mempool as long as you send it to a node that accepts it. Only one transaction will ultimately be added to a block though.

It's always been free, at least from when I knew about it in 2017. I seem to recall there being some conditions, if I'm not mistaken, it shouldn't be a double spend, but there wasn't a min fee as I used it myself several times to push txs for people getting stuck for days/weeks, including 0-fee ones.


Checking it now (https://antpool.com/user/prioritiseTransaction.htm?m=savePrioritiseTx) and it doesn't seem to have changed from when I last remember using. Still free, and able to accept even 2/3 txs. Just submitted a couple of txs and it got accepted for "acceleration". They'll probably naturally get confirmed for others long before antpool though. They've switched off quite a bit of their rigs recently haven't they?

Will try and remember to do a 1 sat/byte tx during next spike and see if antpool will confirm it... and update. Likely to be a few weeks til another mempool spike happens.

Not true, in the last 6 months I did nearly all my transactions at 1sat/B and they usually confirm in about 1 hour or less.

What you should NOT do is use zero, that won't move at all. I remember some people on IRC saying its possible to use half satoshi per byte or so tho. But if you use "twice" of that (ie. 2sat/B) it often gets processed by the next block! (10 mins); or sometimes by 2~4  blocks, but you get the idea...

Electrum (https://electrum.org/) allows the "replace by fee" option in case you get desperate and something gets stuck for weeks, probably other wallets allow this too. It lets you increase the tx fee to an already broadcasted transaction, so you can spend some more satoshis to speed it up yourself.

Does rbf require your input to be used with a smaller output?

I'd be interested in trying to broadcast a 0 fee transaction but I'll have to nail the handshaking process and automate it because I reckon a lot of nodes will reject it.
0.1 sat/byte fee would be interesting too to see what happens,that's about 20 sats for a standard transaction then. It would be nice to see low fees again soon though.

Currently, Bitcoin Core does not allow users to relay transactions with fee lower than 1 sat/b. Fortunately, this is going to change in the next release of Bitcoin Core (0.18.0). Here (https://github.com/bitcoin/bitcoin/pull/13922) you can find more information.


There should be the same inputs and outputs with no change to the amount of coins sent + a new input which will be spent on the increased fee and an extra output if there are any coins left from the new input. However, there are different variants of RBF (https://en.bitcoin.it/wiki/Replace_by_fee).

Anything above 0 is probably good I think. 0.1 would be processed last, but at least not dropped like the zero tx often are...
For the future it is good that wallets allow even lower tx fees, as long as its not zero :)

Has anybody tried to send a zero fee transaction? I might have a go if nobody has attempted it.

So does this mean that it is still possible? If there is a pool of miners who accepts it?

I think we can try because we can set a new tx fee for an already broadcasted transaction anyway.

i just created one, now the problem is with broadcasting it. blockexplorer usually accepts them since they don't have the minrelayfee set but they are rejecting this because of it is creating a SegWit output.
i'll try sending it to a couple of nodes at random and see if they reject it later. for the time being here is the raw hex, feel free to broadcast it:

No luck here, all rejected for the same min relay fee error. Checking your address of input and it seems you've had no luck broadcasting it either. Only way now is for a miner to manually include it I guess and since there were 4 0-fee txs confirmed in the last 24 hours, then there's at least 1 big guy still doing it. Probably just their own txs though. Might spend some time a bit later tomorrow if I remember, to identify the miner/miners who included those txs!

Have you tried bitmain and bitfury? If anyone accepts low or no fees it's going to be them or slush.

I've tried that a few months ago when the network traffic was quite low
I pushed it to a few nodes, some nodes rejected immediately while a few accepted but rejected eventually
those that accepted the tx, removed it after a few minutes... I can tell by checking that node's blockchain explorer

can you tell me their txids? which pool found the block?
would be interesting to try push 0-fee tx into that node and see whether the tx gets through or not :D

Just got my lazy ass to check and all the 0-fee transactions in the past 5 days including today and there's a very curious 2 txs a day, each in 1 block, and all found by F2Pool all between 2.30am to 6.15am UTC... the coinbase data all separate names.

Of those pair of 0-fees, 1 looks like a consolidation, and 1 is split to over 3,800 outputs. So this looks like F2Pool just paying its own participants, so no fees necessary.

Example block 557390 (https://www.blockchain.com/btc/block-height/557390), tx IDs d8e18f5022434269cf59603a70594e5d78485208290a106135ef83f7341cbed8 (https://www.blockchain.com/btc/tx/d8e18f5022434269cf59603a70594e5d78485208290a106135ef83f7341cbed8) and 035d37ece0c7bb52e2f6a9f2c16d06bd6ef6840a73e25c1ffc5357d56eaf0452 (https://www.blockchain.com/btc/tx/035d37ece0c7bb52e2f6a9f2c16d06bd6ef6840a73e25c1ffc5357d56eaf0452)

The 4 other blocks: 557248, 557080, 556931, 556793.

If you want to try that by sending the Satoshi to the Crypto Coin Tree Segwit address, then I'll pay your the Sterling value if it works.

It won't actually work unless I get you to recompile the source :(.
I'm guessing no one else has actually recompiled the source themselves after that edit meaning it probably won't work, unless you fancy downgrading to 0.10.2 or something (I have a copy on my hard drive).

(Don't come here and say you just have to set the mintxfee to zero because IT DOESN'T WORK --aimed at those who believe it does - not jet cash - I wasted half an hour making that transaction if I'm honest electrum does some weird stuff)

---

@Jet Cash, I was going to send you this:

[tx hash removed, @jet cash check your pm on fittotalk.com]

A happy zero fee transaction :).
Someone push that through in a couple of years when the min tx fee is 10 sats per byte because the bitcoin devs are trying to price us out of the market :'(.
"Ooh it will be reduced when v 0.18.0 comes out" they're saying. We've had a release since then of one of their fixes, and it can't be that much to change...

---

I'd like to send $4 without a fee, why is that a stupid ask?

A point release is for bug fixes or soft fork deployment parameter, not major feature changes. Changing the minimum relay fee rate (and everything else that goes with it) would be a major feature release. It definitely would not be first released as a point release. Just because your favorite change isn't in a point release does not mean it is not going to be in the next release.

And just because you think "it can't be that much to change" does not mean it is not. There are a lot more fee rates than just the minimum relay fee and changing fee rates affects a lot of components.

If you can send $4 without a fee, then I can send 100 million 1 satoshi transactions back to myself without a fee. And I can do this constantly and indefinitely forever flooding the blockchain and the mempool. This kind of spam is detrimental the network and has potential to cause major problems. That's the whole reason there's a transaction fee and a minimum relay fee: to prevent this sort of spam.

---

When changing anything in any of Bitcoin's consensus or transaction relay policies and protocols, you can't just think about the short term and how a change affects just you as a user. This is a distributed and decentralized system where major changes to the protocols affect the whole ecosystem. You have to look at the big picture and see how a change affects the entire ecosystem as a whole, not just the individual.

what is all this talk about downgrading upgrading,... can't you already change minrelaytxfee to another number in your bitcoin core settings? i don't have bitcoin core anymore (because of HDD problems, it being full!) to check it out. (as in calling -minrelaytxfee=0 in your console).

you are free to send it with no fees (as i did with the transaction that i posted above) and the rest of the network are also free to chose to either reject or accept it. it just happens that majority of the network is currently rejecting 0 fee transactions and miners love this :)

You get an error with setting mintxfee to zero. Anything higher is fine but... I'm guessing you can't go lower than 0.00000001.

So the second point is also a no to the first part but a yes to the second. I gave up trying to put it through...

I tried sending a few Satoshi to my CoinBase account with a 1 Satoshi/Kb fee, It keeps updating the fee to what I assume is a minimum rate. This applies to 0 Satoshi as well.

Can you send me the raw unsigned transaction hash? I'm not sure how you get it out of core.
If you used electrum I could help put it through much easier.

Instead. Use the mintxfee=0.00000001 when starting the client - I'm not entirely sure on the command pooya said minrelaytxfee so you could try that one too with the operand of 0.00000001.

I kept changing the fee, so I didn't send it.

If I get the chance tomorrow, I'll have a go at changing things, but I might have to fry another pizza instead. :)

Did you manage this?
I need 2 hashes and the amount sent for each (same inputs and same outputs in both just a slightly different fee and amount spent)?

I always tought that btcd allowed the minimum relay fee to be 0:

https://github.com/btcsuite/btcd/blob/86fed781132ac890ee03e906e4ecd5d6fa180c64/mempool/policy.go#L61

Seems like there is no "hardcoded" limit, if the minRelayTxFee == 0, then minFee = 0

source: https://godoc.org/github.com/btcsuite/btcd

BTW: here's a 0 fee non-segwit transaction i made in case somebody wants to try to broadcast it ;)
ps: don't try to include it into a block you're currently trying to solve just to do me a favor, the transaction is using an unspent output funding the exact same address that was funded by the unspent output, so including it into a block would have 0 effect.

I've had this discussion with a couple of new users just last week, If i'm not mistaking 0 fee transactions aren't illegal at the protocol level, it's just most node's and most wallet's implementation that stops 0 fee transactions to be created/broadcasted/... If you use the console of core, and use createrawtransaction signrawtransaction AND broadcast to (for example) a btcd node that has 0 fee relay AND by sheer luck end up with a miner that also uses btcd with 0 fee option AND the miner has non-full block AND includes your tx (not sure if he'd have to rewrite his software to allow this), there is a very small possibility of creating, broadcasting and getting a 0 fee transaction into a block... (a very, very, very, very small possibility).

If you are considering a wallet offer of 0 transaction also consider the safety of such wallet. I believe for those authorizing the transaction the transaction fee get them paid. If you dont want to pay anyone for keeping your bitcoin and for sending it, then think about how too cheap and unsafe you are requesting for everything.

I'm not following your train of toughts here...
The transaction fee is not paying the developers, the miner can add the sum of all transaction fees of all transactions he put in the block he's trying to solve to the current block reward. The wallet developer or maintainer gets nothing.
There have been a couple secure wallets in the past that charged a small fee (not a mining fee) for using their wallet (multibit HD), but nowadays i don't think any truely vetted community supported wallet still charges a fee.

Also, if you're using a wallet where a thirth party is keeping your funds, do yourself a favour and migrate to a decent wallet straight away.

This makes no sense at all.
A block reward is already paid and people who want high transaction priority will pay higher fees for it.