[Nov 2024] Fees are low, think about Consolidating your small inputs @2.0 sat/vb

[tranthidung]

Chill out mate, wait a couple days and you'll be good. Look at this graph: https://jochen-hoenicke.de/queue/#BTC,2w,weight

You can compare the spike right now to the one on 09/30. It will last a few days but it will be nice and cheap soon, I'm sure

It is not true. The increase of demand on bitcoin transactions will be reflected on the bigger size in the mempool and higher fee rate for 1 vMB from tip of the mempool.

Usually it will occur when Bitcoin has big movements, ups or downs. With recent movements of Bitcoin and the good history in October, I think you will have to accept higher fee rates in this month.

Obviously, no matter the average fee rates are, you will have cheaper ones in weekends, especially on Sundays and early of Mondays.

[pooya87]

It will last a few days but it will be nice and cheap soon, I'm sure

You can not make that prediction.
You see the main reason why fees go up are usually either a spam attack or a big price change (rise or fall). The reason for the current fee spike is the later and when you say "it will be cheap soon" you are actually predicting that price will soon become stable again which is not something you can predict specially now that price is rising faster than before breaking major resistances.
In other words if price rise continues this way, it will be a while before we see 1 sat/vb again.

[NeuroticFish]

wait a couple days

This would basically mean that the price changes will slow down again.
It's too difficult and risky to make such predictions, I don't believe that yours will be accurate.
I am sure that the mempool will clear up again, but I surely won't risk telling how long will take until that'll happen, especially now when the price is expected to go towards new ATH.

[n0nce]

You can compare the spike right now to the one on 09/30. It will last a few days but it will be nice and cheap soon, I'm sure

There is one big difference between that spike and this one, the fees paid by these 40k transactions waiting.
During the peak of that previous one, a 5ast/vb would still have only around 0.2-0.5 Mvb to compete with, right now the same rate of 5sat would put you under 13Mvb.

Those were consolidations or tumblers or whatever, but they went for the minimum fee, this one might actually be real usage by a lot of poeple, not just a few services, the 20k confirmed transactions difference means they are smaller in size so more likely to be single payments.

I don't understand why it should be cleared less quickly though. That's purely a matter of [#new transactions], mempool size and [#mined blocks / unit time]. Higher transaction costs shouldn't affect the speed of the miners in a negative way; if anything, more people would mine if there is more reward to be had.

Regarding BTC trading and stuff, I'm not sure. Yeah there might be a correlation, but I'm not sure that the majority of people actually transfers their funds on and off the exchange when buying & selling. While it's not good practice, especially higher-frequency traders tend to leave funds on exchanges to avoid deposit / withdrawal fees all the time.

Finally - Sorry guys for speculating / predicting how long it will take to clear the mempool, of course I'm just guesstimating based on my experiences We shall see how much it really takes, of course nobody can predict the future However, size has already come down a bit, actually! From almost 30 vMB it went under 20 vMB in a matter of hours.

[LoyceV]

I don't know where you got the idea that we should be able to have an empty mempool next weekend).

Experience: fees are usually lowest on Sunday.

especially now when the price is expected to go towards new ATH.

I'm always expecting that Still, at the previous new ATHs this year, fees were nowhere near what they were at the end of 2017. If I zoom out on Johoe's Bitcoin Mempool site, there was a massive amount of blue (low-fee) transactions unconfirmed around April this year, while 4 years ago there were many high-fee transactions unconfirmed for a very long time. I remember paying around $25 (much more than 0.001BTC) back then for a small transaction (one input, one output).
It probably helps that exchanges like Binance trick people into withdrawing their own centralized shitcoin tokens instead of real Bitcoin. Those suckers keep fees low by accepting basically counterfeit Bitcoin.
And maybe Bitcoin users (and companies) in general have become smarter selecting fees.

[NeuroticFish]

I'm always expecting that

LoL! So optimistic! 

Still, at the previous new ATHs this year, fees were nowhere near what they were at the end of 2017.

There are many differences between 2017 and 2021, from SegWit to centralized services batching transactions and also using SegWit. Also I still believe that the end-of-2017 situation was caused by a spam attack too.
I compare 2021 only with 2021, maybe 2020 ... and when the price jumps, the mempool gets congested rapidly too. And for a new ATH this year.. there have to be a couple more price jumps during these few months.

It probably helps that exchanges like Binance trick people into withdrawing their own centralized shitcoin tokens instead of real Bitcoin. Those suckers keep fees low by accepting basically counterfeit Bitcoin.

Exchanges always asked for unreasonable withdrawal fees lying that they're tx fees. And this whether the tx fees were big or small.
Those who withdraw on smaller fees for various uses.. well done; those holding whatever token because of that.. they deserve their fate, really...

And maybe Bitcoin users (and companies) in general have become smarter selecting fees.

And using SegWit, and batching transactions, and keeping an eye on the mempool before making transactions... yes, I also expect users and (especially!) companies learn and improve. It's after all, about avoiding to waste money too.

[stompix]

I don't understand why it should be cleared less quickly though. That's purely a matter of [#new transactions], mempool size and [#mined blocks / unit time]. Higher transaction costs shouldn't affect the speed of the miners in a negative way; if anything, more people would mine if there is more reward to be had.

Ironically, although the price per BTC is up and fees are also up something weird is happening, probably is just luck but the block time is increasing, there were only 121 blocks mined in the last 24hours.

As for why this is different is because of what was causing it, in the first situation you have a dump of 40k in a matter of seconds which has started to be chewed immediately as there was no real usage on top of it, in this one you had poeple adding transactions over the average capacity for a while, as you said, a matter of [#new transactions] added.



You can see in the above example how peaks that are made just from one dump thousands of 1sat/b are clearing far faster and have just a sudden peak and die down while the one that is caused by random poeple rushing and trying to get their tx as fast as possible like this one has a different pattern.

That being said I'm quite sure the mempool will be empty by Saturday unless something really weird will happen with the hash rate or the price.

And maybe Bitcoin users (and companies) in general have become smarter selecting fees.

Just because everyone would use the lowest fee it won't do a thing for the mempool, instead of having 100k transactions from 1 to 500sat/b you will have 100k with 1-2sat/b the one paying the minimum will wait for just as much in both situations.

[LoyceV]

the block time is increasing

There's a site for that too: bitrawr.com: it shows there are actually a few more blocks found than expected since the last difficulty adjustment.

[stompix]

the block time is increasing

There's a site for that too: bitrawr.com: it shows there are actually a few more blocks found than expected since the last difficulty adjustment.

Hihi, that's what we use instead of chart buddy in the difficulty speculation thread 
You can see in this post of that the pace is actually going down slowly and steadily, we started the epoch with a 105-110% pace and we slowly dipped into negative. I would assume something really bad happened with some miners somewhere, some blackout, grid failure or something

But the ones paying attention probably have done what they could to avoid paying higher fees, we had months for consolidation, opening channels, securing your coins in cold wallets, we had half a year to buy that damn coffee with onchain transactions....
Or maybe that's the cause...it's getting colder and poeple buy coffee and hot tea with bitcoin...
Mystery solved, case closed!

[n0nce]

I don't like to be the one always saying 'told you so', but look, 2 days after I said 'wait a couple days', 1sat/vB transactions are mined again..

[Poker Player]

I don't like to be the one always saying 'told you so', but look, 2 days after I said 'wait a couple days', 1sat/vB transactions are mined again..

Are you always right?

Let's see if we don't have another nullius here.

I don't know if you have a crystal ball with which you can predict the future, but the assumption that the fees were going to remain high was to assume that we were entering a more or less sustained price increase, which seems quite reasonable according to what has happened in previous cycles and the predictive models that point to the fact that in this last quarter we are going to break the $65k ath and go beyond it. Minimum $100k.

As pooya87 says: 

You can not make that prediction.
You see the main reason why fees go up are usually either a spam attack or a big price change (rise or fall). The reason for the current fee spike is the later and when you say "it will be cheap soon" you are actually predicting that price will soon become stable again which is not something you can predict specially now that price is rising faster than before breaking major resistances.
In other words if price rise continues this way, it will be a while before we see 1 sat/vb again.

[Timelord2067]

Prices can also go up simply by people withdrawing their bitcoin from the market (without actually selling) leaving others asking higher prices and getting those ask prices hence the number of transactions (and corresponding fees paid) aren't surging in size.

[n0nce]

Are you always right?

I'm not, I never said that

The assumption that the fees were going to remain high was to assume that we were entering a more or less sustained price increase, which seems quite reasonable [...]

I am seriously interested if there is any data that backs this up; I mean the idea that there is not only a correlation between mempool size and Bitcoin price increase, but also that it's the causation for it. Why I ask this? Because (unfortunately) many people leave their funds on the exchange. So even if many people buy or many people sell, if it's all done on-exchange, there are no Bitcoin transactions needed. Now, this is only a hypothesis as well; I don't have numbers for it.

In case you don't have any, I would look more into this topic when I find the time. It would be really interesting to see how many people actually hold their keys.

[jackg]

The assumption that the fees were going to remain high was to assume that we were entering a more or less sustained price increase, which seems quite reasonable [...]

I am seriously interested if there is any data that backs this up; I mean the idea that there is not only a correlation between mempool size and Bitcoin price increase, but also that it's the causation for it. Why I ask this? Because (unfortunately) many people leave their funds on the exchange. So even if many people buy or many people sell, if it's all done on-exchange, there are no Bitcoin transactions needed. Now, this is only a hypothesis as well; I don't have numbers for it.

There must be some data showing it because a volatility rise normally always increases the fees (even some crashes have afaik).

The first wave of the current market sent us up to about 150 sats per byte. There also seem to be people tracking outflows from exchanges (although apparently glassnode is a paid for service) there might be ways to programmatically determine this if you wanted to though by looking at volume increases from an exchanges cold funds or making a withdraw (as a lot will be batched together or traceable).

[stompix]

I don't like to be the one always saying 'told you so', but look, 2 days after I said 'wait a couple days', 1sat/vB transactions are mined again..

And just 4 hours alter the mempool still has 14Mb and the last block minimal fee was 12sat/vb 

Again, it's a different situation than the one at the end of September, there was no rush to then to send transactions, there was no continuous spike in fees.
You can see that the previous spikes have not caused anything remarkable on the fees, just slight variations, these one is a bit more serious, the first one in months.

Not that the weekend isn't going to clear it up but still, it's different.

So even if many people buy or many people sell, if it's all done on-exchange, there are no Bitcoin transactions needed. Now, this is only a hypothesis as well; I don't have numbers for it.

You can start from here 
https://bitinfocharts.com/bitcoin/address/1NDyJtNTjmwk5xPNhjgAMu4HDHigtobu1s
This is Binance's known address for withdrawals, if anything has changed in the pattern of usage it must be here.

[n0nce]

You can start from here 
https://bitinfocharts.com/bitcoin/address/1NDyJtNTjmwk5xPNhjgAMu4HDHigtobu1s
This is Binance's known address for withdrawals, if anything has changed in the pattern of usage it must be here.

Okay, so looking from June till now, it seems it went down in a relatively straight line until August, then up again until now. No insane movements in October, when looking at the big picture.
Looking at the mempool chart, ever since July, when the mempool first 'hit the floor' in a long time, there have been lots of little spikes in relatively regular intervals, but purely pattern-based, the two charts don't seem very correlated.

However, even if we saw a correlation, it still wouldn't prove causation.

Binance address:


Mempool:

[Poker Player]

However, even if we saw a correlation, it still wouldn't prove causation.

I will partially agree with you that causation is probably not proven. But it is very difficult to demonstrate causation, and even more so through the simple observation of statistical phenomena.

I understand that the other forumers who affirm that it exists is by the observation of many years. If you see that a phenomenon happens only once after another, you cannot establish a causal relationship. But if years go by and you always see that one phenomenon happens after the other, and you've never seen an exception, you can begin to think that there is a causal relationship.

But if you mean proving it, as in being absolutely sure, it probably cannot be demonstrated.

[pooya87]

I am seriously interested if there is any data that backs this up;

It is just logic and observation.

I mean the idea that there is not only a correlation between mempool size and Bitcoin price increase,

Not price increase, but price change which can be either increase or decrease. Also not all rise and falls would cause a spike, it has to be semi-significant. For example a 5% drop or 5% rise is not going to cause a fee spike but a 20% will.

Because (unfortunately) many people leave their funds on the exchange.

True but when we are seeing a significant "movement", specially after a period of stability, there are a lot of traders who move their funds too. For example there is a large number of day traders who leave bitcoin market when it is stable to go to shitcoin market and pump and dump altcoins. As soon as bitcoin starts making rising they dump those shitcoins and move back to bitcoin market (which is why all altcoins dump against bitcoin). In many cases this requires a bitcoin withdrawal.

So even if many people buy or many people sell, if it's all done on-exchange, there are no Bitcoin transactions needed.

That's true and we are ignoring those people as they have no effects on-chain. We focus on the other group who move their coins to and from exchanges to sell or have bought respectively.

[n0nce]

Because (unfortunately) many people leave their funds on the exchange.

True but when we are seeing a significant "movement", specially after a period of stability, there are a lot of traders who move their funds too. For example there is a large number of day traders who leave bitcoin market when it is stable to go to shitcoin market and pump and dump altcoins. As soon as bitcoin starts making rising they dump those shitcoins and move back to bitcoin market (which is why all altcoins dump against bitcoin). In many cases this requires a bitcoin withdrawal.

Okay, yes, this makes sense to me; that if you're not going purely between BTC and fiat, there's withdrawals needed.

Anyhow - 3 days after I stated 'wait a couple days', mempool is empty!


It seems price has stabilized as well, so this would be evidence for that theory:

[LoyceV]

This deserves a quote here:

Fee estimation:

Low fee:

Code:

[img]https://blackhatcoiner.com/lowfee/[/img]

Medium fee:

Code:

[img]https://blackhatcoiner.com/mediumfee/[/img]

High fee:

Code:

[img]https://blackhatcoiner.com/highfee/[/img]