Maybe one thing that I might add is that if the transaction fees are currently coming back down and eating through the 20 sats per vbyte backlog, then we might be able to conclude that whatever was being done in the last 3 months or so in order to keep the fees high might not be sustainable (and maybe costing the attackers too much), and it may well be better to not change anything in any kind of substantial ways rather than to be forced (or rushed) into some kind of a change that is largely being artificially justified by the three month backlog in order to cause people to speculate that something in bitcoin is broken or some urgent change is needed (when it is not)... and so yeah, there might not be as much urgency as some complainers (cough, cough franky1) are trying to spout out about.
observing mempool you will see the fees didnt come down to allow blocks to eat through the 20sat/vb backlog
Maybe my language was not precise enough for uie-pooie..
I did not say that they ate through the 20 sats per vbyte territory.. I had meant to say that it has been in the process of eating through.. and so maybe you are getting caught up upon language in order to quibble about meaningless matters.
My subsequent chart had shown that there are ranges of backlog, and I had ONLY provided the last week which shows that largely the widdling away of backlog has been in the 20-25 sats per vbyte range, and prior to that it was in higher ranges such as 25-30 sats per vbyte, 30-40 sats per vbyte, 40-50 sats per vbyte etc etc etc...
So we can even zoom out further on that same chart that I had already provided earlier and see the backlog continuing to get widdled away, including that currently, in the past week or so, we having the 20-25 sats per vbyte backlog widdled away.
instead the 20sat/vb got purged. then some 16sat/vb added and got purged , then some 20sat/vb added and got purged because the min went upto 21 before new fresh tx added to mempool from 16sat/vb up.. and then purged again back to 20
- fastestFee: 26 sat/vB
- halfHourFee: 25 sat/vB
- hourFee: 25 sat/vB
- economyFee: 25 sat/vB
- minimumFee: 16 sat/vB
- fastestFee: 26 sat/vB
- halfHourFee: 25 sat/vB
- hourFee: 25 sat/vB
- economyFee: 25 sat/vB
- minimumFee: 20 sat/vB
- fastestFee: 24 sat/vB
- halfHourFee: 24 sat/vB
- hourFee: 24 sat/vB
- economyFee: 24 sat/vB
- minimumFee: 16 sat/vB
- fastestFee: 28 sat/vB
- halfHourFee: 26 sat/vB
- hourFee: 25 sat/vB
- economyFee: 25 sat/vB
- minimumFee: 21 sat/vB
you can see that jochen-hoenicke retains tx below 20sat/vb because HIS mempool is set at 550MB instead of standard 300
but by seeing the straight line(unnatural) of the 20-21 colours they are purged by majority which means only jochen-hoenicke is retaining them hence the straight line. where most natural transactions of 300mb mempools are purging cheap fees
(if very cheap tx were being added you would see wiggly lines of tx's coming and going as they get relayed in - confirmed out)
I am not presuming that transactions are being processed for lower than the backlog amounts, so right now, if you are trying to be economical as you can, then to be safer, it would likely be better to place your transaction fees in the 25-30 sats per vbyte range rather than in the 20-25 sats per vbyte or any lower than 25-30 sats per vbyte... .so of course, as the backlog gets widdled away then we are able to put our economic transactions lower, but if want to be sure that they go through quickly we have to go a bit higher than the backlog and hope that any spike is not sustainable.. since it does seem to take a bit of time to build up the backlog in the various ranges, and I still am not even claiming to have much clues about the causes, even though as many of us here, I am not suspecting the backlog to be as organic as it might be being presented to be.. but hey I don't know enough.. so I am more than willing to wait to attempt to weigh what others think or what kinds of evidence is presented regarding plausible causes and theories of sustainability of keeping fees up like that and if shenanigans might be going on or just shenanigans that are costing the attackers a lot of money to keep up the ruse..
if you zoom right in on the jochen graph you might see the very rare one pixel change meaning maybe a little amount of cheap tx got into some lucky block. but the overall straight line shows there is no real coming of going of tx of that fee point because they get purged.. not eaten by block
the "min" is more of a purge line.
Maybe there are better charts to show your purge theory? I admit that I heard about some of these purging ideas but I have not really looked into it.. Also, I am not opposed to the idea that low ball transactions might sometimes get purged or some software might not allow fees to get set below certain amounts, but I just don't see how transactions would be being purged at anywhere within 30% or so of the backlog line (referring to the jochen graph).. and then also if the backlog line is continuing to go down, then there seems to be less and less reason to be engaging in any purging.. except maybe way below 5 sats per vbyte or something like that, and I had also heard that some wallets might not allow transactions to get set when they are at certain low levels.. .. but to me the backlog still seems to be going down... at least for now...and really in the last 30 days, even though the there has been a lot of transaction fee-related suffering in the last 3 months.