📅 Original date posted:2015-05-10
📝 Original message:On Sun, May 10, 2015 at 9:21 PM, Gavin Andresen <gavinandresen at gmail.com> wrote:
> a while I think any algorithm that ties difficulty to block size is just a
> complicated way of dictating minimum fees.
Thats not the long term effect or the motivation-- what you're seeing
is that the subsidy gets in the way here. Consider how the procedure
behaves with subsidy being negligible compared to fees. What it
accomplishes in that case is that it incentivizes increasing the size
until the marginal "value" to miners of the transaction-data being
left out is not enormously smaller than the "value" of the data in the
block on average. Value in quotes because it's blind to the "fees"
the transaction claims.
With a large subsidy, the marginal value of the first byte in the
block is HUGE; and so that pushes up the average-- and creates the
"base fee effect" that you're looking at. It's not that anyone is
picking a fee there, it's that someone picked the subsidy there. :)
As the subsidy goes down the only thing fees are relative to is fees.
An earlier version of the proposal took subsidy out of the picture
completely by increasing it linearly with the increased difficulty;
but that creates additional complexity both to implement and to
explain to people (e.g. that the setup doesn't change the supply of
coins); ... I suppose without it that starting disadvantage parameter
(the offset that reduces the size if you're indifferent) needs to be
much smaller, unfortunately.
Gregory Maxwell [ARCHIVE] /
npub1f2…crwet
2023-06-07 15:34:01
in reply to nevent1q…t5et
Author Public Key
npub1f2nvlx49er5c7sqa43src6ssyp6snd4qwvtkwm5avc2l84cs84esecrwetPublished at
2023-06-07 15:34:01Event JSON
{
"id": "f9216e585959ed427a0298c9b8c11553827bd4fdbc4f0a9b1078a2d66ad99304",
"pubkey": "4aa6cf9aa5c8e98f401dac603c6a10207509b6a07317676e9d6615f3d7103d73",
"created_at": 1686152041,
"kind": 1,
"tags": [
[
"e",
"aec396df7693e37c124fbd2891fe6c4a3b28f46e7fbc3f304a7b1d78d2f5ffbe",
"",
"root"
],
[
"e",
"288b4066a767d19a108bf67e9da1e412e0fe862a3e7034e9fe8f271105f8e13f",
"",
"reply"
],
[
"p",
"857f2f78dc1639e711f5ea703a9fc978e22ebd279abdea1861b7daa833512ee4"
]
],
"content": "📅 Original date posted:2015-05-10\n📝 Original message:On Sun, May 10, 2015 at 9:21 PM, Gavin Andresen \u003cgavinandresen at gmail.com\u003e wrote:\n\u003e a while I think any algorithm that ties difficulty to block size is just a\n\u003e complicated way of dictating minimum fees.\n\nThats not the long term effect or the motivation-- what you're seeing\nis that the subsidy gets in the way here. Consider how the procedure\nbehaves with subsidy being negligible compared to fees. What it\naccomplishes in that case is that it incentivizes increasing the size\nuntil the marginal \"value\" to miners of the transaction-data being\nleft out is not enormously smaller than the \"value\" of the data in the\nblock on average. Value in quotes because it's blind to the \"fees\"\nthe transaction claims.\n\nWith a large subsidy, the marginal value of the first byte in the\nblock is HUGE; and so that pushes up the average-- and creates the\n\"base fee effect\" that you're looking at. It's not that anyone is\npicking a fee there, it's that someone picked the subsidy there. :)\nAs the subsidy goes down the only thing fees are relative to is fees.\n\nAn earlier version of the proposal took subsidy out of the picture\ncompletely by increasing it linearly with the increased difficulty;\nbut that creates additional complexity both to implement and to\nexplain to people (e.g. that the setup doesn't change the supply of\ncoins); ... I suppose without it that starting disadvantage parameter\n(the offset that reduces the size if you're indifferent) needs to be\nmuch smaller, unfortunately.",
"sig": "593ac318cde30d9810ffe5a48508d872254815c3ba36cf558a51c1808bec36dacf4338f81592ddb3643de5a427a237634ae1c4d4a708bafe334690d4de8b9b3d"
}