<articmine> @boog900: I am planning to double the fees

<articmine> It is needed to support tx weights up to 20000 bytes.

<articmine> For smooth scaling up to 8 in 16 out FCMP++ transactions

<articmine> This is with a penalty free zone of 1000000 bytes

<articmine> By the way tx weights drive the penalty which in turn drives fees. So tx weights are very closely related to fees

<articmine> This is completely different from Bitcoin type crypto currencies.

<ofrnxmr> Double per byte, or per input etc? > <@articmine> I am planning to double the fees

<ofrnxmr> If you look at stressnet, the fees are already quite high

<ofrnxmr> about 15c if pretending that testnet was 300$ per xmr on normal (lvl 2) fee

<ofrnxmr> Fee (per_kB): 0.000519840000 (0.000084925999)

<ofrnxmr> Tx size: 6.1211 kB

<namenet:matrix.org> @ofrnxmr: Is it possible to lower this fee? I noticed my testnet fee was also approximately 10c

<articmine> When all the weights are sorted out the fee for a 2 in 2 out tx will over 2x what it is now.

<articmine> There are two factors:

<articmine> We are increasing the scaling rate from 1% to 2% 1000000 bytes for the penalty free zone with a reference tx size of 20000 bytes.[... more lines follow, see mrelay.p2pool.observer/e/voiUw8EKZ0pVQnV6 ]

<articmine> I am opposed to a lower penalty free zone than 1000000 bytes for FCMP++

<articmine> This is to keep the fees from raising higer

<articmine> Higher

<articmine> What are the scaling parameters for testnet?

<articmine> Penalty free zone, tx weights etc

<articmine> Reference transaction size?

<ofrnxmr> @articmine: Untouched

<articmine> From now with a tx weights of say 9000 bytes?

<ofrnxmr> @ofrnxmr: this is what a lvl 2 (normal) 1 in 2 out tx looks like

<articmine> Then 0.15 USD is about what I would expect

<articmine> If not more

<ofrnxmr> what would this look like under proposed changes?

<ofrnxmr> Approximately

<articmine> For the low fee ~0.025 USD based upon 300 USD/XMR

<articmine> For the normal fee 4x that

<articmine> So 0.10.USD

<articmine> This is with a penalty free zone of 1000000 bytes

<articmine> I suspect that some people will think it is too low and other too high

<articmine> This y for 1 in 2 out

<gingeropolous> it seems fine to me. I think the fee calculations should satisfy these requirements: 1. reasonable for the average user making average transactions, 2. designed such that industrial users are incentivized to create optimized transactions (it shouldn't cost less to make 64 transactions that are chain-heavy (storage and compute) that could be handled by 1 transaction that is chain-lite.

<gingeropolous> well, i guess heavy could cover it. because there's chain-heavy (storage) and network-heavy (compute, bandwidth).

<ofrnxmr> 1. Is a matter of $/xmr 2. Agree

<ofrnxmr> At $300, 10c for a "normal" tx fee might sound a bit high, but at $3000, its prohibitively high.

<ofrnxmr> Imo, i have np with raising the low fee from 0.00004 to 0.00008, but id also reduce the normal multiplier from 4x to 2x. So instead of 1 4 16 1000, id do smthn like 1 2 20 100 1000

<sgp_> I have a radically simple fee proposal that I can organize if there's interest. The highlights are:

<sgp_> Eliminate min fee. Create a small penalty free zone (e.g. 400 kB), and allow miners to grow blocks from that zone to X size (e.g. 2x the penalty zone size, in this example 800 kB) with linear penalty from the tail emission to X until tail emission is fully depleted at X. Allow growing X by y% per block (e.g. 10% per year). Req [... too long, see mrelay.p2pool.observer/e/4aeE4sEKZWRyOGpC ]

<sgp_> This removes the need to "guess" what a fair USD/XMR rate is. It allows for market competition for block space. And it allows for continuous growth over time to account for technological advancements

<sgp_> mrelay.p2pool.observer/m/monero.social/HcsfjofWLvrwQRLheLBindZQ.png (image.png)

<sgp_> works for XMR @$1, and XMR @$100,000

<elongated:matrix.org> @sgp_: We need to be realistic

<sgp_> @sgp_: In case this wasn't clear, I mean it works for all values