Better sats/vb conversion #86
Conversation
| 253 => 1.0, | ||
| _ => fee_sats_per_1000_wu as f32 / 250.0, | ||
| } as f32; | ||
| let sat_per_vb = std::cmp::min(1.0, fee_sats_per_1000_wu as f32 / 250.0); |
There was a problem hiding this comment.
I'm still not entirely sure this is right either. I'm pretty sure 253 "fee_sats_per_1000_wu" is supposed to be 1.0 sat_per_vb. If you always just divide by 250.0 you get 1.012. Which I guess probably gets rounded to 1 or I guess even if not it means you'd pay an extra sat for every 100 vbytes? Not really a big deal.
Matt ended up opening this issue in bdk bitcoindevkit/bdk#608
I'm still not entirely sure the best way to do this for now but maybe your simplification is fine?
There was a problem hiding this comment.
This is what we do in bitcoin-s so I copied it here:
There was a problem hiding this comment.
@johncantrell97 @benthecarman given that the minimum fee in sats_per_kwu is 253. Will std::cmp::min(1.0, fee_sats_per_1000_wu as f32 / 250.0) not always produce a fee rate of 1.0. Is there something am missing here?
Use min of 1 because of policy not relaying transactions under this limit