In the current Bitcoin mining landscape with proxy pools like AntPool & friends, 6 mining pools produce and mine more than 95% of the block templates.
Bitcoin mining is highly centralized today.
b10c.me/blog/015-bit...
new @OrangeSurfBTC tool https://orange.surf/grid/
This (x.com/boerst/statu...) post by boerst inspired me to dig into the invalid "AntPool & friends" mining jobs a bit deeper. Findings:
- happens during forks
- likely bug in AntPool coinbase building
- probably not selfish mining
- proxy pools affected too
b10c.me/observations...
👀@mempool
cc @mononautical @OrangeSurfBTC
.@mempool the accelerated transaction mempool.space/tx/1e3674b914e20c8ac2c62... shows 8.33 s/vB in the block-audit but only 3.40 s/vB on the transaction page.
Based on the position in the block 8.33 s/vB should be correct. Based on the accelerated...
testnet4 (aka forknet4) right now
https://fork.observer/?network=4
@mempool's block audit also expected the transactions and considers it as excluded due to "marginal fee rate" too
here's an RSS feed for anyone interested miningpool.observer/template-and-block/sanct...
miningpool-observer alerting about a "Sanctioned Transaction missing from Block" on @f2pool_offical's block 875575.
I think it was reasonable for F2Pool not to include it as they knew about transactions paying a higher fee - but I'll keep an eye on it
https://miningpool.observer/template-and-bl
fyi @mempool this address shows a negative balance for me
node210 fra (v3.1.0-dev) [01df22ef][17255d15]
mempool.space/address/3JP3zF7LoeoAotqk...
cc @wantclue @skot9000
Someone working on something like this?
testnet4 looks more like forknet4 right now
hm @FoundryServices ?
I had another look at similarities between mining pool block template by comparing the Merkle branches in the mining jobs. This confirms suspicion about proxy pools and raises the question if some pools are directly peering with each other.
https://b10c.me/observations/12-template-similarity/
really like this image from the chapter on open-source development of @kallerosenbaum's Bitcoin development philosophy
https://bitcoindevphilosophy.com/#opensource
During the fork at height 853051 between Foundry and Binance Pool, Foundry mined on their own block and won.
The first version 3 Bitcoin transaction😉
mempool.space/tx/110da331fd5336038316c...
No drop in inbound connections to my Bitcoin P2P monitoring nodes since the start of the #cloudstrike incident.
(matches my assumption that most nodes probably don't run on Windows servers with cloudstrike installed)
also as blog post with a few recent examples
b10c.me/blog/014-mining-pool-gam...
mining pool game theory during forks
Todays fork between Foundry (winning) and AntPool again shows that AntPool (& friends/proxies) isn't mining on their own block during a fork.
AntPool first send an empty job mining on their own block but then switched to Foundry's block and found a block on it.
My data might not show the full picture here as it shows ViaBTC only mining on AntPool's block. This is possibly due to ViaBTC running multiple stratum servers (e.g. "VIP" servers for high-hashrate miners) distributing different jobs.
Another fork between AntPool (loosing) and Foundry (winning). While my data shows that all pools besides Foundry mined on AntPool's block, ViaBTC found a block on Foundry's block.
Also, we see forks now almost every other day now..
