Interesting Block 679115

[c_atlas]

I was on mempool.space looking at the current sat/vbyte when I noticed that block 679115 had only 1 transaction. I've never seen this happen before, it was mined by AntPool, do any of you know why they would mine a block like this? What incentive was there to mine an empty block? They just missed out on +1BTC in fees...

Hash: 0000000000000000000679cde96f90985c8b23f2880b46af017596f67cb4cdfe
Here's a link to the block

[BayAreaCoins]

Miners profit from the fees staying high and the Bitcoin wallet estimating the traffic is heavy.

Empty blocks, with block rewards, are smart when it causes people to pay 10x to 25x the "normal" fee for confirmation within the next block.

It's the reverse of Pool Hopping basically, but against the network users rather than the miners this time.

[c_atlas]

Miners profit from the fees staying high and the Bitcoin wallet estimating the traffic is heavy.

Empty blocks, with block rewards, are smart when it causes people to pay 10x to 25x the "normal" fee for confirmation within the next block.

So they're increasing network congestion in an attempt to increase the fees? That seems a bit extreme, especially since higher fees in the next block could go to another pool. Does this happen often, or only when the network activity is higher than usual (say because the price is up considerably for the day)?

[BayAreaCoins]

So they're increasing network congestion in an attempt to increase the fees?

I don't know, but it certainly feels that way to me.  I believe they contribute to the congestion as well.

That seems a bit extreme, especially since higher fees in the next block could go to another pool.

Perhaps, but perhaps the pools are in on it together?  Seems like the most profitable for every one mining and there aren't very many that need to agree.

The way the pools are divided up could be just to appear to make Bitcoin look decentralized as well.

Bitcoin is MUCH better off with a bunch of solo miners.

Does this happen often, or only when the network activity is higher than usual (say because the price is up considerably for the day)?

It happens really often.  

MoparMining talks about it a lot.

https://www.theblockcrypto.com/linked/1964/close-to-19-of-mined-bitcoin-blocks-were-empty

[odolvlobo]

Miners profit from the fees staying high and the Bitcoin wallet estimating the traffic is heavy.
Empty blocks, with block rewards, are smart when it causes people to pay 10x to 25x the "normal" fee for confirmation within the next block.
It's the reverse of Pool Hopping basically, but against the network users rather than the miners this time.

No.

Miners profit from the fees staying high and the Bitcoin wallet estimating the traffic is heavy.

Empty blocks, with block rewards, are smart when it causes people to pay 10x to 25x the "normal" fee for confirmation within the next block.

So they're increasing network congestion in an attempt to increase the fees? That seems a bit extreme, especially since higher fees in the next block could go to another pool. Does this happen often, or only when the network activity is higher than usual (say because the price is up considerably for the day)?

No.

So they're increasing network congestion in an attempt to increase the fees?

I don't know, but it certainly feels that way to me.  I believe they contribute to the congestion as well.
Perhaps, but perhaps the pools are in on it together?  Seems like the most profitable for every one mining and there aren't very many that need to agree.

No. Unlikely.

Empty blocks occur because miners start mining the next block immediately after receiving a new block, and they don't wait to process the new block and gather the transactions for the next block. However, once the miner is ready, they will switch to mining a full block -- they don't want to miss out on the transaction fees.

A miner that gives up transaction fees effectively gives that revenue to other miners, so there is unlikely to be collusion because miners that are not colluding would take all the fees from those that are. Also, it doesn't happen very often.

[c_atlas]

So they're increasing network congestion in an attempt to increase the fees?

I don't know, but it certainly feels that way to me.  I believe they contribute to the congestion as well.
Perhaps, but perhaps the pools are in on it together?  Seems like the most profitable for every one mining and there aren't very many that need to agree.

No. Unlikely.

Empty blocks occur because miners start mining the next block immediately after receiving a new block, and they don't wait to process the new block and gather the transactions for the next block. However, once the miner is ready, they will switch to mining a full block -- they don't want to miss out on the transaction fees.

A miner that gives up transaction fees effectively gives that revenue to other miners, so there is unlikely to be collusion because miners that are not colluding would take all the fees from those that are. Also, it doesn't happen very often.

Check the block out.

AntPool didn't mine the next 2 blocks, but it seems like they managed to raise the fees substantially since block 679115.

[odolvlobo]

Check the block out.
AntPool didn't mine the next 2 blocks, but it seems like they managed to raise the fees substantially since block 679115.

Can you show that AntPool gained more than 1.03 BTC by not including transactions? Because, that's how much the miner of the next block got from including transactions.

[c_atlas]

Check the block out.
AntPool didn't mine the next 2 blocks, but it seems like they managed to raise the fees substantially since block 679115.

Can you show that AntPool gained more than 1.03 BTC by not including transactions? Because, that's how much the miner of the next block got from including transactions.

Check the blocks before 679115, I went back until I was rate limited but I didn't see a block with +1BTC in fees. Several blocks following the empty one earned +1BTC in fee's although AntPool wasn't one of the lucky ones. Since the fees have come back down and most of the higher fee blocks went to other miners, I'm back to being skeptical that AntPool did this to raise fees. They would've made more if they just filled the block, although maybe their model's said it was the right decision...

[redwine49]

Check the block out.
AntPool didn't mine the next 2 blocks, but it seems like they managed to raise the fees substantially since block 679115.

Can you show that AntPool gained more than 1.03 BTC by not including transactions? Because, that's how much the miner of the next block got from including transactions.

Check the blocks before 679115, I went back until I was rate limited but I didn't see a block with +1BTC in fees. Several blocks following the empty one earned +1BTC in fee's although AntPool wasn't one of the lucky ones. Since the fees have come back down and most of the higher fee blocks went to other miners, I'm back to being skeptical that AntPool did this to raise fees. They would've made more if they just filled the block, although maybe their model's said it was the right decision...

I don't like the fact that miners can hog the transaction fees and and purposely mine empty blocks to keep the network busy.
I don't think that's good habit for bitcoin ecosystem.

with new people getting into bitcoin right now. experience for the first time sending money with bitcoin.
Then they think bitcoin transaction is very slow and very expensive.

Btw i feel the same

[jackg]

I'm thinking this could be a "distributed luck" strategy.

Make x3/6 miners mine the next block as it's selected algorithmically (based off fees).
Make x2/6 miners mine a block we've put together randomly.
Make x1/6 miners mine an empty block to see if we can push that out (it'd realistically be a lot lower than 1/6 because difficulty is a lot lower for smaller blocks afaik since there's only one hash to compute - I *think*).

I think the more transactions there are in the mempool the more the fee has less to do with what gets confirmed by these pools (unless the fee is very high or very low).

[int03h]

The difficulty of mining each block is random so it's not difficult to understand that block 69715 has only 1 confirmed transaction.
Please observe that the preceding block is 69714. In this block, there are 2437 confirmed transactions. The time to exploit the block immediately following the above block is 69715 almost at the same time. So it has only one confirmed transaction.
Check out the links to see the answer.
https://blockchair.com/bitcoin/block/679114
https://blockchair.com/bitcoin/block/679115

[odolvlobo]

...it'd realistically be a lot lower than 1/6 because difficulty is a lot lower for smaller blocks afaik since there's only one hash to compute - I *think*...

Your thinking is wrong. The difficulty is the same regardless of the size of the block.

[jackg]

...it'd realistically be a lot lower than 1/6 because difficulty is a lot lower for smaller blocks afaik since there's only one hash to compute - I *think*...

Your thinking is wrong. The difficulty is the same regardless of the size of the block.

https://en.bitcoinwiki.org/wiki/Merkle_tree

I'm assuming the tree is balanced compared to being one sided. If it's not balanced then it will take fewer hash functions once the initial ones are done. But non balanced trees seem quite inefficient.

[odolvlobo]

...it'd realistically be a lot lower than 1/6 because difficulty is a lot lower for smaller blocks afaik since there's only one hash to compute - I *think*...

Your thinking is wrong. The difficulty is the same regardless of the size of the block.

https://en.bitcoinwiki.org/wiki/Merkle_tree
I'm assuming the tree is balanced compared to being one sided. If it's not balanced then it will take fewer hash functions once the initial ones are done. But non balanced trees seem quite inefficient.

I guess I misunderstood your point. If you are counting the number of times that any SHA-256 hash is computed when building a block, then that depends on several factors (including the number of hashes needed to compute the merkle root). However, those are still only a fraction of the number of hash calculations needed to find a hash that is less than the target hash, and the target hash (which determines the difficulty value) is the same regardless of the size of the block.

[jackg]

However, those are still only a fraction of the number of hash calculations needed to find a hash that is less than the target hash, and the target hash (which determines the difficulty value) is the same regardless of the size of the block.

The functions used in mining blocks are probably best described similar to radiation.

You can't sit with a carbon 14 atom on your desk and say "I know for sure this one will be carbon 12 within 1400 years" but you're able to sit with a mol of carbon 14 and say for sure that the radioactive emmission will have halved within the next 1400 years.

This is similar with blocks and the programmers at the top of producing miners and their algorithms know this. They're not using all their miners to solve the same exact block and just changing nonces and version numbers. They're attacking whatever they can to be the ones to successfully mine the next block because there's money and the reliability of the network in it.

Sure an empty block isn't very productive this time around but if it'd take 2 hours to find a non empty block you may as well find the empty one and move on.

(in extreme terms, a long wait for a block could lead to a higher chance of a 51% attack or confirmations being undone if miners' algorithms decide to traverse back over the chain - as can be done with orphaning but I think 2 block orphans are rare these days in most chains).

Edit: can't decide if this is fully on topic so op feel free to report it for a split from your topic.

[BayAreaCoins]

If you look at that image of the Bitcoin network... it's a razor-thin line keeping the miners earning 20x (or more) from people who want confirmation in 10 minutes vs 24 hours... (each block is roughly 1mb, there are 144 blocks roughly a day.)

Imagine Bitcoin is like a bus with 20 seats and there is a line of 200 people trying to get on.  Basically, each ticket "could" cost $1, but the person who owns the bus and collects the fees keeps putting 15 people on the bus for $100 each and five normal people who just want to get home pay $100 as well.  Now the bus is full.  The bus owner just netted $500 from five people rather than $20 from 20 people.