Empty blocks

[nepaluz]

Johnybravo, I do not agree with you on the specific issue of the process of SPV mining being unstable in itself, I rather think it is the people who code it that flagrantly implement it - and that is a nod with particular reference to the mentioned forking.

On top of that, forking is not exclusively caused by SPV mining but more importantly there's no harm to "rest of us" when a fork of this nature happens. Infact, forks do happen and the protocol knows about it, thus it can not, by definition, be detrimental.

[ShrykeZ]

Cool. Is it possible for you to open-source this program? I have a few ideas for it.

Can you tell what pool(s) are currently mining 0 transaction blocks consistently?
Thanks,
Sam

Are you saying that some pools can opt out of mining empty blocks? I thought you don't know what's coming until you finish mining the block :-/

I guess you learn something new everday day!

The contents of the 'block to be' are decided by the pool before it sends out the work to the miners.

It's rare if ever that there are no transaction available on the network.

Many pools send out work to their miners that includes no new transactions other that the necessary coinbase transaction required to build a block.
That's what all the "Empty Block" stats are about.

My pool https://kano.is/ doesn't send out work containing empty blocks
(unless there are no transaction available in the pool's bitcoind - which is probably never)

This is why people should switch from the larger usually Chinese based pools over to pools like Kano's or others that operate similarly.

[kano]

Johnybravo, I do not agree with you on the specific issue of the process of SPV mining being unstable in itself, I rather think it is the people who code it that flagrantly implement it - and that is a nod with particular reference to the mentioned forking.

On top of that, forking is not exclusively caused by SPV mining but more importantly there's no harm to "rest of us" when a fork of this nature happens. Infact, forks do happen and the protocol knows about it, thus it can not, by definition, be detrimental.

Well it only takes one miner to produce a block with 1 bad transaction in it, and that can push ~65% of the mining power off the bitcoin blockchain.
Not detrimental?

The law covers the issue of dealing with someone killing someone else.
However, killing someone else is usually considered detrimental ...... by definition.

[nepaluz]

Ofcourse it does, and mangoes are prunes though just like not all blockchain forking is due to SPV mining.

[kano]

Ofcourse it does, and mangoes are prunes though not all blockchain forking is due to SPV mining.

At least reread what you wrote here:
https://bitcointalk.org/index.php?topic=1085800.msg13205770#msg13205770
Yes you can't delete it coz I quoted it also.

[nepaluz]

Is it possible to delete messages? Well I'll spell it out for you then, not all prunes are mangoes.

[jonnybravo0311]

I don't think it was ever claimed that forking was exclusively caused by SPV mining.  You made that statement.  However, what is true is that SPV mining can and has caused forks.

Just for fun, I re-ran things again to see where we are.  As of block 387696:

AntPool 7.61% of their blocks are empty
BW.com 4.14%
f2pool 3.55%
KnC 3.23%
Eligius 2.3%

[nepaluz]

You indeed never claimed such, and your numbers should tell a good story ... the one you refuse to see and tell, so I'll ask for some pointed stats.
1. What is the percentage of forks / empty blocks?
2. What is the percentage of forks / all blocks?
3. What is the percentage of forks / empty blocks in a timespan of the highest concentration of forks (in which timespan at least 2000 blocks have been mined)?
4. What is the percentage of forks / all blocks in a timespan of the highest concentration of forks (in which timespan at least 2000 blocks have been mined)?

The answer to all of those is bound to be in fractions of a percentage; extrapolations will without doubt yield similar results, meaning the likelihood of forks happening due to empty block mining is statistically negligible, and that will be true for a very long time.

So if you base your argument of empty block mining being detrimental due to the evil that is forking, which said forking's occurence is statistically negligible, you may have an agenda but you definitely have no point.

[jonnybravo0311]

I welcome you to perform those calculations.  I'm simply providing the data.  I've also never made the conclusion you came to:

base your argument of empty block mining being detrimental due to the evil that is forking, which said forking's occurence is statistically negligible, you may have an agenda but you definitely have no point.

There is nothing in this thread stating empty block mining is detrimental due to forking.  In fact, I've never even tied the two together.

I'm not sure what agenda you think I may have, other than to report the numbers.

[philipma1957]

We had the largest block count day in years on the 10th of dec

 215 blocks.

we had more then 20

 1 and 2 transaction blocks.

The  purpose of BTC is to move payments via transactions.


a 10+ percent of no transaction blocks is not good.

It clogs up the payment system.   If the 20 tiny transaction blocks all had 1000 transactions we would have moved  20,000 transactions.

We would not have this happening

blockchain info skipped auto payment of fees pulled excess coin out of account sent it back causing an un confirmed .0101 amount

from a fully confirmed original balance of .202

https://blockchain.info/tx/92a7cba0eac341534e79362b2d9c2dba42ee3479d69bd347c036f5c778186c6d

thus 1 minute later when blockchain info made this payment with a fee they did won't confirm this .151 transfer

so .202 fully confirmed btc is now locked up as unconfirmed due to 2 issues. first blockchain info client forget to pay a fee and needlessly casued an unconfirmed .0101

second dozens of empty blocks passed by these two transactions which if you do both of them you do get .0001 btc

So the entire point of mining which is moving coins via transactions is lost here.  Why some pools just want block rewards.

Well we will see how long the .202 stays tied up and pools continue to mine empty blocks to pass by the fees to get rewards.

More then 30 hours now.

https://blockchain.info/tx/3989f4a2d7cdbcd146e5b9df949c514fd5920e66e9486d4c18bf1681ab7ba3b7

[Rizky Aditya]

I don't really understand the issue here. Does the miner earn less if he/she finds/mines an empty block? I'm pretty sure that he/she would earn the same amount.

[kano]

I don't really understand the issue here. Does the miner earn less if he/she finds/mines an empty block? I'm pretty sure that he/she would earn the same amount.

No.

[leowonderful]

lol antpool mines yet another empty block https://blockexplorer.com/block/000000000000000001a7ce5d59bd2b32422884114adeadbdb5427ffb3836e47e

[-ck]

lol antpool mines yet another empty block https://blockexplorer.com/block/000000000000000001a7ce5d59bd2b32422884114adeadbdb5427ffb3836e47e

No need to point out each and every one of them. It's a regular occurrence.

[jonnybravo0311]

So... with the backlog of transactions currently plaguing the network, one would think that the pools would be constantly pushing full blocks onto the chain.  Yeah.  Here's how it is really stacking up since last Monday (block 399383 to block 400871):

AntPool - 44/341 blocks mined are empty (12.9%)
bw.com - 8 of 103 blocks mined are empty (7.77%)
f2pool - 4 of 382 blocks mined are empty (1.05%)
KnC - 3 of 58 blocks mined are empty (5.17%)
Eligius - 2 of 9 blocks mined are empty (22.22%)
Slush - 1 of 69 blocks mined are empty (1.45%)

If we take an average of 1500 transactions per block, which is low by the way, then had AntPool had even half as few empty blocks there would be no transaction backlog at all.  Everybody's transactions would be confirmed.  There wouldn't be pages and pages of questions in the forums asking why a transaction hasn't been confirmed after days of waiting.

For the love of all that's holy - STOP MINING ON POOLS THAT PRODUCE EMPTY BLOCKS!!!

[os2sam]

Thanks for the update.

[valkir]

They want bigger block so they do empty block.  So idiot.

[TracerX]

So... with the backlog of transactions currently plaguing the network, one would think that the pools would be constantly pushing full blocks onto the chain.  Yeah.  Here's how it is really stacking up since last Monday (block 399383 to block 400871):

AntPool - 44/341 blocks mined are empty (12.9%)
bw.com - 8 of 103 blocks mined are empty (7.77%)
f2pool - 4 of 382 blocks mined are empty (1.05%)
KnC - 3 of 58 blocks mined are empty (5.17%)
Eligius - 2 of 9 blocks mined are empty (22.22%)
Slush - 1 of 69 blocks mined are empty (1.45%)

If we take an average of 1500 transactions per block, which is low by the way, then had AntPool had even half as few empty blocks there would be no transaction backlog at all.  Everybody's transactions would be confirmed.  There wouldn't be pages and pages of questions in the forums asking why a transaction hasn't been confirmed after days of waiting.

For the love of all that's holy - STOP MINING ON POOLS THAT PRODUCE EMPTY BLOCKS!!!

F2Pool has really cleaned up it's act, at least with BTC mining.  They had a higher than 1% weekly average for a long time--I suppose I'll start a thirty/60/90 day stat. 

[jonnybravo0311]

Yes, they've certainly gotten better about submitting empty blocks.  Unfortunately, they still SPV mine and they are still submitting empty blocks, albeit far less frequently.

[watashi-kokoto]

they are still submitting empty blocks, albeit far less frequently.

empty blocks result from bursts