[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool

[kano]

This subject has come up before and there were some p2pool members who made it clear that they don't care about confirming transactions ...
I wonder how many of them ever used any BTC ... ever ...

Blasphemy!

Seriously though, while we have (as a pool) mined a few 0 tx blocks in my time, the vast majority of nodes include transactions, which in my opinion is very important to do.

I'd also further argue that the vast majority of P2Pool miners do it for a love of the decentralized nature of Bitcoin and the fact that zero-trust is required.

No offense to anyone on another pool, but when you consider the mindset of your average miner, P2Pool miners care more about the network then most.

Just look at BTC Guild or GHash when they were getting close to 50%, average miners flocked to them in droves to reap the rewards without concern for the potential impact while P2Pool chugged along completely decentralized.

Yes, indeed.
I guess my comment could have come off as a negative to p2pool - but it wasn't meant to be at all.
That 'some' was only a few ... not lots.

Edit: some of the big pools start blocks changes with empty blocks also.
This is an issue I've reiterated about anyone using the eloipool software ... like Eligius ...
Antpool also does this.
They produce empty blocks on occasion and do it to reduce their rejects at the pool coz their software is slow.
They clearly fall under the heading of pools doing things inherently bad for bitcoin.
We don't ever do this with ckpool - I always use the transactions in the block template if they are there from bitcoind.

[manfred87]

so what if there is one really ugly block in the sharechain, which leads to a "rejection" of every found block.. would we see this?

What would an "ugly block" be? And how could any type of block lead to the "'rejection' of every found block"? If it's a block, it's ipso facto valid.

i am in the altcoin scene.. there are many possbilies where a block has the right difficulty, but is not acepted in the network.. the coin daemon simply rejects it although you have the right difficulty.. maybe a block which does not include masternoes ( darkcoin aka dash ) or has other "problems". so you see "block rejected" in your log.. maybe you can ignore that for 500 myriadcoin.. but it would be really hard if 25 BTC were lost this way.. every aspect of a block has to be valid.. all transactions, etc...

[Geremia]

I wish I had more data to work with, but here's what the luck distribution looks like (as of today):


(data courtesy Minefast.CoinCadence.com P2Pool stats)

[kano]

Does that include (as I said above)
The valid stale blocks?
Most of the stale share-chain shares?

The pool by design has a much higher stale rate ... MUCH higher.
That doesn't mean the work wasn't done to find the blocks.
The stale shares are usually ignored on a normal pool since they are usually under 0.5% - even as low as 0.2%
Here on p2pool they are of the order of 10%
So effectively you would be ignoring 20 to 50 times the stale shares as a normal pool - so that would falsely skew luck higher for p2pool.
Then to make things even worse, the blocks found for those stale ignored shares are accounted to p2pool as extra blocks ...

So yeah you need to include both of those to produce anything that's not a false representation of p2pool luck.

[Geremia]

`cgminer` says my difficulty with P2Pool is 715 and 746 for my two identical miners, respectively. Mining with Eligius, my difficulty seems to be in increments of powers of two (either 512 or 1024 for my miners). It thus seems P2Pool allows a much finer resolution on the local miners' difficulties. Why is this?

[-ck]

`cgminer` says my difficulty with P2Pool is 715 and 746 for my two identical miners, respectively. Mining with Eligius, my difficulty seems to be in increments of powers of two (either 512 or 1024 for my miners). It thus seems P2Pool allows a much finer resolution on the local miners' difficulties. Why is this?

There's nothing "fine" about it. It's a coding decision to set the share submission rate closer to 60 than 20 (like regular pools use), and p2pool's diff is meaningless apart from providing you with an estimated hashrate.

See: https://bitcointalk.org/index.php?topic=274023.0

[Geremia]

Does P2Pool support IPv6?

[MissouriMiner]

Great.  A Block!   I also sent some BTC using sendmany.  Let's hope for some good luck! 

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

[idonothave]

Great.  A Block!   I also sent some BTC using sendmany.  Let's hope for some good luck! 

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

well done

[squidicuz]

Finally! \o/

[Geremia]

Great.  A Block!   I also sent some BTC using sendmany.

Thank you for the 0.39755 mBTC!

Let's hope for some good luck!  

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

Although the block only has 2 transactions… I suppose that's better than the previous block, which only had 1 (the coinbase one)…

I just joined P2Pool on 4/5/15 and am pretty pleased.

How do we determine the address corresponding to the person in the P2Pool who actually mined this block so we can thank him/her directly?

[jonnybravo0311]

Great.  A Block!   I also sent some BTC using sendmany.

Thank you for the 0.39755 mBTC!

Let's hope for some good luck!  

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

Although the block only has 2 transactions… I suppose that's better than the previous block, which only had 1 (the coinbase one)…

I just joined P2Pool on 4/5/15 and am pretty pleased.

How do we determine the address corresponding to the person in the P2Pool who actually mined this block so we can thank him/her directly?

In cases like this last block, you cannot because the share that solved the block was either DOA or orphaned - so it didn't make it onto the p2pool share chain, and hence there's no record of the share itself.  However, if the share that solves the block does make it onto the share chain, then you can see the payout address of the share (I'm just taking a random share here and not one that solved a block):

Code:

Share data

Timestamp: Sat Apr 18 2015 14:58:23 GMT-0400 (EDT) (1429383503)

Difficulty: 2852459.6919484762

Minimum difficulty: 2852459.6919484762

Payout address: 1Mag5XQcm81yP3rz8qJvXbU1KSWypKmPur

[kano]

Great.  A Block!   I also sent some BTC using sendmany.

Thank you for the 0.39755 mBTC!

Let's hope for some good luck!  

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

Although the block only has 2 transactions… I suppose that's better than the previous block, which only had 1 (the coinbase one)…

I just joined P2Pool on 4/5/15 and am pretty pleased.

How do we determine the address corresponding to the person in the P2Pool who actually mined this block so we can thank him/her directly?

No, it had lots of transactions.
If it only had 1 transaction it would be pretty much as bad for bitcoin as the last one.

As for thanking them ... well it is completely random finding a block ... and they get thanked by getting a block finder reward by getting a small % of everyone's expected payout from everyone mining on p2pool

[Geremia]

In cases like this last block, you cannot because the share that solved the block was either DOA or orphaned - so it didn't make it onto the p2pool share chain, and hence there's no record of the share itself.

That's interesting. I suppose wouldn't be necessary to make it to the P2Pool chain; that'd just slow down the announcement of the found block?

[jonnybravo0311]

Great.  A Block!   I also sent some BTC using sendmany.

Thank you for the 0.39755 mBTC!

Let's hope for some good luck!  

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

Although the block only has 2 transactions… I suppose that's better than the previous block, which only had 1 (the coinbase one)…

I just joined P2Pool on 4/5/15 and am pretty pleased.

How do we determine the address corresponding to the person in the P2Pool who actually mined this block so we can thank him/her directly?

No, it had lots of transactions.
If it only had 1 transaction it would be pretty much as bad for bitcoin as the last one.

As for thanking them ... well it is completely random finding a block ... and they get thanked by getting a block finder reward by getting a small % of everyone's expected payout from everyone mining on p2pool

You looking at something different than I am?  Last block found by p2pool was 352654, and that block had only 2 transactions, of which the generation transaction was one... so it's almost just as bad as the last one

[jonnybravo0311]

In cases like this last block, you cannot because the share that solved the block was either DOA or orphaned - so it didn't make it onto the p2pool share chain, and hence there's no record of the share itself.

That's interesting. I suppose wouldn't be necessary to make it to the P2Pool chain; that'd just slow down the announcement of the found block?

No.  P2Pool tracks shares, which is how it determines what each miner will be paid when one of those shares is also the solution of the block.  In the case of orphaned shares, you add a share to the chain, but a longer chain has already been accepted by the network.  Therefore, your share is not added to the share chain and is orphaned.  However, because your share also solved the block of Bitcoin, the new block is created and added to the blockchain.  Miners are paid (because the generation transaction information is part of the block), but the share itself is not recognized by p2pool.  That's why virtually every single p2pool UI out there doesn't show this block - they all base block finds off of accepted p2pool share chain shares.  The exception is windpath's node - he scrapes the blockchain data to get the p2pool blocks regardless of whether the share that solved it is on the p2pool share chain.

[Geremia]

The exception is windpath's node - he scrapes the blockchain data to get the p2pool blocks regardless of whether the share that solved it is on the p2pool share chain.

Where's windpath's node?

[jonnybravo0311]

The exception is windpath's node - he scrapes the blockchain data to get the p2pool blocks regardless of whether the share that solved it is on the p2pool share chain.

Where's windpath's node?

http://minefast.coincadence.com

[kano]

Great.  A Block!   I also sent some BTC using sendmany.

Thank you for the 0.39755 mBTC!

Let's hope for some good luck!  

https://blockchain.info/tx/cf1207fdf10be1d86db56f89b8155250f0b1269d8ec321fd1203716739816f67

Although the block only has 2 transactions… I suppose that's better than the previous block, which only had 1 (the coinbase one)…

I just joined P2Pool on 4/5/15 and am pretty pleased.

How do we determine the address corresponding to the person in the P2Pool who actually mined this block so we can thank him/her directly?

No, it had lots of transactions.
If it only had 1 transaction it would be pretty much as bad for bitcoin as the last one.

As for thanking them ... well it is completely random finding a block ... and they get thanked by getting a block finder reward by getting a small % of everyone's expected payout from everyone mining on p2pool

You looking at something different than I am?  Last block found by p2pool was 352654, and that block had only 2 transactions, of which the generation transaction was one... so it's almost just as bad as the last one

I was (incorrectly) looking at that one linked.
I didn't actually check the list of p2pool blocks.
Though, if p2pool starts regularly making tiny blocks, it would lose it's whole "Good for Bitcoin" stand and be worse for bitcoin than the few good pools that don't do that https://kano.is/ ... as an example

[jonnybravo0311]

Ahh... I was thinking I wasn't seeing it right

By the way, I've certainly mined in your pool.  Currently my gear is on MRR and has been rented pretty much non-stop.  When it's not rented its pointed to ck's solo pool.  The only gear I've got on p2pool any more are 2 S3s that have been running the long-term test on OgNasty's node.