p2pool - Decentralized, Absolutely DoS-Proof, Pool Hopping-Proof Pool [archival]

[BombaUcigasa]

Yes, and another 0.5% to the block finder to disincentivize withholding blocks.

I didn't read all the code thoroughly, but why can't peers change their assigned shares to include a higher proportion, so if they do find the block, they get more payout? And still on the subject, alter the code to check for the blockchain difficulty (and not that of the sharechain) and mine for themselves while signalling less difficult shares to the p2poll?

[BombaUcigasa]

A slight benefit of using p2pool that should be mentioned: rewards are truely anonymous, you get the virgin bitcoins directly from generation blocks, they have no history, when you use them for the first time, nobody will know where they come from, they can't be traced.

[forrestv]

I didn't read all the code thoroughly, but why can't peers change their assigned shares to include a higher proportion, so if they do find the block, they get more payout? And still on the subject, alter the code to check for the blockchain difficulty (and not that of the sharechain) and mine for themselves while signalling less difficult shares to the p2poll?

If they change their payout proportion in any way, other nodes won't accept their shares and they will register as stale, essentially resulting in solo mining.

The blockchain difficulty isn't used for anything except determining if a share solved a block and how far back to set the payouts... and you can't selectively mine over the block difficulty range solo and the share difficulty range with p2pool. The output hash is unknown and depends how you're mining.

[Sukrim]

Isn't requiring your users to run a fullsized bitcoind node going to kill this pool sooner or later? Scalability issues come to mind...

Block withholding would anyways not really work in a PPLNS pool like this one, as you only hurt the operator for his 0.5% - you can't like in PPS suck his wallet dry.
A 0.5% "lottery" however is still ok imho, even though it adds unnecessary variance.

[BombaUcigasa]

Isn't requiring your users to run a fullsized bitcoind node going to kill this pool sooner or later? Scalability issues come to mind...

It's explained that you can use a low bandwidth mode, the sharechain has an increased difficulty (>1 but < blockchain) thus share broadcasting does not flood the network, and you don't need to broadcast to all peers, relaying takes care of stats propagation.

But yes, scalability issues come to mind, the best way to do it is to have a pushpool mine in p2pool, and smaller individual miners participate in the pushpool.

I didn't read all the code thoroughly, but why can't peers change their assigned shares to include a higher proportion, so if they do find the block, they get more payout? And still on the subject, alter the code to check for the blockchain difficulty (and not that of the sharechain) and mine for themselves while signalling less difficult shares to the p2poll?

If they change their payout proportion in any way, other nodes won't accept their shares and they will register as stale, essentially resulting in solo mining.

The blockchain difficulty isn't used for anything except determining if a share solved a block and how far back to set the payouts... and you can't selectively mine over the block difficulty range solo and the share difficulty range with p2pool. The output hash is unknown and depends how you're mining.

Ah, I see, so if you send out altered payouts with fake shares, you won't get paid at all by the peers. But what if YOU find the block and give yourself a bigger payout? Since you pay everyone, what can the network do to punish you, when you determine your share solved the block and give yourself a bigger cut? Obviously some people will entertain the ideas or profits of "customizing" their code instance.

[Sukrim]

I'm not so much concerened about the p2pool shares but the whole bitcoind that also needs to run. There you already need ~1 GB of space and it'll grow.

[bal3wolf]

lol your complaining about 1-3gigs of used space when a typical computer has 20-30gigs of free space even on low end computers.

[Tril]

Ah, I see, so if you send out altered payouts with fake shares, you won't get paid at all by the peers. But what if YOU find the block and give yourself a bigger payout? Since you pay everyone, what can the network do to punish you, when you determine your share solved the block and give yourself a bigger cut? Obviously some people will entertain the ideas or profits of "customizing" their code instance.

You are welcome to do this. Since you are solo mining you can pay yourself as much as you want. Most solo miners pay themselves 100%...

[bal3wolf]

I'm not so much concerened about the p2pool shares but the whole bitcoind that also needs to run. There you already need ~1 GB of space and it'll grow.

Also you could mine on the public p2p pool.


https://bitcointalk.org/index.php?topic=38421.0

[forrestv]

Important new release! UPGRADE NOW 3f4e4fa
Windows binary: http://u.forre.st/u/xjsqpiql/p2pool_win32_3f4e4fa.zip

This release changes the share generation code so that the main Bitcoin p2pool can switch from a 5 second to a 10 second share period. This is an effort to decrease the number of stales, following a successful trial on the namecoin p2pool, where we had a constant 9% stale rate with a 10 second time between shares.

In three days, the main Bitcoin p2pool will switch from a 5 second share period to a 10 second share period. You MUST upgrade before then to continue mining.

Other changes:

• Much more obvious notification if the connection to bitcoind has been lost
• Each network now has its own log file name and worker interface port, so it's easier to run multiple p2pool nodes in parallel
• Several small bugfixes and UI improvements

[jonathan]

I guess more people would try p2pool if it worked with most miners (like cgminer or poclbm).

I will stop mining with mtred to move to p2pool just as soon as poclbm works at p2pool. I check for any progress on this daily! I've nothing against mtred, they have been fine. I would just prefer p2p mining for the usual predictable reasons.

[forrestv]

I guess more people would try p2pool if it worked with most miners (like cgminer or poclbm).

I will stop mining with mtred to move to p2pool just as soon as poclbm works at p2pool. I check for any progress on this daily! I've nothing against mtred, they have been fine. I would just prefer p2p mining for the usual predictable reasons.

Poclbm has worked with p2pool for a long time - you just need to download the alternative version from https://github.com/forrestv/poclbm and get the fast_long_polling branch. Do you need me to build a Windows binary?

[bal3wolf]

I guess more people would try p2pool if it worked with most miners (like cgminer or poclbm).

I will stop mining with mtred to move to p2pool just as soon as poclbm works at p2pool. I check for any progress on this daily! I've nothing against mtred, they have been fine. I would just prefer p2p mining for the usual predictable reasons.

Poclbm has worked with p2pool for a long time - you just need to download the alternative version from https://github.com/forrestv/poclbm and get the fast_long_polling branch. Do you need me to build a Windows binary?

That would be nice i tried but i suck at anything coding lol.

[forrestv]

That would be nice i tried but i suck at anything coding lol.

Uh .. can you explain the problem in a bit more detail? There is no coding involved. Are you using Linux or Windows?

If you're using Windows, tell me, and I'll make a binary.

If you're using Linux, all you have to do is download https://github.com/forrestv/poclbm/tarball/fast_long_polling and use it like normal poclbm.

[bal3wolf]

windows i mean you gota have  all kinds of packs to compile it i never got them all installed right  kept getting errors when id run it or make it a exe.

[Christian Pezza]

how I can see the actual speed of the pool without join it?

thank you

[BombaUcigasa]

how I can see the actual speed of the pool without join it?

thank you

on the irc channel

[Christian Pezza]

can u please give me the channel name thanks

[BombaUcigasa]

I'm available on IRC on Freenode for any questions/discussion (freenode, nick: forrestv, channel: #bitcoin-dev or #p2pool).

[deti]

I'm a bit confused with the fees taken. Is it right how I understood it?:

Code:

  50 BTC/block
+ X.XXXXXXXX BTC from fees of transactions in the solved block
= Total BTC to be distributed

Now:
0,5% of 'Total BTC' goes to the one who finds the block
0,5% of 'Total BTC' goes to forrestv
What remains is PPLNS (Payed Per Last N Shares)

Please confirm or correct me. Thanks.