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

[kavjlaeg]

To connect to me:  whortonda.noip.me:9333

Hello.
I try to connect Avalon6 , node not responding...

[m_a17]

Success is obvious
 http://picua.org/img/2017-05/04/r43fd62zfbzs3gl0j1m0cjvay.gif

[whortonda]

@kavjlaeg: 9333 is the p2p port; I am using a non-standard worker port 3000.  I edited my original post to clarify.  Anyone is welcome to mine to my node.  I can't guarantee 100% uptime, but if it becomes unresponsive I'll work to restore as soon as possible.

Full disclosure: I am merge mining several coins on my node. Merge mined coins are not shared among miners, so by mining to my node you will be generating free coins for me.

Second block on jtoomim fork found: https://blockchain.info/block/000000000000000001a22e8f0a4b0cc86e6a301d3b0bbc6022a7d1a67139eb5c
998kB / 13.82763859 BTC

Hashrate on the fork increasing as well-- 15PH for the last few hours.

[m_a17]

Hashrate on the fork increasing as well-- 15PH for the last few hours.

fork = Global Pool Hashrate: 20.26 PH/s (Rejected 1.56 PH/s / 7.69%) 
The exodus of the hordes into fertile lands

[in2tactics]

I guess I will be reconfiguring my node for the fork sooner rather than later.

[jtoomim]

fork = Global Pool Hashrate: 20.26 PH/s (Rejected 1.56 PH/s / 7.69%)  

So much for doing a controlled small-scale performance test. I guess it's trial by fire. Tally-ho!

So far, the jtoomimnet fork has generated four blocks. All four blocks were larger than 998 kB. Two separate mining addresses have found two blocks each. I think both of them are Nicehash miners, but I only know one of the miners. (whortonda)

I'm back at my Washington mine, but I've got a bunch of mine-related stuff to take care of first before I can work on the DNS seeds and the peer-banning issue that some people are having with joining the fork.

Some preliminary observations on fairness: the large miners (at least whortonda) appear to have a substantial advantage in terms of orphan rate (0.83% for him vs 3.6% to 5.3% for my nodes). However, as nicehash renters, they have a huge disadvantage in terms of DOA rates (4.5% for him vs 0.0% to 0.9% for my nodes). So overall, we are pretty much at parity. That's pretty much the ideal result, but not for the right reasons. Oh well, good enough for now.

The most relevant question for the fork isn't whether jtoomimnet is fair, but whether jtoomimnet is more fair or less fair than mainnet is. Unfortunately, I don't have any good data on that question right now. When I get time, I might try to look through some of our logs and make some comparisons. The baseline orphan rates are much higher on mainnet, and the noise from that makes it much harder to compare orphan rates than on jtoomimnet, so I'm not sure I'll be easily able to get enough precision to definitively answer the question.

[KorbinDallas]

Still having issues connecting to the hardfork. Not able to establish a connection with any peers. As p2poolers we need to stick together. Don't forget about us on mainet 

[nreal]

Share difficulty is adjusted too low. Theres no advantage if shares get old before block is found.. Like if now expected time to block is 10 days and shares ttl is 2 days :/ It should be much more difficult to find a share and maybe the share could be calculated to live at least 2 blocks time. So share would be valid for 20 days... Keeping easy to find shares is stupid and prevents big miners as well as small miners to get decent payout.. Min share difficulty is now 1234113 an it should be at least 30x that.. Every share counts but not pays..

[-ck]

Share difficulty is adjusted too low. Theres no advantage if shares get old before block is found.. Like if now expected time to block is 10 days and shares ttl is 2 days :/ It should be much more difficult to find a share and maybe the share could be calculated to live at least 2 blocks time. So share would be valid for 20 days... Keeping easy to find shares is stupid and prevents big miners as well as small miners to get decent payout.. Min share difficulty is now 1234113 an it should be at least 30x that.. Every share counts but not pays..

I suggest making share life a function of current network difficulty rather than time.

[nreal]

Share difficulty is adjusted too low. Theres no advantage if shares get old before block is found.. Like if now expected time to block is 10 days and shares ttl is 2 days :/ It should be much more difficult to find a share and maybe the share could be calculated to live at least 2 blocks time. So share would be valid for 20 days... Keeping easy to find shares is stupid and prevents big miners as well as small miners to get decent payout.. Min share difficulty is now 1234113 an it should be at least 30x that.. Every share counts but not pays..

I suggest making share life a function of current network difficulty rather than time.

Yes its based on difficulty Everything was calculated right at some point, if I remember right one share was calculated to live 3 blocks time. Nowdays the math is done wrong, share is calculated to live 0,2 blocks time or something Renting hashpower is very risky bacause shares dont pay out etc..

[in2tactics]

Share difficulty is adjusted too low. Theres no advantage if shares get old before block is found.. Like if now expected time to block is 10 days and shares ttl is 2 days :/ It should be much more difficult to find a share and maybe the share could be calculated to live at least 2 blocks time. So share would be valid for 20 days... Keeping easy to find shares is stupid and prevents big miners as well as small miners to get decent payout.. Min share difficulty is now 1234113 an it should be at least 30x that.. Every share counts but not pays..

I suggest making share life a function of current network difficulty rather than time.

I agree with -ck. It should be a function of current network difficulty.

Yes its based on difficulty Everything was calculated right at some point, if I remember right one share was calculated to live 3 blocks time. Nowdays the math is done wrong, share is calculated to live 0,2 blocks time or something Renting hashpower is very risky bacause shares dont pay out etc..

However, I disagree with nreal. It should not be related to anything higher than 1 block time. I am all for renting hash on p2pool mainnet, but renting hash is always a risk and that risk should be reflected in share life on mainnet. I do realize that jtoomim is the one working on the updated fork so ultimately what he codes is up to him, but it is my opinion that p2pool for mainnet should always favor long-term p2pool miners not renters. If that is a problem, then the solution may be to have two separate p2pool pools, one for mainnet and one for renters.

[nreal]

Share difficulty is adjusted too low. Theres no advantage if shares get old before block is found.. Like if now expected time to block is 10 days and shares ttl is 2 days :/ It should be much more difficult to find a share and maybe the share could be calculated to live at least 2 blocks time. So share would be valid for 20 days... Keeping easy to find shares is stupid and prevents big miners as well as small miners to get decent payout.. Min share difficulty is now 1234113 an it should be at least 30x that.. Every share counts but not pays..

I suggest making share life a function of current network difficulty rather than time.

I agree with -ck. It should be a function of current network difficulty.

Yes its based on difficulty Everything was calculated right at some point, if I remember right one share was calculated to live 3 blocks time. Nowdays the math is done wrong, share is calculated to live 0,2 blocks time or something Renting hashpower is very risky bacause shares dont pay out etc..

However, I disagree with nreal. It should not be related to anything higher than 1 block time. I am all for renting hash on p2pool mainnet, but renting hash is always a risk and that risk should be reflected in share life on mainnet. I do realize that jtoomim is the one working on the updated fork so ultimately what he codes is up to him, but it is my opinion that p2pool for mainnet should always favor long-term p2pool miners not renters. If that is a problem, then the solution may be to have two separate p2pool pools, one for mainnet and one for renters.

Share should always pay out - thats the idea i guess. Dont mix things with rented or not rented. If one uses 2ph for 24 hours and dont get any payment for that the p2pool sucks because these easy to find shres makes sharechain go too fast round.. Too small share diff doesnt help anyone.

[frodocooper]

Yes its based on difficulty Everything was calculated right at some point, if I remember right one share was calculated to live 3 blocks time. Nowdays the math is done wrong, share is calculated to live 0,2 blocks time or something Renting hashpower is very risky bacause shares dont pay out etc..

However, I disagree with nreal. It should not be related to anything higher than 1 block time. I am all for renting hash on p2pool mainnet, but renting hash is always a risk and that risk should be reflected in share life on mainnet. I do realize that jtoomim is the one working on the updated fork so ultimately what he codes is up to him, but it is my opinion that p2pool for mainnet should always favor long-term p2pool miners not renters. If that is a problem, then the solution may be to have two separate p2pool pools, one for mainnet and one for renters.

I concur with in2tactics on this one.

Share should always pay out - thats the idea i guess. Dont mix things with rented or not rented. If one uses 2ph for 24 hours and dont get any payment for that the p2pool sucks because these easy to find shres makes sharechain go too fast round.. Too small share diff doesnt help anyone.

Shares do pay out. If they didn't, we wouldn't be here. The difference between (short-term) rented hashpower and non-rented hashpower on P2Pool is that the renter is essentially gambling that P2Pool finds a block while his shares remain in the sharechain (about 3 days). Their payouts therefore rely heavily on chance. Just because someone uses 2PH/s for 24 hours on P2Pool and doesn't get a payout, doesn't mean that P2Pool "sucks." All it means is that his or her gamble didn't pay off.

Also, if you're unsatisfied with P2Pool's default share difficulty, you can always set your own by appending "+", followed by your preferred difficulty level, to the end of your P2Pool mining address. And if you prefer to only have "difficult-to-find" shares on P2Pool's sharechain, simply append "/", followed by your preferred difficulty level, to the end of your P2Pool mining address.

[in2tactics]

Also, if you're unsatisfied with P2Pool's default share difficulty, you can always set your own by appending "+", followed by your preferred difficulty level, to the end of your P2Pool mining address. And if you prefer to only have "difficult-to-find" shares on P2Pool's sharechain, simply append "/", followed by your preferred difficulty level, to the end of your P2Pool mining address.

What nreal is talking about is the p2pool pseudo-shares. The "+" and "/" options do not determine what shares get submitted to the p2pool share chain.

Share should always pay out - thats the idea i guess. Dont mix things with rented or not rented. If one uses 2ph for 24 hours and dont get any payment for that the p2pool sucks because these easy to find shres makes sharechain go too fast round.. Too small share diff doesnt help anyone.

I think you are missing the point. Setting the share life to 1 block as I suggested would make things fairer for everyone to include renters by increasing the probably that anyone mining during the current round will get a payout by the time p2pool finds its next block, but it does not guarantee it. Making the share life higher than 1 block as you suggested only serves to benefit renters by lowering their risk and decreasing potential earnings for long-term p2pool miners. I do not think that arbitrarily decreasing renter risk at the expense of loyal long-term miners is the right answer. This very topic has been a source of contention for pools since their inception and in my opinion using a share life of 1 block seems to be the best compromise for everyone.

[frodocooper]

Also, if you're unsatisfied with P2Pool's default share difficulty, you can always set your own by appending "+", followed by your preferred difficulty level, to the end of your P2Pool mining address. And if you prefer to only have "difficult-to-find" shares on P2Pool's sharechain, simply append "/", followed by your preferred difficulty level, to the end of your P2Pool mining address.

What nreal is talking about is the p2pool pseudo-shares. The "+" and "/" options do not determine what shares get submitted to the p2pool share chain.

Actually, they do. Or at least, I've been under the impression that they do.

The specified difficulty value after the "/" tells the miner to send only shares above the specified "/" value to the sharechain. If the specified difficulty value after the "/" is lower than the current P2Pool minimum difficulty, then the "/" value is ignored. So, if a miner desires to have only high-difficulty shares in the sharechain, he or she may do so using the "/" function.

The specified difficulty value after the "+" tells P2Pool to send only work of the specified "+" difficulty value to the miner, i.e., the pseudo-shares.

Share should always pay out - thats the idea i guess. Dont mix things with rented or not rented. If one uses 2ph for 24 hours and dont get any payment for that the p2pool sucks because these easy to find shres makes sharechain go too fast round.. Too small share diff doesnt help anyone.

I think you are missing the point. Setting the share life to 1 block as I suggested would make things fairer for everyone to include renters by increasing the probably that anyone mining during the current round will get a payout by the time p2pool finds its next block, but it does not guarantee it. Making the share life higher than 1 block as you suggested only serves to benefit renters by lowering their risk and decreasing potential earnings for long-term p2pool miners. I do not think that arbitrarily decreasing renter risk at the expense of loyal long-term miners is the right answer. This very topic has been a source of contention for pools since their inception and in my opinion using a share life of 1 block seems to be the best compromise for everyone.

I'm getting a little confused here. Do correct me if I'm wrong. From what I am able to understand from your suggestion, you are suggesting that a share's lifetime be limited to the current round of mining, i.e., the sharechain is reset after P2Pool finds a block?

If that's the case, then the payout system would be more along the lines of the "Proportional" payout system rather than P2Pool's current PPLNS system. If what you are suggesting is a "Proportional"-like system, then I think it would be unfair to long-term miners, since the "Proportional" payout system is susceptible to pool-hopping.

Then again, I'm probably misunderstanding the whole discussion. Some clarification would therefore be nice, and much appreciated.

[flameruk]

Hi.

Ive been working away for a week so im missing all the "hard fork" talk on here.
I see the pool hash has dropped and read a thread that the hard fork option now has the greater hash rate.
Is it time for me to jump to the new fork?

Yes im being very lazy not reading back through the post for which I apologize right now.
Im about to head out again for a weekend job away but this is playing on my mind now.

[notabeliever]

Maybe its the nature of p2pool's but what bothers me is the strain on my avalon A721 getting push to excessively high difficulties where it starts to under perform.
The flip side of that  is when I set my own difficulty to 4096 I can't find any shares and on several occasions have missed 4 rewards from blocks because my miner was unable to find just 1 share.

Are their any nodes that have controlled the pool difficulty as when I mine on other types of pools my miner chuggs along nicely.

So what would be the recommended difficulty to use.

[windpath]

Maybe its the nature of p2pool's but what bothers me is the strain on my avalon A721 getting push to excessively high difficulties where it starts to under perform.
The flip side of that  is when I set my own difficulty to 4096 I can't find any shares and on several occasions have missed 4 rewards from blocks because my miner was unable to find just 1 share.

Are their any nodes that have controlled the pool difficulty as when I mine on other types of pools my miner chuggs along nicely.

So what would be the recommended difficulty to use.

Without being sure I suspect the problem your having is that the Avalon can not handle P2Pools frequent work restarts. Every time a share is found (~30 seconds) the pool tells your miner to throw away what's it's working on and start fresh on the next share. This happens much more often on P2Pool, and the Avalon may not be able to handle it.

[nreal]

Yes its based on difficulty Everything was calculated right at some point, if I remember right one share was calculated to live 3 blocks time. Nowdays the math is done wrong, share is calculated to live 0,2 blocks time or something Renting hashpower is very risky bacause shares dont pay out etc..

However, I disagree with nreal. It should not be related to anything higher than 1 block time. I am all for renting hash on p2pool mainnet, but renting hash is always a risk and that risk should be reflected in share life on mainnet. I do realize that jtoomim is the one working on the updated fork so ultimately what he codes is up to him, but it is my opinion that p2pool for mainnet should always favor long-term p2pool miners not renters. If that is a problem, then the solution may be to have two separate p2pool pools, one for mainnet and one for renters.

I concur with in2tactics on this one.

Share should always pay out - thats the idea i guess. Dont mix things with rented or not rented. If one uses 2ph for 24 hours and dont get any payment for that the p2pool sucks because these easy to find shres makes sharechain go too fast round.. Too small share diff doesnt help anyone.

Shares do pay out. If they didn't, we wouldn't be here. The difference between (short-term) rented hashpower and non-rented hashpower on P2Pool is that the renter is essentially gambling that P2Pool finds a block while his shares remain in the sharechain (about 3 days). Their payouts therefore rely heavily on chance. Just because someone uses 2PH/s for 24 hours on P2Pool and doesn't get a payout, doesn't mean that P2Pool "sucks." All it means is that his or her gamble didn't pay off.

Also, if you're unsatisfied with P2Pool's default share difficulty, you can always set your own by appending "+", followed by your preferred difficulty level, to the end of your P2Pool mining address. And if you prefer to only have "difficult-to-find" shares on P2Pool's sharechain, simply append "/", followed by your preferred difficulty level, to the end of your P2Pool mining address.

Those were the days when a share used to be valid for 3 days. Everyone was whining at the time that its too difficult to find a share and then it was adjusted so a share is nowdays valid for a day. It doesnt help if a big miner raises his shares to max which is 30x the minimium. Its like flooding the system with tiny shares. And no you dont get paid for every share you get like you should because shares are too fucking small and small shares flood the system.. But share should be valid not 3 days but somewhere near the shares needed to get 2 blocks, 1 block at least which was earlier today 8 days for 1 block now its 7.9 days - so difficulty should be so that shares would be valid for lets say 12 days. Then it would be like fun to get few shares and then few blocks..

[flameruk]

Hello, add p2p node

classic p2pool http://low-doa.mine.nu:9332/static/
hardforked p2pool http://low-doa.mine.nu:9334/static/

Location Russia, Moscow city, installed and configured by sawa.

------------------
CPU: i7 7700k 4.2 GHz
RAM: 16GB DDR4
HDD: 2x PCI-E NVMe SSD
Pipe: 50m/50m optic cable

Hi Kavjlaeg

Im using your hard fork node at the moment from the UK so I can mine the new chain while I re-configure my node server and do a little maintenance.
Working OK from the UK.

Theres another guy using my node at ukp2pool.uk which is going to go down for a mod to the hardfork version.
Hope you got a backup pool configured .......