kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
June 06, 2014, 03:15:30 AM |
|
The last that was heard from the lead developer was that a pretty significant redesign was on the way to fix the high variance problem for smaller miners. I would assume he's still working on it.
I think the point is that p2p pooled mining is a relatively hard problem to solve. Centralised pools can make use of known, easily implemented infrastructure designs, but p2p mining needs innovative software to make the network topology viable for the purpose. This design may not now and may never be perfect, but it's a very good choice for owners of large hashrates, and even for those with smaller or middle sized rigs that already feel confident they will profit on their investment.
What were you smoking when you wrote that?
|
|
|
|
norgan
|
|
June 06, 2014, 06:02:35 AM |
|
I have a bit of hosting space left. where are we in most need of a p2pool node? US, EU, Asia, Japan?
I already have au and Singapore nodes. was thinking of putting in a US node.
|
|
|
|
Farmer17
|
|
June 06, 2014, 08:29:07 AM |
|
The larger p2pool is, the higher the share difficulty. Right now the share difficulty is 2million. Yes, that makes variance large for little guys. Has always been a downside of p2pool. In today's day and age, 200gh/s isn't that much. At that share difficulty, how much hashrate do I need to have, so I could expect 1 share a day on average? I'm guessing about 100gh/s. M You can figure it out using the following formula: Difficulty * 2**32 / hashrate / 86400 = number of days to find a share Since you want the days to find a share to be 1, you can solve for hash rate. 2272866.97 * 2**32 / hashrate / 86400 = 1 hashrate = 112.98482991097931851851852 GH/s as of right now Thanks very much for the formula and the explanation.
|
|
|
|
IYFTech
|
|
June 06, 2014, 08:46:09 AM |
|
The last that was heard from the lead developer was that a pretty significant redesign was on the way to fix the high variance problem for smaller miners. I would assume he's still working on it.
I think the point is that p2p pooled mining is a relatively hard problem to solve. Centralised pools can make use of known, easily implemented infrastructure designs, but p2p mining needs innovative software to make the network topology viable for the purpose. This design may not now and may never be perfect, but it's a very good choice for owners of large hashrates, and even for those with smaller or middle sized rigs that already feel confident they will profit on their investment.
What were you smoking when you wrote that? I just spent the last 24hrs trying to put into words how much sense that statement didn't make, but I couldn't. Thank you Kano. For the laugh too! Spot on.
|
|
|
|
Carlton Banks
Legendary
Offline
Activity: 3430
Merit: 3080
|
|
June 06, 2014, 09:34:14 AM |
|
I'm not sure that anyone else has come up with a design for p2p pooled mining software, although you're both welcome to prove me wrong. If you can come up with something useful yourselves, I think everyone would thank you.
|
Vires in numeris
|
|
|
norgan
|
|
June 06, 2014, 10:35:11 AM |
|
Since it worked last time (kinda), more blocks please
|
|
|
|
PatMan
|
|
June 06, 2014, 11:31:55 AM |
|
I'm not sure that anyone else has come up with a design for p2p pooled mining software, although you're both welcome to prove me wrong. If you can come up with something useful yourselves, I think everyone would thank you.
Actually, I was so sick & tired of the lack of development on p2pool over a year ago that I put out some feelers with the aim of doing exactly that: https://bitcointalk.org/index.php?topic=213051.0I was just testing the waters, but the main aim was to give the dev here a bit of a kick up the backside & hopefully get p2pools problems sorted out. Unfortunately, due to other commitments, I was unable to push on with the project. It seems however that the dev here didn't get the message, because since then, absolutely no progress has been made, there has been zero development on p2pool whatsoever. It's just crazy. In software terms, even 6 months of no development is a lifetime - but over 12 months is software suicide, especially in the Bitcoin environment where development is moving at breakneck speed. If software devs are not moving forwards constantly, then they are going backwards - meaning p2pool is currently going in reverse at lightspeed. I was also in discussions with a dev friend of mine not so long ago about creating a p2pool installer in an effort to make installing p2pool easier for the average user to install & run p2pool & thus generate more potential miners, but this project was also shelved because it just seemed pointless introducing people to software that has so many incompatibility issues & is not maintained/updated or been improved upon in over a year. Madness. It wouldn't be so bad if the dev here bothered to post more often on what, if any, work was being done to the code to improve things. There's a lack of discussion between the dev & the users of p2pool generally, leaving every p2pool user who posts on this thread having to "assume" that something is being done because they haven't heard anything otherwise. There are many talented programmers/coders on this thread who I'm sure would be more than willing to help the dev with any issues/ideas/suggestions for improvement in p2pool - but the lack of conversation from the dev make this impossible. We are now in the unenviable position where large scale miners are unable to use p2pool for the reasons stated by member flound1129 ( https://bitcointalk.org/index.php?topic=18313.msg7118016#msg7118016) and small miners can't use p2pool because of variance - leaving us with the bit in the middle, of which all are happy to be ripped off & ignored by the likes of ghash.io because it's just easier & has a pretty web interface. If it was bad a year ago, it's even worse now. It's not looking good at all.
|
|
|
|
windpath
Legendary
Offline
Activity: 1258
Merit: 1027
|
|
June 06, 2014, 01:55:56 PM Last edit: June 06, 2014, 02:51:26 PM by windpath |
|
Hey Folks! While I did not get as much done as I wanted I promised an update this week including access to the data from all historical p2pool blocks. There is still a lot in the development pipeline, but wanted to get this out today: For all p2pool miners, regardless of what node you mine on:You can now view your total p2pool earnings and your payouts per found block on our miner dashboard for all time. If you do mine on my node you will have access to all your data on the miner dashboard, I plan to make a separate dashboard soon for those that do not mine on my node so that they can simply monitor their payouts. Note: This is not a balance checker, it reports ONLY income made from mining on p2pool, as such only payments from found blocks will be shown (also does not include any mass payments from those donating to p2pool miners). To access your stats go to: http://mining.coincadence.com/index.php and enter your mining address in the "Miner Dashboard" field: Miners working on my node will see full stats: Those mining on another node will see only payout stats: More to come, I hope you all enjoy it.
|
|
|
|
norgan
|
|
June 06, 2014, 02:25:25 PM |
|
Pretty cool, nice work. This provides some nice insight.
|
|
|
|
Carlton Banks
Legendary
Offline
Activity: 3430
Merit: 3080
|
|
June 06, 2014, 04:28:19 PM |
|
I'm not sure that anyone else has come up with a design for p2p pooled mining software, although you're both welcome to prove me wrong. If you can come up with something useful yourselves, I think everyone would thank you.
Actually, I was so sick & tired of the lack of development on p2pool over a year ago that I put out some feelers with the aim of doing exactly that: https://bitcointalk.org/index.php?topic=213051.0I was just testing the waters, but the main aim was to give the dev here a bit of a kick up the backside & hopefully get p2pools problems sorted out. Unfortunately, due to other commitments, I was unable to push on with the project. It seems however that the dev here didn't get the message, because since then, absolutely no progress has been made, there has been zero development on p2pool whatsoever. It's just crazy. In software terms, even 6 months of no development is a lifetime - but over 12 months is software suicide, especially in the Bitcoin environment where development is moving at breakneck speed. If software devs are not moving forwards constantly, then they are going backwards - meaning p2pool is currently going in reverse at lightspeed. I was also in discussions with a dev friend of mine not so long ago about creating a p2pool installer in an effort to make installing p2pool easier for the average user to install & run p2pool & thus generate more potential miners, but this project was also shelved because it just seemed pointless introducing people to software that has so many incompatibility issues & is not maintained/updated or been improved upon in over a year. Madness. Then you are only confirming what I said: there is no other design for p2p pooled mining. Well done for taking the initiative to create a competing p2pool, but I think the forrestv design still wins against a pool that doesn't exist yet. It wouldn't be so bad if the dev here bothered to post more often on what, if any, work was being done to the code to improve things. There's a lack of discussion between the dev & the users of p2pool generally, leaving every p2pool user who posts on this thread having to "assume" that something is being done because they haven't heard anything otherwise. There are many talented programmers/coders on this thread who I'm sure would be more than willing to help the dev with any issues/ideas/suggestions for improvement in p2pool - but the lack of conversation from the dev make this impossible. It's difficult to assess if the project lead can spend huge amounts of time responding to concerns about development, I'm not aware of his other responsibilities outside of p2pool (I don't think p2pool donation income could pay the bills right now). He certainly provides appropriate information as and when it's necessary, but I would agree that he is pretty economical with the time he spends in this thread (although I don't totally agree that it's a bad thing). I do know that there are other p2pool dev discussions in other subs and on IRC and possibly mailing lists, although any progress has not been added to any new branch on p2pool github. The Bitcoin Core devs (such as gmaxwell, Peter Todd, wtogami & Jeff Garzik) seem to take an interest in discussing and promoting p2pool, although it has been a few months since there has been any feedback from them on anything p2pool related. If there were any good ideas coming out of their discussions, I don't think they'd hold back. We are now in the unenviable position where large scale miners are unable to use p2pool for the reasons stated by member flound1129 ( https://bitcointalk.org/index.php?topic=18313.msg7118016#msg7118016) and small miners can't use p2pool because of variance - leaving us with the bit in the middle, of which all are happy to be ripped off & ignored by the likes of ghash.io because it's just easier & has a pretty web interface. If it was bad a year ago, it's even worse now. It's not looking good at all. I don't think small miners are unable to use p2pool, it's just that it's more of a gamble because of the high share difficulty (combined with rapid multiplying of the block difficulty). You could argue the gamble would be worth it; if they got a lucky streak early on, they could get closer to ROI than with a centralised pool (and most small rig owners have a problem with over-optimistic or plain wrong estimates of ROI anyway...). And I'm not convinced that large miners can't use it either, the largest miner on the pool has somewhere around 100 TH/s. flound1129's problem sounds like hardware problems to me. There were some in this thread who called for forrestv to "fix" p2pool to work with incompatible miners, when the problem lay with the device driver, or the hardware (typically an absence of long polling or stratum support, which are not issues with the p2pool software itself; certain mining protocol features are needed for the p2pool software design to work at all)
|
Vires in numeris
|
|
|
nreal
Full Member
Offline
Activity: 932
Merit: 100
arcs-chain.com
|
|
June 06, 2014, 07:34:48 PM |
|
Can I download the php or whatever it is that P2poolinfo runs on and run it here on my server, it lags there where it now is ? http://p2pool.info/As a backup or something, its annoying when its not working
|
|
|
|
norgan
|
|
June 06, 2014, 09:34:33 PM |
|
Can I download the php or whatever it is that P2poolinfo runs on and run it here on my server, it lags there where it now is ? http://p2pool.info/As a backup or something, its annoying when its not working I already did this at p2poolinfo.azurewebsites.net but its out of date and wont populate. I don't know how to update it.
|
|
|
|
windpath
Legendary
Offline
Activity: 1258
Merit: 1027
|
|
June 06, 2014, 10:30:49 PM Last edit: June 07, 2014, 02:09:30 AM by windpath |
|
Can I download the php or whatever it is that P2poolinfo runs on and run it here on my server, it lags there where it now is ? http://p2pool.info/As a backup or something, its annoying when its not working I already did this at p2poolinfo.azurewebsites.net but its out of date and wont populate. I don't know how to update it. I should have an alternative with all the same data up by end of next week....
|
|
|
|
nreal
Full Member
Offline
Activity: 932
Merit: 100
arcs-chain.com
|
|
June 07, 2014, 01:29:44 AM Last edit: June 07, 2014, 01:44:23 AM by nreal |
|
Im not sure of this , but if you overclock something are you sure its good for your brains - those might melt. Im getting more results here with undervoltaging, might be wrong - you tell me. 1.6kw -> 600w with ants, i love ants! just have to use pencil, and a volt meter. works here https://bitcointalk.org/index.php?topic=526060.0
|
|
|
|
nreal
Full Member
Offline
Activity: 932
Merit: 100
arcs-chain.com
|
|
June 07, 2014, 01:53:17 AM |
|
it works like a charm here, but i cant say if it works for you, join to my node. just now i realised its not possible I get 2x for what im supposed to
|
|
|
|
nreal
Full Member
Offline
Activity: 932
Merit: 100
arcs-chain.com
|
|
June 07, 2014, 02:02:46 AM Last edit: June 07, 2014, 02:27:31 AM by nreal |
|
https://bitcointalk.org/index.php?topic=526060.0Go ahead for gh/ ghash pays for it, and you pay for nothing. Stupid pays for nothing, pretty soon its going to go down with this shit, like 200gh with 180 gear. One can imagine of 200gh rate with 2x 70 blades. Ofcourse you can put it how high its possible to read the stats, but its the thuth that its only 2x 70 that ant s1 is There might be a slight possibility that you hit a diamond when going high, but i dont takr my bets on it. One cant fool P2pool, because.... i have many pictures of shares how in hte earth i can paste it .. This P2pool seems to be foolproof, thank you for it.
|
|
|
|
nreal
Full Member
Offline
Activity: 932
Merit: 100
arcs-chain.com
|
|
June 07, 2014, 02:37:02 AM Last edit: June 07, 2014, 02:48:39 AM by nreal |
|
so tell me what did happen, when you undervolaged.
less shares or more? why
if i put my ants to the limit theres no gain, vs 180gh. i can laugh, because it seems to me that theres no gain going over 0.83v/275mhz/ 1.1v/350mhz
you only use more power, much more power.
350mhz is like 180gh with 360w 275mhz is like 140gh with 160w
|
|
|
|
bitdigger2013
|
|
June 07, 2014, 03:28:20 AM |
|
Hi,
Need some advice on if this seems right. I have been running on my own node for 1.2 days at 1.2 TH and my result so far does not seem right.
Local rate: 1.20 TH/s (DOA 38.57 GH/s / 3.51%) Shares: Total: 12 (Orphan: 1, Dead: 0) Expected payout: 0.01571202
is this just variance related or an obvious issue with my node?
was getting .04+ daily with ghash.io
|
|
|
|
CaptEmulation
|
|
June 07, 2014, 03:46:29 AM |
|
Hi,
Need some advice on if this seems right. I have been running on my own node for 1.2 days at 1.2 TH and my result so far does not seem right.
Local rate: 1.20 TH/s (DOA 38.57 GH/s / 3.51%) Shares: Total: 12 (Orphan: 1, Dead: 0) Expected payout: 0.01571202
is this just variance related or an obvious issue with my node?
was getting .04+ daily with ghash.io
Variance. You need 3 days to fill out your PPLNS share queue-- so you are only at 1.2/3 = 0.4 of your expected block payout. You had bad luck for 1.2 days, you could do much better. I have been purring along at 530GH/s and per-block between 0.02 to 0.028 the past several blocks.
|
|
|
|
bitdigger2013
|
|
June 07, 2014, 03:57:10 AM |
|
Hi,
Need some advice on if this seems right. I have been running on my own node for 1.2 days at 1.2 TH and my result so far does not seem right.
Local rate: 1.20 TH/s (DOA 38.57 GH/s / 3.51%) Shares: Total: 12 (Orphan: 1, Dead: 0) Expected payout: 0.01571202
is this just variance related or an obvious issue with my node?
was getting .04+ daily with ghash.io
Variance. You need 3 days to fill out your PPLNS share queue-- so you are only at 1.2/3 = 0.4 of your expected block payout. You had bad luck for 1.2 days, you could do much better. I have been purring along at 530GH/s and per-block between 0.02 to 0.028 the past several blocks. Anything I should be tuning on my end? or just need to have luck average out over time?
|
|
|
|
|