raskul
|
|
June 22, 2014, 12:32:21 PM |
|
Are you coming home today or tomorrow Tonight, but feel free to ask murdof to do the run command when he wakes from his afternoon nap. He has the login info Yeah, sorry about nagging but i am actually so hyped by the low ping i got before, so i might just try to message him. No worries mate.
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
murdof
|
|
June 22, 2014, 02:06:12 PM |
|
Iceland node up and running guys!
|
|
|
|
Collider
|
|
June 22, 2014, 02:09:23 PM Last edit: June 22, 2014, 02:22:27 PM by Collider |
|
Iceland node up and running! Preliminary stats page: http://192.71.218.197:9332/static/ Connect your miner to: 192.71.218.197:9332 The workername is your Bitcoin address. EXCELLENT ping from Thor Datacenter (Advania).
|
|
|
|
zvs
Legendary
Offline
Activity: 1680
Merit: 1000
https://web.archive.org/web/*/nogleg.com
|
|
June 22, 2014, 02:30:54 PM |
|
I wish people would realize the fallacy of this whole ping time being the most important factor nonsense. Here's my post copy & pasted from another board, in reply to someone that questioned what I was saying. Keep in mind that this is for DOGE, the shares are farther apart for Bitcoins, so latency will have less of an impact. Sure,
Find a p2pool with at least 200 shares, preferably 500+. See what the orphan rate is, ignore the DOA rate (unless it's incredibly abnormal, like p2pool.org's).
Pool A has an orphan rate of 15%, pool B has an orphan rate of 10%. You have 30ms latency to pool A, 200ms latency to pool B (0% packetloss to both). P2pool attempts to time shares 15s apart, but from my experience it's more than that... so, let's be liberal with it and say 6 shares per minute, plus one block per minute. That's seven interruptions per minute.
There's a 170ms latency difference, which ends up being 340ms because of the round trip. Re: Your machine gets work, an interrupt occurs a nanosecond later, your machine keeps working on the old work for ~170ms, gets the new work and then there's another 170ms until that new work can be submitted... so then 7*340ms, 2380ms, 2.38s of dead time per minute. That's roughly a 4% DOA differential between Pool A and pool B.
So then you have, say, expected DOA of 1% on pool A, expected DOA of 5% on pool B.... adding those to the orphan rates, it becomes 16% for pool A and 15% for pool B, which means you'd be better off on pool B.
|
|
|
|
Collider
|
|
June 22, 2014, 02:45:17 PM |
|
DOA rates (for shares) at Bitcoin p2pools are however atleast 2.5%.
Reducing the ping from 40ms to under 1ms improved my DOA rate from ~5% to 2.5%.
The main thing you should look out for is that your DOA rate is lower than the average P2Pool rate, as you will produce less invalid shares, which will in turn increase your payout per block.
Am I correct in the assumption that only shares "expire" within the 10 second timeframe and found blocks don´t expire that fast?
|
|
|
|
raskul
|
|
June 22, 2014, 07:48:42 PM |
|
I wish people would realize the fallacy of this whole ping time being the most important factor nonsense.
Ping time is not the most important factor, but it does make a difference, as you will acknowledge. Our miners do realise there are other important factors and what we aim to do is ensure that there are nodes in regions of the world where there are may be no nodes, or the nodes available are not as stable; to allow the option for miners to mine on p2pool. The more public nodes across the world, the more options miners will have, in the future and the more stable the p2pool network will be. We simply aim to have powerful servers which run stable and reliable nodes for p2pool bitcoin mining. It's not about being the only participant, or to monopolise on p2pool nodes; rather, the opposite - if you are able to find a lower rate of rejects somewhere else, then please do not feel that centralcavern nodes are the preferable option for you. In some regions of the world it can be more costly to run a VPS for single miner use, so we just want to try and help in cases such as this.
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
RoadStress
Legendary
Offline
Activity: 1904
Merit: 1007
|
|
June 22, 2014, 08:29:46 PM |
|
I jumped on BTCGuild luck's train, but I will be back.
|
|
|
|
CentralCavern (OP)
Newbie
Offline
Activity: 28
Merit: 0
|
|
June 22, 2014, 08:35:32 PM |
|
No problem RoadStress - milk it! As noted earlier, by the most excellent murdof; our ICELAND node is now active. http://is.centralcavern.uk:9332 just waiting on the blockchain downloading on our Middle East node - watch this space!
|
|
|
|
raskul
|
|
June 22, 2014, 11:23:15 PM |
|
Middle East node will be ready tonight
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
CentralCavern (OP)
Newbie
Offline
Activity: 28
Merit: 0
|
|
June 23, 2014, 10:05:59 AM |
|
Middle East p2pool node is now active and accepting miners point your miners to; http://me.centralcavern.uk:9332username is your bitcoin wallet address password can be anything. good luck!
|
|
|
|
raskul
|
|
June 23, 2014, 11:57:34 AM |
|
Now working on a US(west) Bitcoin p2pool node - and also altcoin pools.
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
CentralCavern (OP)
Newbie
Offline
Activity: 28
Merit: 0
|
|
June 23, 2014, 03:56:18 PM Last edit: June 24, 2014, 10:14:53 AM by CentralCavern |
|
|
|
|
|
RoadStress
Legendary
Offline
Activity: 1904
Merit: 1007
|
|
June 23, 2014, 04:25:25 PM |
|
IS node
LOST CONTACT WITH BITCOIND for 3.1 minutes! Check that it isn't frozen or dead!
|
|
|
|
raskul
|
|
June 23, 2014, 04:27:11 PM |
|
IS node
LOST CONTACT WITH BITCOIND for 3.1 minutes! Check that it isn't frozen or dead!
restarted bitcoind. this is the 2nd time that this has happened. i'll need to investigate ADD: i'd advise making the EU node your failover, that way you won't lose any shares if this happens again.
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
Acejam
|
|
June 23, 2014, 04:42:17 PM |
|
IS node
LOST CONTACT WITH BITCOIND for 3.1 minutes! Check that it isn't frozen or dead!
restarted bitcoind. this is the 2nd time that this has happened. i'll need to investigate ADD: i'd advise making the EU node your failover, that way you won't lose any shares if this happens again. People aren't smart. Failover should always be server-side.
|
|
|
|
raskul
|
|
June 23, 2014, 04:47:55 PM |
|
well the issue wasn't to do with the server; we are running at nowhere near capacity on that node... i'll ask the boss what he thinks once he's walked the dog and had his meal.
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
raskul
|
|
June 23, 2014, 05:38:36 PM |
|
I've contacted the VPS host to request an upgrade to the memory, Iceland node will be restarted in due course once memory is upgraded.
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
Collider
|
|
June 23, 2014, 06:01:37 PM |
|
Is the Tekcoin pool a p2pool node?
If yes, where is it located?
|
|
|
|
raskul
|
|
June 23, 2014, 06:05:24 PM |
|
Is the Tekcoin pool a p2pool node?
If yes, where is it located?
no it is MPOS on PPLNS so latency isn't an issue
|
tips 1APp826DqjJBdsAeqpEstx6Q8hD4urac8a
|
|
|
Collider
|
|
June 23, 2014, 06:08:38 PM |
|
Is the Tekcoin pool a p2pool node?
If yes, where is it located?
no it is MPOS on PPLNS so latency isn't an issue What is MPOS?
|
|
|
|
|