Quote from: mahrens917 on February 06, 2015, 02:05:24 AM

Quote from: windpath on February 05, 2015, 07:29:48 PM

Don't think of it as competing with miners on the same node, you are completing (and collaborating) with the whole pool.

Also, node efficiency is a poor indicator on any node with more then 1 miner. A single miner with a high latency can bring node efficiency way down, while your miners may be mining very efficiently.

Look at your latency to the node, the node getBlockTemplate latency, and your stale/DOA shares to get a feel for your particular efficiency...

After a few days of mining you can look up your miner in our "active miners" list for p2pool. That estimated hashrate, if higher then expected, means you are mining efficiently/currently lucky. If it is lower then expected you are mining inefficiently or in a streak of bad luck. (this only works after you have been mining for the length of the share chain, i.e. 3 days)

http://minefast.coincadence.com/p2pool-stats.php

Thanks Windpath. Is node efficiency with perfect miners (the stat) a good indication of the quality of the node? I agree just plain node efficiency can be skewed by individual miners.

I don't think any 1 factor can really determine the quality of a node for a given miner, it's a combination of factors...

The "efficiency with perfect miners" stat simply excludes DOA shares from the "efficiency" stat.

If you have a miner with a high stale/orphan rate it will go down just like the efficiency stat, and could be caused by various problems.

In addition to your latency to the node, the node getBlockTemplate latency, and your stale/DOA shares mentioned above; you would also need to take into account node stability/response time, the number of incoming/outgoing p2pool peers and their latency from the node, and some other factors I'm probably forgetting to really get the whole picture.

There is must be a great formula to weight and calculate all this and put it in simple terms, I just don't know it Wink