Re: PhoenixMiner 2.8c: fastest Ethereum/Ethash miner with lowest devfee (Windows)

Quote from: PhoenixMiner on April 05, 2018, 06:02:17 AM

Any plans for CryptoNote miner or integration, would be great?

Not in the immediate future but in a few months a lot could and probably would change. Wink

Quote from: Jbodz83 on April 03, 2018, 08:20:34 PM

This is my hashrate and stale share percentage in Claymore 11.6

https://image.ibb.co/iz6vBc/ETC.png

Phoenix can tap my RIG up to 395Mhs (13GPU) however, it gives me 5-6% stale share..

i like the hashrate spike from PhoenixMiner, but i guess effeciency goes to Claymore 11.6 - hope that PhoenixMiner could catch up soon.. Cool

This problem seems to impact a fair share of users so we tested like crazy but the results are really puzzling. For example, in the most pure form of testing - with a test network proxy that eliminates all latencies, the difference is less than 1% (as it should be) and virtually zero when using the new 2.9 branch which lowers the internal stale shares to about 0.1%. Also, when checking out one of the many devfee addresses of Claymore, we get the following on ethermine, ethpool and ethermine-etc at the moment of this writing (this is a great way of comparison as represents the average stale shares of many rigs):

https://www.ethermine.org/miners/34FAAa028162C4d4E92DB6abfA236A8E90fF2FC3/dashboard	4%
https://www.ethpool.org/miners/34FAAa028162C4d4E92DB6abfA236A8E90fF2FC3/dashboard	3%
https://etc.ethermine.org/miners/34FAAa028162C4d4E92DB6abfA236A8E90fF2FC3/dashboard	4%

Now, compare this to our devfee wallet (we have three on ethermine, and one on ethpool and ethermine.etc):

https://www.ethermine.org/miners/00d4405692b9F4f2Eb9E99Aee053aF257c521343/dashboard	5%
https://www.ethpool.org/miners/00d4405692b9F4f2Eb9E99Aee053aF257c521343/dashboard	3%
https://etc.ethermine.org/miners/d549Ae4414b5544Df4d4E486baBaad4c0d6DcD9d/dashboard	4%

Quote

The difference is way smaller than what is observed by a lot of users (and confirmed by our own tests when we use a single mining rig or small group of rigs). As we said, in version 2.9, the internal stale shares are almost zero (the job change latency from the moment the new job is received from the pool to the moment first hash is computed is less than 30 ms), so if this doesn't solve the problem, it would be very strange indeed.

appreciate the time comparing and clear justification, i am looking forward to see 2.9 outperform Claymore.
what i like about PM is that it could maximize my GPU's capability to hash up to 31.2Mhs for 580's which has a difference of .5 Mhs from Claymore.
having 1-2% of stale share is reasonable enough to shift to PM, i am still trying out other settings using PM though.

thanks and appreciate your feedback.