Status as of Monday, September 4, 2017

[Steve Sokolowski]

Good morning!

• We issued a new release yesterday, as is customary on Sundays. This time, we modified the sign-in phase for miners so that when they connect the mining server, they will now wait to be issued work until the next work assignment happens for all miners, instead of immediately receiving work. This change significantly reduces network traffic and CPU load that occurs when many miners connect to the server at once, while the delay incurred by the miners is often less than 1s.
• While some miners have complained about lower profits today, we have not found any evidence to support theories that bugs caused the decline. Today's reduction in profits has been caused by normal difficulty increases as scrypt slowly returns to its breakeven level of around 0.5 cents due to more miners coming on. This is just how mining works.
• We are taking a different approach to the connectivity issues. Since we've spent weeks trying to figure out the problem without complete success, we're going to stop devoting time to it. Instead, we will focus on eliminating other issues that cause problems and see if any of those issues was responsible for side effects that caused some of the network connectivity problems.
• Planning continues after Saturday's meeting that resulted in a tentative decision on the project's future. While we can't announce the decision just yet, we have outlined the ways in which service will improve no matter which way the decision goes.

[werewolf21]

why I do not work DGMS pool????

[joshn712]

Good morning!

• We issued a new release yesterday, as is customary on Sundays. This time, we modified the sign-in phase for miners so that when they connect the mining server, they will now wait to be issued work until the next work assignment happens for all miners, instead of immediately receiving work. This change significantly reduces network traffic and CPU load that occurs when many miners connect to the server at once, while the delay incurred by the miners is often less than 1s.
• While some miners have complained about lower profits today, we have not found any evidence to support theories that bugs caused the decline. Today's reduction in profits has been caused by normal difficulty increases as scrypt slowly returns to its breakeven level of around 0.5 cents due to more miners coming on. This is just how mining works.
• We are taking a different approach to the connectivity issues. Since we've spent weeks trying to figure out the problem without complete success, we're going to stop devoting time to it. Instead, we will focus on eliminating other issues that cause problems and see if any of those issues was responsible for side effects that caused some of the network connectivity problems.
• Planning continues after Saturday's meeting that resulted in a tentative decision on the project's future. While we can't announce the decision just yet, we have outlined the ways in which service will improve no matter which way the decision goes.

Our daily revenue went down 25% in 1 day, The difficulty did not go up 25% in 1 day... there is something else wrong.

[mmfiore]

To me it is clear that there is a payout problem. The pro hashing website is telling me that it is averaging over 7 cents per 1 MHz. I calculate that actual amount of profit I should make for that day. It turns out to be for the amount of hashing power I have to be around about 180 to 200 dollars a day. But when I calculate the rate I am actually earning at its about 125 to 145 a day. That is significantly below what the PH website is predicting. I then go to NH and mine for a day and they payout exactly the amount per day they predict which is in the range of 180 to 200 dollars a day. Something is siphoning the profits or they are being lost some how.

[Steve Sokolowski]

To me it is clear that there is a payout problem. The pro hashing website is telling me that it is averaging over 7 cents per 1 MHz. I calculate that actual amount of profit I should make for that day. It turns out to be for the amount of hashing power I have to be around about 180 to 200 dollars a day. But when I calculate the rate I am actually earning at its about 125 to 145 a day. That is significantly below what the PH website is predicting. I then go to NH and mine for a day and they payout exactly the amount per day they predict which is in the range of 180 to 200 dollars a day. Something is siphoning the profits or they are being lost some how.

What is your hashrate and how does it compare to what the manufacturer promised? Because we switch coins frequently, hashrate is often lower here than at pools which mine a single coin with long block times, like litecoins.

We're continuing to investigate issues as they come up, and once stability is achieved, we are going to write an automated test suite to prove that these numbers always exactly match hashrate. The only fact we can promise right now is that people are making the most the market can put out, since there's not huge amounts of profit showing up that is unaccounted for.

[mmfiore]

My has rate is between 2.2 to 2.4 GHS I have Titans. A question for you, if my miners are set to hash litecoin all day versus switching between coins will the rate of pps drop for me because I am mining a hard difficulty coin versus switching between easier to mine coins?

[mmfiore]

I have noticed that some days the payout is good, it's about right and unexplainable the next day it drops by 60 to 70 dollars.

[Steve Sokolowski]

I also wanted to add here that we think that there may also be WAMP data surges at disconnect. We only fixed the issue for connect. That means that it's possible that if a lot of miners disconnect at once, it could explain the large bursts of data. There is also unnecessary data being sent at miner disconnect, which we reduced.

Chris will deploy a change for that this afternoon to see if we can further reduce WAMP bandwidth usage and CPU load.

[mmfiore]

I still need to know how to best configure my miners. Is pps different when we mine a difficult coin versus an easy to mine coin? This may explain why there is profit swings in mining at pro hashing

[micca410evo]

@mmfiore, it shouldn't work like that. This is the formula of PH:
https://prohashing.com/help.html#payouts-computation




So difficulty adjusts your earnings.. So it doesn't care if you do cheap or high diff.. Just put the difficulty in that gives your miner the best performance.