Status as of Monday, December 1

[Steve Sokolowski]

I apologize for not providing an update in a few days due to the holiday. Here's what's new:

• It turns out that there is a race condition that is causing the profitability statistics to be underreported on the site and to PoolPicker. The problem is that when a coin switch occurs, sometimes the switch is recorded in the database before the difficulties and prices of the coins are updated. That means that, while miners are paid the correct amount, the out-of-order execution of these instructions results in the database recording the older (lower) profitability, which existed before whatever changed to cause it to make sense to mine the new coin. It looks like the underreporting is around 10%, but on some days where we reported 0.0001 or less, the profitability was too low by 50% or more. The problem is worse when non-litecoins are the most profitable, and I'm working on a fix for that issue now.
• It seems that improving scrypt profitability is the most important thing that we can do to create happy customers, so that's what we'll do. I'm looking at what we can do during the holiday break, when I'll have 9 days in addition to the three weekends to make significant progress forward. After resolving these concurrency issues, the next step will be to divert a proportion of hashrate to easy coins, followed by adding multiple exchanges.
• The holiday sale has ended, and thanks to those who participated. One of the major reasons for the sale was to determine what effect fees would have on hashrate. It seems that the fees have no effect on how many people use the pool, which is good, because a 1% fee results in losing money due to exchange fees and slippage and makes it unprofitable to operate. The profitability metrics went up after the fees returned to normal, because the value of altcoins rose over the weekend.
• Chris enabled 10 new coins over the weekend, which has increased profitability further. He is working on additional coins now.