Status as of Wednesday, September 17

[Steve Sokolowski]

Here is today's status:

• Chris received and set up the ASICs yesterday. He still has to work on the dispute case with the seller on eBay. After setting them up, he was quickly able to determine that our pool is not responsible for the problem with these miners. Instead, it turns out that the difficulty of the shares requested from the miner is responsible. For some reason, the miner performs best submitting shares of difficulty 2048, not 16384, even though there is no reason the miner should need to submit shares of such a low difficulty.
• After finding this out, we determined we can improve the performance of Zeus miners by 25% by allowing those miners to mine at 2048 difficulty. For now, I'm going to ask Chris to set the maximum difficulty of the pool to 2048, which does not require a restart. He'll do this during the afternoon, and that will improve performance. In the future, we may allow people to override the difficulty, or we may just leave the maximum difficulty this low, but this solution will work for now.
• I completed implementation of the mining server part of the multiple exchanges, and now I'm testing it. My goal by this weekend is to be able to release a new version of the mining server that does exactly what it does now with Cryptsy, but using the new multiple-exchange architecture. Then, we can take a break from that and fix a few of these other issues that came up and perform the much simpler task of implementing actual Poloniex trading after that.
• I got jimlite paid out by giving him an extra dollar. After talking with Chris, we decided on a compromise that will satisfy both jimlite and Kristof: the addition of a "close account" button. When clicked, the user is charged a small fee of about five cents to cover transaction costs, and then the entire balance is paid out during the next payout interval (which is currently at 2:00am), even though it doesn't meet the payout threshold. This will ensure that we hold less money from inactive miners, will provide people with assurance they won't have small balances left over, and will satisfy Kristof by not reducing the payout threshold too much.
• Unless someone points out something, I'm going to have Chris increase the maximum number of users to 75, and begin significantly ramping up. There isn't a reason anymore not to start making money, as the site seems to be stable and there will always be a low number of issues outstanding.
• The Middlecoin pool disappeared recently. They said that BTC-e shut down their account, and they lost 58 bitcoins. Instead of paying out, they decided to take the unethical route and run off with the money. These guys made $3m over the past year, and people like them give mining pools a bad name by not honoring their commitments. I hope that someone sues him and recovers damages; if I had any money there, I would be willing to pay a lawyer nearly all of my recovery to make that point.
• There is a 0.18 bitcoin bounty for being the first pool to implement litecoindark payouts. Chris has been talking with the litecoindark developer to implement payouts in litecoindark and to claim the bounty, but the developer requires a newer version of gcc than Debian supports. He's going to continue to talk with him, but we probably won't be able to claim this bounty and gain the publicity associated with it because we cannot compile the daemon.

[cryptorific]

• Alot of pools still operate on much lower difficulties, wemineltc's asic only pool has a fixed difficulty of 1000. Intuitively it seems like a waste of bandwidth but perhaps they're just too comfortable with the status quo.
• Let me know what you think is the best route for getting more miners? some more reddit posts?
• I noticed middlecoin went down but I'm not sure I buy their explanation. They've been sucking for a long time, they were at their peak when doge peaked above 200 satoshi's and after a series of delayed payouts and poor earnings (they were only mining the handful of coins on btc-e) their hashrate crashed. Considering they had up to 30k+ users and were running the site off ec2, their server costs probably were pretty big and probably have been operating at a significant loss for a while. Unfortunately 58 btc isn't really enough to fund a law suit over, not to mention identifying the interested plaintiffs would be nearly impossible

[Steve Sokolowski]

It really is a shame that these people can get away with 58 bitcoins and nobody caring about it. If he was losing money, then he should have shut down. Why would someone operate a business that is declining and has started to lose money? He already made a fortune during the last bubble, so if he managed his wealth appropriately, he shouldn't even need to work now anyway.

If you'd like to make more reddit posts, then that's great. At this point, we can handle a lot more people, so we don't have to hide the pool like we did before. Chris will set the maximum number of registrations to some high number just so that the server's CPU usage doesn't get overloaded, but it's unlikely that we could get to that point. It will be interesting to see how many people join in.

Chris changed the maximum difficulty to 2048. While originally we had thought about allowing the user to override that and set a higher level, I'm not sure that's worthwhile. It just makes things more difficult for the user to remember, we have to spend time writing that code, and the mining server is not a significant usage of resources compared to the website, database, and other parts of the system. For your miners, it's the difference between 60 bytes per minute and 400 bytes per minute, which doesn't matter.