Prohashing Mining Pool Forums

Here's the status after the weekend.

In short, we completed testing and validated all of the features that are listed below, from last week. These work and are ready for release.

• The ability to ignore Coinwarz's data for specific coins, since their data is not always correct.
• The backend infrastructure for multiple exchange support, although multiple exchanges won't actually be enabled.
• An option to pay coins below the payout threshold, at the cost of the transaction fee.
• The ability to hide usernames in the "found blocks" table, at jimlite's request.
• A more accurate BTC per day reading on the top bar of the site; the current reading takes the dollar earnings and converts them to BTC, which makes the pool appear to be less profitable on days that BTC value increases and the other way around.
• An option to reduce the number of blocks required until mined blocks reach maturity on a per-coin basis, to reduce slippage so that mined coins can be sold faster than 120 blocks when possible.
• A new 19th error condition that disables a coin when the last 30 blocks average twice as profitable than the second most profitable coin's last 30 blocks, as another way of detecting forks.

However, we decided to hold the release because jimlite pointed out that dogecoin merge mining can be performed with any coin. While we are already close to #1 all the time, dogecoin merge mining should return us to being #1 by 20% again, and it explains the declining profitability recently.

I was successfully able to put the merge mining information into the primary coin on Sunday, but I expect that getting the dogecoin blocks to work will be the most difficult task that we have accomplished so far, because we need to find a way to test the code (and their testnet doesn't work). To speed this up, I'm going to try to push my work on this project up to 40 hours per week by eliminating daily "bitcoin thoughts" posts and paying people to do chores for me. Imagine how fast this would go if I could spend 85 hours on this project, instead of having to spend the other 45 hours at another job working and commuting. Merge mining is a huge pain that has significantly delayed our plans, but I won't let it defeat us.

Adding dogecoin support needs to be done the right way, because I suspect that other coins are going to start doing this merge mining, and we need to be able to support multiple merge mining coins so that the Wafflepool doesn't gain an advantage by doing that first. This also involves reworking how coins are credited to users to a concept of "inputs and outputs" - for example, a share could be mining grandcoins and dogecoins, while requesting payouts in netcoins and litecoins. Previously, there was only one "input" coin.

I'm going to be quiet around here for the next few days so I can be as efficient as possible. In the meantime, Chris is going to work on dealing with incorporating the business and performing marketing while he has no testing to do. Thanks to jim for his hard work with pointing out these suggestions!

Sounds like you are very devoted and will have a winning pool behind you. I'm going to have my miners on you off/on, and you can always find me in IRC rooms, #bittrex I am almost always there. A thought to make things more profitable quicker, I don't know if this is easier, but why not just pay out the merged doge in doge only, until you can get the two coins to convert to whatever is requested. I had my account set up to split between btc and doge anyways, so maybe at first just send the doge to our doge addresses? I haven't coded in 25 years, so this may or may not be an easier temp. solution.

Sorry I've been MIA for a bit, between school and my upcoming wedding my time has been stretched. I will be going to http://launch.coinsinthekingdom.com/ bitcoin conference this weekend and wanted to help advertise the site. Any ideas? Print up a few tshirts? flyers? I wish gridseed still sold those little 50khs usb asics to give away.

If you have a suggestion for something you would like me to buy that you can use for PR, I would be willing to pay for it. Looking at the conference schedule, I interpret that it is mostly focused around speakers as opposed to some of the other conferences that have large exhibition areas. My interpretation is that most of the publicity would be from you just talking with people as opposed to them coming across a display, so I would think that giving the people you come across something to remember the site would be a good idea. Do you have suggestions for what could be given away and also can be prepared in three days?

Basically tshirts or business cards are pretty much it on such short notice. I dunno why I only now just heard about it, I would have prepared something more substantive if I had known but I agree a exhibitors table would probably be pointless and unnecessary.

As far as merged mining goes, can't PesetaCoin, UnitedScryptCoin, OrgCoin, HunterCoin and Syscoin be merged as well?

The problem with merge mining multiple coins is that we can't enable a coin unless it is on Cryptsy. We have the infrastructure for multiple exchanges tested, but implementing the buying and selling on those exchanges still needs to be done. We can't do anything with coins that aren't listed on the exchanges we support, so there isn't any reason to mine such coins.

I think that, for this release, we'll just enable dogecoin mining. Then, we'll finish Poloniex support, which will increase profitability by adding more coins. Once we do that, we can evaluate which coins are supported on one or the other exchange and add merge mining for more coins. Otherwise, we will have too much feature creep. This release is going to require a full shutdown of the pool for up to 2 hours to make the 47 database changes as it is.

My guess is that the features that will get the most users are (in this order):

-Merge mining with dogecoins
-Increasing marketing
-Multiple exchanges so we can mine more coins (infrastructure already tested, but none other than Cryptsy implemented)
-Figuring out what the Kimoto Gravity Well is so we can mine those coins
-Mining many small coins at the same time by distributing different work to everyone (infrastructure will go into this release, but will not be tested in order to speed things up, so we can't enable it yet)
-Creating the algorithm-switching custom client
-Merge mining with other coins


Do you agree with the order?

You will find that for new coins, bittrex and c-cex are the first ones to list. Poloniex and Cryptsy take their time. Much profit is to be made with new launches.

Unfortunately, those two exchanges don't have "push" APIs. We can only work with exchanges that push real-time data to subscribers.

The alternative to a "push" API is a "polling" API, where the mining server has to call a webpage every few seconds and parse it to discard all the data it doesn't need, a CPU-intensive process that wastes huge amounts of bandwidth. More importantly, this data is already five seconds out of date by the time we get it.

I was thinking of putting out an "open request" to one of the big discussion forums asking for exchanges to prioritize a push-based API for the reasons that you mentioned.