Status as of Monday, May 11

Discussion of development releases of Prohashing / Requests for features
Forum rules
The Development forum is for discussion of development releases of Prohashing and for feedback on the site, requests for features, etc.

While we can't promise we will be able to implement every feature request, we will give them each due consideration and do our best with the resources and staffing we have available.

For the full list of PROHASHING forums rules, please visit https://prohashing.com/help/prohashing- ... rms-forums.
Post Reply
User avatar
Steve Sokolowski
Posts: 4585
Joined: Wed Aug 27, 2014 3:27 pm
Location: State College, PA

Status as of Monday, May 11

Post by Steve Sokolowski » Mon May 11, 2015 9:24 am

I completed everything I wanted to get done this weekend and more. Here's what was accomplished. All of the features will be rolled out once Chris completes his testing.
  • Multiple merge mining has been completed. We are aware of seven scrypt coins that will be merge mined beginning later this week. Chris believes that there are 3 x11 coins that can be merge mined when that algorithm is implemented.
  • Comkort was finished and verified. Chris deployed the trading code for Comkort, but it is unknown if the mining server is selecting Comkort coins yet. If not, he'll release that when he gets up.
  • There are two bugs that were discovered that explained some accounting issues. First, shares for found dogecoin blocks were not being credited to their finders, so that fix will increase profitability by about 1/1 million or whatever small percentage that is. Second and more importantly, dogecoin blocks were lost when the primary coin had more than one transaction in its block. That issue will also be fixed with the next release.
  • There is a new password argument, "r=," which allows miners to manually declare a work restart interval. "r=0.23," for example, informs the mining server that the work restart interval for the miner is 0.23s, and skips work restart testing. The purpose of this flag is to address a customer complaint where cloud mining services do not allow miners to be assigned the low difficulties used for work restart testing. Note that other customers who are not having a problem should not use this flag; it is very difficult for a human to determine the optimal value and you will probably cause yourself to lose money by setting it manually.
  • A bug is present in the current release that can cause exchanges to permanently lose connectivity after a single disconnection to the exchange occurs. The problem has been resolved in the next release. Bittrex, which often returns 500 errors downloading its price data, was susceptible to this problem.
  • We introduced the concept of a "sell-only" exchange, but did not actually implement any sell-only exchanges yet. Examples of such exchanges are ccex and bter - both of which have profitable coins for mining but which do not allow automatic withdrawals. It isn't feasible for Chris to manually withdrawal the volume of coins that customers request, so we will not be able to buy or pay out in coins solely available on these exchanges.
We anticipate releasing all of these changes Tuesday evening. A restart of the database server will be necessary to apply an unrelated patch, resulting in 10m of downtime.

After that's done, here is the plan for the following few weekends:
  • May 16-17: Resolve issues introduced by this release, implement ccex and bter as "sell-only" exchanges, and implement "w=" worker name password argument (but there will not yet be any way to display this data)
  • May 23-25: First weekend off of the year
  • May 30-31: Weekend dedicated to customer requests of non-profitability features, like display of worker data and some API calls
  • First two weekends in June: Implement support for "a=" password argument to specify algorithm to mine (default will remain scrypt)
  • Before end of July: Redesign website to support the data required for multiple algorithms, using websockets for real-time display
We still have not yet decided how the website will be redesigned (suggestions will be needed for that), nor whether we should go immediately to automatic algorithm switching or just support static algorithms with "a=" first. Chris says that there are rumors that x11 will soon be taken over by ASICs, so perhaps we should just support x11 as a static algorithm first. The difficult part about algorithm switching is that we will need to propose extensions to the stratum protocol.
Post Reply