pool mechanics and ethermine artificial rates

jetpackjetpack Member Posts: 88
Idk if you noticed but say your running 80mhs and you get xxx shares a hour and your doing 120mhs effective. Then it almost strategically hits 70mhs after that effective. Do they actually program and control that to make it more fair for everyone? It seems too systematic. Say luckily you have a 47ms or 62ms ping/connection so your in the win vs others... thus you can get more share submitted faster vs others. But it seems if your doing 120mhs with only 80 mhs they are diluting your power to make the system balance for others? And then some how at the end of the day your "actual" effective is the same as your actual. I was just wondering your thoughts or knowledge on that. Just seems odd

Comments

  • Ericjh801Ericjh801 Utah, USAMember Posts: 361 ✭✭
    I'd say at the lower hash rates there's a bit of luck involved. You are probably going to see an average of your hashrate over a long period of time but day to day it could flux a lot. There is no system in place to give you less/more shares per hour, all based on you solving.
  • cidmocidmo Member Posts: 444 ✭✭✭
    edited February 13
    because of the VARDIFF its really hard to tell what any pool is doing
    unless ur tracking each share at whatever difficulty it was submitted
    this is also one reason why i will never trust pools
    its just way to easy for them to fudge small numbers without anyone ever knowing
    and in terms of a pool like ethermine with 100000s of workers
    its like office space
  • jetpackjetpack Member Posts: 88
    I was just wondering what you thought as its statically hits 120mhs then drops to 70mhs as if its dummbed down to that rate for a purpose.
  • Ericjh801Ericjh801 Utah, USAMember Posts: 361 ✭✭
    Yeah it's not on purpose. Luck of the draw for the most part. Usually just evens out at the end.
  • MadKernelMadKernel Member Posts: 49
    edited February 14
    Some random considerations about mining , pool, miner, the network ...

    FACTORS THAT CONTRIBUTE TO LOWERING THE REAL HASHRATE VS HASHRATE REPORTED BY THE MNER
    • NEW BLOCK PROPAGATION SPEED ON THE ETHEREUM NET(can vary from 0.2 to 5 secs!) https://ethstats.net/
      (Here we can do no thing)
    • POOL LATENCY TO DISPATCH NEW WORK (??? > 100 ms)
      Depends of the bounty of implementation and the load of the pool .
      I suggest also to control the long term 'luck' of the pool.
      if >100% probably is because well implemented.
      Why some pools doesn't public these statistics?
    • NETWORK LATENCY TO AND FROM THE MINER (ping time)
    • - LATENCY OF THE MINER TO ABORT CURRENT WORK AND INITIATE THE NEW (flush of the work queues, depends on the size of the CL queue GROUP_SIZE, WORK_SIZE, on my card I measure it ~47 ms).
    Since these thing happens once a new node is mined (about 14.5 secs) the real time dedicated by the miner to mine the current block may be in the best scenarios:

    14.5-0.2-0.1-0.05-0.05 = 14.1
    14.1 / 14.5 = 97 %

    With a phisiologic numbers of almost 3% hash stale reported by the pools, for the worst done yourself the calculation.

    For masking what shit ETH network is, the pools started to accept as valid also stale share (valid shares of the penultimate block).

    So here was listed and analyzed (may be wrongly) the factors that contribute to lowering the miner theorical hashrate vs pool reported one.
Sign In or Register to comment.