ethermine.org - The fastest way to mine ether

191012141531

Comments

  • MajorTorleyMajorTorley Member Posts: 29
    Fixed-NotFixed?

    Yesterday in the 12 hours after dr_pra announced blocks were fixed to all be paying out, eth payment to me reached 90% of expected, a big increase and only 10% short- possibly within pool luck on a off day?
    Today over the last 8 hours payout has reached exactly 100% of expected, Whew. If that holds a statistical while more, or stays anywhere close i'd say things are FIXED for sure. Looking forward to when the 25-30% unpaid blocks from last week are released to us.
  • the_doritothe_dorito Member Posts: 16
    "Looking forward to when the 25-30% unpaid blocks from last week are released to us"

    Couldn't agree more.
  • newkidONdablocknewkidONdablock interwebzMember Posts: 121
    @MajorTorley you think they will reimburse us for them?
  • MajorTorleyMajorTorley Member Posts: 29

    @MajorTorley you think they will reimburse us for them?

    Yes. Because dr_pra said so: "The issue has been identified. Due too a problem in the rounds calculation module certain blocks were not credited. We were able to identify the problematic rounds and will start to reimburse them as soon as possible. "

    Thats great for us because surely they know when the code was changed that messed up the rounds calculation module, and they can find all those blocks that weren't paid and pay them. Easy for me to say and guess though as I dont know their servers or methodology. It could turn out to be really really hard for them to sort.

    Its good that they havent payed it out to us yet because that is giving us time to see that payouts are correct again, which mixing in old payments with current ones would make near impossible to tell.
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    Yes not paid blocks will for sure be paid out. We are still figuring out the best way on how to distribute them. Most likely we will double pay the block closest to the unpaid block to reward every miner according to their hashrate at during the time the block was found.
  • newkidONdablocknewkidONdablock interwebzMember Posts: 121
    @dr_pra
    i'm impressed man, sorry for the misunderstanding, there was no ill-will
  • myminermyminer Member Posts: 8
    Great Pool! Very easy and stable.. also design wise looks great!

    Mining with this : https://forum.ethereum.org/discussion/7106/mining-rig#lates:smile:

    feel free to pitch in and help out in the thread:):smile:
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    The value of stale shares had been adjusted to 65% of normal shares. This was done based on an analysis of the average uncle reward the pool receives.
  • JPSJPS Member Posts: 63
    edited May 2016
    ... I guess it's a coincidence that suddenly NOW I get stales/invalids? :|
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    No, you were submitting them before as well, we just have added the respective statistics and show them now.
  • newkidONdablocknewkidONdablock interwebzMember Posts: 121
    @dr_pra
    is it normal that im sending stales almost as much as valids? I have 90-100MH/s and it show im mining @ 66-70 :S
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    No, you should not have more than 5% of stales. Over the whole pool ~3% of all shares submitted are stale. You need to check your setup, maybe you are using longer kernel run times or have a bad internet connection.
  • deliciousminingdeliciousmining Member Posts: 49
    I had 40% invalids on 16.5.1 driver. 16.2.1 0%. I tried so hard to make this new driver work but there is no hope beyond 16.2.1
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @dr_pra Given that many on the pool setup guides and other mining guides and forum advice suggest huge global/local work sizes, without regard for the hash rate of the GPU, it's not surprising that some miners see larger stale share percentages.
  • BiodomBiodom Member Posts: 693 ✭✭✭
    edited May 2016
    few observations:

    1. setting stale % at 65%, which is average what pool receives is reasonable.
    2. did something else changed? I am getting less invalids (almost none), but speed is also very slightly slower (maybe by 1-1.5%)
    3. from my perspective this is much better because before people were jamming their stale shares and getting 100% while pool and the rest of people were getting paid LESS (if pool only gets 65% for stales).
    4. someone (@adaseb?) was actually suggesting that pool was attacked by stale shares resulting in less payment. Although owner concentrated of some other module, I am sure from current development that payment was also decreased because of excessive stale share. In most cases it was probably due to not knowing that it is happening, but there could have been bad actors as well-we would not know, unless @dr_pra tells us.

    @dlehenky
    re guides-some numbers there are ridiculous. I tried before some at 2 times less than suggested and GPU/screen almost froze and invalids increased.
  • newkidONdablocknewkidONdablock interwebzMember Posts: 121
    @dr_pra
    I still think there are some problems or is this working as intended, before all this miss-calculation happened the bottom line in the personal stats page showed shares like a lot 1500-2000 daily and then it resets probably daily but now i didnt see a number higher than 100, it varies like 20mins before i had 84 and now i have 76 and 4 stale and after that 0 stale and 40 and my miner is working the whole time without disconnect or anything. Care to clarify?
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @newkidONdablock I believe the new stats are the most recent 60 minutes at the point you look at them. Don't quote me, though. :)
  • newkidONdablocknewkidONdablock interwebzMember Posts: 121
    @dlehenky
    it appears so, you might be on to something
  • MajorTorleyMajorTorley Member Posts: 29
    Monday payments were accurate within margin of error. Today after the new stale adjustments payment is 12-15% low. Page says I have sent 4 stales in that time.
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    Low compared to what? Blocks found by the pool during the last 24h match what is expected given the pools hashrate.
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    @dlehenky yes, valid/stale/invalid statistics are from the last hour. Previously they were reset every time the worker reconnected to the pool. It is much better comparable this way.
  • ethfanethfan Member Posts: 458 ✭✭✭
    dr_pra said:

    @dlehenky yes, valid/stale/invalid statistics are from the last hour. Previously they were reset every time the worker reconnected to the pool. It is much better comparable this way.

    Why not a rolling 24h stales/invalids like the Effective Average Hashrate?
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    This will come soon. If you want to have an idea how it will look like check out the ethpool.org thread on the forum.
  • MajorTorleyMajorTorley Member Posts: 29
    edited May 2016
    dr_pra said:
    Heck Yea. That would pretty much be the most informative useful bunch of information that any pool that I know of offers. All I can think of that I would really really like with that is the ability to see a day or two of that hour info somewhere, instead of only having access to the most recent hour. Dwarf does a nice job of doing that.

    Any ETA on the unpaid blocks getting paid? How many days worth were there? I think I noticed 5 or so? When they get paid will the stales in them be paid at the rate they were then, or the new lower one?
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    Missing rounds will be paid starting on Thursday in small batches in order not to overload our payment system and to monitor the process. ~185 blocks/uncles are due to be paid. We will use the historic hashrate distribution as if the blocks were paid at that time.
  • po_etpo_et Member Posts: 4
    The stats are strange. I have checket one of top miner

    Time: 08:37:00pm
    0/3/49/0%
    0/2/49/0%
    0/1/34/0%

    refresh browser
    Time: 08:31:17pm
    0/3/49/0%
    0/2/48/0% ??? it's not typo
    0/1/32/0% ??? it's not typo

    refresh browser
    Time: 08:31:32pm
    0/3/49/0%
    0/2/48/0%
    0/1/33/0%

    I think you should add the stale/share and not subtract.
    It is very mysterious.
  • dr_pradr_pra Member Posts: 445 ✭✭✭
    @po_et it is a sliding window (from now to now - 60 minutes). It is normal that the values fluctuate a bit up and down as share submission is subjected to luck.
  • po_etpo_et Member Posts: 4
    OK, understood, that makes sense.
  • MajorTorleyMajorTorley Member Posts: 29
    dr_pra said:

    Low compared to what? Blocks found by the pool during the last 24h match what is expected given the pools hashrate.

    Low payment compared to estimated payment. All day today i've gotten steady .075/hr average from the pool, but with 195 mh/s at current difficulty. Math and the top four eth calculators all say it should be .0859 (cryptoyeti), .086 (etherscan), .0833 (coinwarz), or (.0807 ethermine). This is way way way better than when random blocks weren't being paid but ~12% low still (between 15% and 7 1/2%).

    When the ethermine earning estimates on the status page was tweaked to give lower estimates Saturday, what was that based on? It now gives lower estimates than any working calculator out there.
  • MajorTorleyMajorTorley Member Posts: 29
    dr_pra said:

    ~185 blocks/uncles are due to be paid.

    We are completely at your mercy since only you can tell how many werent paid, but could you possibly please look at them again? That many comes to about 25% of 1 day worth of mining?, which in my specific case is about half a eth. I know I was not getting paid a even greater percentage than that, generally 35% missing, and over at least 4 but probably 6 or 7 days not the equivalent of one. My tiny rig was underpaid by at least 3 eth that I knew about and probably some I didn't.

    With multiple people mentioning 25-30%+ underpays across at least 3 days, 185 blocks has to be way way off, or there was another unfound problem siphoning our coins.
Sign In or Register to comment.