Ethereum Mining Pool Hub - 0% Fee, pays all kind of mining rewards, supports all miner programs

1568101122

Comments

  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    shutfu said:

    what is difference between porns 20535 and 20536 when connecting with claymore (only uses stratum, readme says to use 20536) or with genoil in -S mode?

    @shutfu

    20535 is compatible with genoil's stratum mode (-S) and eth-stratum-mining-proxy (made by coinotron).
    20536 is compatible with eth-proxy (made by dwarfpool)

    Claymore supports both, but you need to set appropriate option for each port.
    shutfu said:

    also @Delle54 , your failover server in your commandline is wrong, it is missing the 1 in "us-east1", it should read "us-east1.ethereum.miningpoolhub.com:20535"

    Actually us-east is same as us-east1 currently. I added "us-east" dns host as well because many miners would forget to add "1" at the end.
    The original idea of "us-east" was to distribute us-east1, us-east2, us-east3 servers from us-east host automatically (load balancing). There's no us-east2, us-east3 servers currently, but someday us-east server would behave a bit differently.

    Thanks.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    bitcanuck said:

    shutfu said:

    what is difference between porns 20535 and 20536 when connecting with claymore (only uses stratum, readme says to use 20536) or with genoil in -S mode?

    Despite what @miningpoolhub says, claymore doesn't really support both. At least up to v4.2b only port 20536 with -esm 1 works without errors. If you try the other ports and stratum modes you'll get json errors or it will just fail to connect your worker.
    @bitcanuck

    Ah.. actually I haven't tested claymore with 20535 yet. I only heard that it supports both proxy mode.
    Thank you for correct info.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    bitcanuck said:

    Any ETA for the stale share stats on the dashboard?

    I'm still working on it. Seems like it will be finished within this weekdays.
  • GenoilGenoil 0xeb9310b185455f863f526dab3d245809f6854b4dMember Posts: 769 ✭✭✭
    MPH has become my favorite testing pool because of the vardiff and high number of shares within a short timeframe. Testing my 1.1 now. I notice that unlike other pools, calculated hashrate based on submitted shares shows a much higher hashrate (about 10%) than what the miner reports. Is this normal, due to vardiff? I would expect that over time it would even out to the 'real'rate as reported by the miner, but it doesn't.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    @Genoil @bitcanuck

    This is not related to vardiff.
    As bitcanuck already stated, MiningPoolHub adjusts reported hashrate to match the numbers they see from client. Currently pool bumps about 12% from collected shares.

    Things have changed from early days.

    (1) Miner program changed.
    From not optimized settings with 280x gpu, same catalyst version, same OS.
    latest v1.0.7 is showing 12~14 Mh/s
    but v1.0.4b3 is showing 16~18Mh/s

    Same epoch DAG, but ethminer's hashrate level is quite different from my test pc.
    Can you tell me why this difference is made?


    MiningPoolHub adjustment matched for 1.0.4b3 but someday this hashrate changed and this is the main reason why you see incorrect hashrate level from pool.


    (2) Stale share counting is applied.
    It's not that big, but this made some shares wasted before are now counted. This results higher hashrate, higher block finding percent from pool.
    This improvement was made from latest ethminer-genoil and pool both. This change would increased hashrate few.


    I think it's good time to adjust the error correction value as many people updated to latest ethminer.
    I just want to know the reason of this hashrate changes between ethminer version.
    Will this happen later too?
  • GenoilGenoil 0xeb9310b185455f863f526dab3d245809f6854b4dMember Posts: 769 ✭✭✭
    Thanks @miningpoolhub for the explanation. I changed main development GPUs between those versions, so I'll have to go back and compare what is shown. For now, something seems wrong with 1.0.7 and your 280X. My 7950 is much faster :)
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    edited May 2016
    Genoil said:

    Thanks @miningpoolhub for the explanation. I changed main development GPUs between those versions, so I'll have to go back and compare what is shown. For now, something seems wrong with 1.0.7 and your 280X. My 7950 is much faster :)

    @Genoil

    My 280x is from xfx (mining performance not good), undervolted, optimize settings not specified. This is not intended for mining but to test pool from my development pc.
  • GenoilGenoil 0xeb9310b185455f863f526dab3d245809f6854b4dMember Posts: 769 ✭✭✭
    just tested 104b3 and 107 on the 7950. Both give the same hashrate output (about 16.3MH/s). From version 108 onwards the default display has changed, so it looks to be a little less. But the kernel itself hasn't changed a bit, so it is the same.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    Genoil said:

    But the kernel itself hasn't changed a bit, so it is the same.

    @Genoil

    I'm curious that why my 280x hashrate varies much between ethminer versions. All other environments are same, only ethminer binary changed.
    Even the command line is same.

    Any idea about it? Am I the only one experiencing this weird result?
  • GenoilGenoil 0xeb9310b185455f863f526dab3d245809f6854b4dMember Posts: 769 ✭✭✭
    Well coincidentally I'm talking to someone with 1 out of many identical rigs that shows similar behaviour. Weird.
  • Delle54Delle54 Hannover, GermanyMember Posts: 16
    edited May 2016
    Since yesterday, I sometimes get a message like "unauthorized worker":After I restart genoil 1.0.8, it runs normal for an undefined period of time. What could be the cause?
    Post edited by Delle54 on
  • ImAMiner?ImAMiner? Member Posts: 208 ✭✭
    edited May 2016

    bitcanuck said:

    shutfu said:

    what is difference between porns 20535 and 20536 when connecting with claymore (only uses stratum, readme says to use 20536) or with genoil in -S mode?

    Despite what @miningpoolhub says, claymore doesn't really support both. At least up to v4.2b only port 20536 with -esm 1 works without errors. If you try the other ports and stratum modes you'll get json errors or it will just fail to connect your worker.
    @bitcanuck

    Ah.. actually I haven't tested claymore with 20535 yet. I only heard that it supports both proxy mode.
    Thank you for correct info.
    I run claymore on port 20535 with no errors...

    With 4.3 now but I could've sworn I ran it before 4.2b.
  • ImAMiner?ImAMiner? Member Posts: 208 ✭✭
    bitcanuck said:

    @ImAMiner? I just stopped genoil-ethminer to test 4.3b again. I get errors with all stratum modes (-esm 0, 1, & 2). Here's the error I get with esm 0 & 1. With esm 2 I get auth errors.

    Weird, here's what I use

    EthDcrMiner64.exe -wd 1 -r 0 -allpools 1 -estale 1 -ftime 5 -mport 3333 -epool us-east1.ethereum.miningpoolhub.com:20535 -esm 2 -ewal Login.Worker -epsw 1234 -eworker Login.Worker -etha 0 -ethi 16 -dpool stratum+tcp://dcr.coinmine.pl:2222 -dwal Login.Worker -dpsw 1234 -dcri 40

    Relevant https://bitcointalk.org/index.php?topic=1433925.msg14829094#msg14829094
  • workwork Member Posts: 2,084 ✭✭✭✭
    @bitcanuck I use qtminer mode at the qtminer port and it works fine.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    Genoil said:

    Well coincidentally I'm talking to someone with 1 out of many identical rigs that shows similar behaviour. Weird.

    @Genoil

    Maybe I would have to relive undervolt things and test with optimized settings.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    bitcanuck said:


    Any idea about it? Am I the only one experiencing this weird result?

    Seems like you're chasing a red herring. What matters is how the current version of ethminer performs. Nevertheless, I also tried 104b3, and got almost identical hashrates compared to 1.1: 20.93 vs 20.95 on Win7/64 with Catalyst 15.7.1 drivers. I used the same command line as well (except for removing -FS since 104b3 didn't support failover).
    @bitcanuck

    Well, I just wanted to know my issue because I use this gpu to test and set some criteria values like starting diff level. I understand that this is not important problem. But now I can't rely on my GPU only but have to collect some real miner's opinions.

    Back to the point.
    Originally, I assumed that hashrate performance was same between ethminer versions internally, but some time averaging (like 15 seconds) method was added on display. Now, it doesn't look like what I assumed.

    The problem is, many miners reported that they see lower hashrate from pool compared to their client numbers at early days. This is why I inserted error correction value at that time.

    Maybe adding stale share counting fixed this problem. Because this makes some hashrate increase.

    Did your hashrate got 10% up than real hashrate before the stale share counting? stale share counting was added around 16th May. Or you experienced constant 10% up than real hashrate from the beginning.


    Anyway, thank you for your support and feedback. Your comment helps much.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    Delle54 said:

    Since yesterday, I sometimes get a message like "unauthorized worker":After I restart genoil 1.0.8, it runs normal for an undefined period of time. What could be the cause?

    @Delle54

    Not sure why this problem happens.

    "Unauthorized worker" message will be shown when miner connects and submits shares before authorization is succeeded. It can happen at startup by timing issue but have no idea why it happens during the successful mining.

    (1) Did it have some network issue before "Unauthorized worker" and tried reconnection?
    (2) Does unauthorized worker state stays forever and not recovered automatically? Or Do you have to always restart the miner to work?
  • TermieTermie Member Posts: 130
    edited May 2016
    >>> ETH: Stratum - socket send failed 10053, disconnect

    yesterday evening I restarted mining at MHP again. Since then I´d noticed several disconnects, or server switches by miner, e.g. from europe1 to us-east1 and back. I´m mining with Claymore's 4.2 with option -esm 0 and using MHP port 20536. Is this combination the right one? Or is it anyway better/more stable to switch to port 20535 and -esm 1?

    Can I also use HOST: hub.miningpoolhub.com which is not listed at your ETH configuration help page, but at your "Getting started" page? Where is this server's location?

    Next question: right now I can see more detailed round share stats for the first time:
    Your Valid - 18,388.5397
    294.2773 - (1 level stale 1.55%)
    Your Invalid - 302.0521 (1.59%)

    Until now I always had 0% invalid shares, now I have 1.59% - is this rate too high or within normal range?
    Post edited by Termie on
  • Delle54Delle54 Hannover, GermanyMember Posts: 16
    @miningpoolhub
    (1) Yes, I had a network issue.
    (2) I had to restart manually.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    Termie said:

    >>> ETH: Stratum - socket send failed 10053, disconnect

    yesterday evening I restarted mining at MHP again. Since then I´d noticed several disconnects, or server switches by miner, e.g. from europe1 to us-east1 and back. I´m mining with Claymore's 4.2 with option -esm 0 and using MHP port 20536. Is this combination the right one? Or is it anyway better/more stable to switch to port 20535 and -esm 1?

    Can I also use HOST: hub.miningpoolhub.com which is not listed at your ETH configuration help page, but at your "Getting started" page? Where is this server's location?

    Next question: right now I can see more detailed round share stats for the first time:
    Your Valid - 18,388.5397
    294.2773 - (1 level stale 1.55%)
    Your Invalid - 302.0521 (1.59%)

    Until now I always had 0% invalid shares, now I have 1.59% - is this rate too high or within normal range?

    @Termie

    Pool restarted about 5~7 hours ago to collect and show the stale share statistics on webpage. Seems like this restart caused some disconnects.

    The two 20535 and 20536 ports are all stable. You can choose any port you like.
    I prefer 20535 since this protocol is compatible with stratum. Pool can switch mining coins easily. This is not useful feature these days, but will be useful when other coin's profitability rise up. Auto switching, auto exchange feature is the main strong point of miningpoolhub.


    hub.miningpoolhub.com is also possible but I do not recommend you to try it. It works as ethereum backup server located in US virginia.
    I hope you to use ethereum dedicated server near your location.

    In theory, invalid shares must be 0% in normal case with latest genoil-ethminer. Since pool collects all late shares upto 6 level as valid, there should be no invalid shares in statistics.
    Well, it's just theory. There's some probability that would cause invalid shares rise.
    Please tell me if you are experiencing issue even you are using latest genoil-ethminer (1.0.8 or higher). I will analyze your shares.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    Delle54 said:

    @miningpoolhub
    (1) Yes, I had a network issue.
    (2) I had to restart manually.

    @Delle54

    Seems like that miner didn't request authorization after network is disconnected.

    As I can't wait ethminer to be fixed in near future, I changed pool's strict policy. Now we allow submitting shares before authorization. There should be no similar issue from now on.

    Thank you for reporting this issue.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    bitcanuck said:



    Did your hashrate got 10% up than real hashrate before the stale share counting? stale share counting was added around 16th May. Or you experienced constant 10% up than real hashrate from the beginning.

    Yes, it was about a month ago that I mentioned it, then you admitted the numbers were grossed up.
    I think it was after I started using Genoil's ethminer in direct stratum connection mode instead of the official ethminer + eth-proxy. I doubt there are any serious miners using eth-proxy anymore. I think you should even update the pool info to state that only direct stratum connections (i.e. claymore, qtminer, or genoil) are recommended, and remove the information on how to setup eth-proxy.
    @bitcanuck

    Actually, there are many miners using eth-proxy still. They use eth-proxy to make failover to dwarf.

    Also, as you already know, there's reconnect timing bug in genoil-ethminer which makes miner program hung when network disconnects. This makes me hesitate to recommend this miner currently.
  • miningpoolhubminingpoolhub Member Posts: 308 ✭✭
    edited May 2016
    [Ethereum Mining Pool Hub Notice]

    - Added stale share statistics at Dashboard page.
    - Adjusted hashrate error correction value. (Other statistics may be affected little bit)
  • workwork Member Posts: 2,084 ✭✭✭✭
    @bitcanuck I actually know of at least a few people with multi GH/s farms that still run eth-proxy. It's consistent, stable, and does fail-over well. It's far from great, but it does the job.

    @miningpoolhub Awesome to see the stale stats! Sweet!
  • workwork Member Posts: 2,084 ✭✭✭✭
    edited May 2016
    @bitcanuck probably just bad luck. It seems like overall the stale share changes affected returns by maybe a percentage point or two, based on my rough estimates.

    I'm getting under 3% 1-step stales, which seems about right.
  • BubbaPumpsBubbaPumps Member Posts: 31
    bitcanuck said:

    Pool returns have been terrible in the last 24hrs, about 76% of ideal returns. While this could be just a typical spell of bad luck, it's an odd coincidence for such bad luck to happen just after stale shares get accounted for.
    I'm hoping a bug didn't get introduced into the pool software causing block solutions to be lost.

    Well this was a bad time to start mining here.
  • workwork Member Posts: 2,084 ✭✭✭✭
    @BubbaPumps it'll even out. The previous week had a few days of very good luck. Luck is a fickle thing - it comes and goes.
  • retherrether Member Posts: 258 ✭✭
    How has MPH compared to ethermine for those of you who have used both?
  • workwork Member Posts: 2,084 ✭✭✭✭
    @rether I used to use ethermine, but mostly use MPH now. Vardiff for less variance is a big attraction IMO. I also like MPH's global block stats chart.
  • PalmzipperPalmzipper GermanyMember Posts: 24 ✭✭
    Just a fun info I came along while testing GlobalBoost-Y CPU-mining on MPH.

    While running ETH-mining with genoil you can open a 2nd instance and let the CPU-miner for GlobaBoost-Y run.
    Tested with an i5 CPU and a NVIDIA GTX970.
    (But it didn`t work with the older AMD Athlon 64 X2 CPUs.)
    If you are concerned about too much drain on CPU-performance you can reduce the threads you want to offer for CPU-mining with the --threads=x command.

    Besides, its so cute, you can even CPU-mine with an Atom based tablet.
Sign In or Register to comment.