Seedhash is always 0

sandedsanded Member Posts: 20
this seems to prevent me from mining a block. :(

Same for this guy on reddit

Does anyone have a clue?


  • MarioFortierMarioFortier Boston, MAMember Posts: 30
    miner 13:39:27|ethminer Got work package:
    miner 13:39:27|ethminer Header-hash: e3297af310836367bb7b1dbc184d6454f9f5767
    miner 13:39:27|ethminer Seedhash: 000000000000000000000000000000000000000000
    miner 13:39:27|ethminer Target: 000000000e6cfeab16bfb475e0f8d5d09763f880f3e8

    Same for me... but I did mine a block so it seems to work.
  • sandedsanded Member Posts: 20
    I just assumed that when a lot of people mine with the same seed, it's very unlikely to mine a block
  • MarioFortierMarioFortier Boston, MAMember Posts: 30
    sanded, you might be right.

    Me worrying now...

  • MarioFortierMarioFortier Boston, MAMember Posts: 30
    edited July 2015
    FYI, for a test I did force regenerating the DAG (by deleting .ethash) and still observing seedhash of zero.

    Update: Found another block. Assuming I am not in forked world, the block finding rate seems OK in my case.
  • sandedsanded Member Posts: 20
    maybe I'm just impatient then..... thanks for the update!
    Also, maybe the seedhash of 0 is correct for genesis.

    A confirmation would be highly appreciated though!
  • sandedsanded Member Posts: 20
    edited July 2015
    ethminer shows the hashrate. It seems to be working fine.
    (hankndank, you're also using ethminer -G, right!?)

    By this time I assume that's the seedhash of 0 everybody uses at the moment and it's related to the block era.
    I interpreted "Seedhash" as something that should vary... but apparently the "random offsetting" takes place elsewhere.
  • ChrJentzschChrJentzsch Member Posts: 17
    Seedhash: 000000000000000000000000000000000000000000 is just fine. don't worry. The seed for the nonce is a different one.
Sign In or Register to comment.