Ploting with two NVIDIA Quadros M6000 using Direct mode.


  • admin

    @Njcfamily31907 31K nonces is easily doable by a 480. This is from one of my 470's - writing to an SSD:

    100% (753664 nonces), 29730.34 nonces/minutes, 25m 21s



  • @Wolf No, unfortunately I don't have one (:sigh), only an inefficient gtx 560ti. But I remember haitch said here that an r9 280x can do 40.000 nonces/min (167 MB/sec, more than most hdds maximum write speed). I suppose that rx480 can do better, because the hash function behind burstcoin is sha256, same of bitcoin, and I read benchmarks where rx480 performs better than 280x in bitcoin mining...
    About directly optimization vs optimize after plotting is difficult to say... In my case my CPU can plot much faster on /async mode of janror cpu plotter(unoptimized plot output) than my GPU, so for me it's better to optimize after plotting: but probably this is not your case. Also remember that you need extra space for use the plot optimizer



  • I also do some plotting on the side for my friends who mine burst and an RX480 can easily break 30K+++

    Generating nonces...
    61.59% (9388032/15243648 remaining nonces), 32625.65 nonces/minutes, ETA: 2h 59m 28s...



  • And as for configs, I believe they rely on the type of card and your overall system configuration,

    Just try plotting maybe a hundred thousand nonces .

    use
    8192 64 8192

    and increase x2 every time to see if your system can handle it.

    If it can't start, then use the last configuration that worked.

    It's not really a one-size-fits all solution.



  • @kohai said in Mining with two NVIDIA Quadros M6000 using Direct mode.:

    I also do some plotting on the side for my friends who mine burst and an RX480 can easily break 30K+++
    Generating nonces...
    61.59% (9388032/15243648 remaining nonces), 32625.65 nonces/minutes, ETA: 2h 59m 28s...

    Thats buffer or direct with that speed?



  • also do anyone know if I have to lower the stagger size when plotting more that one drives.

    Example: 1 Drive
    gpuPlotGenerator generate direct D://3225037558013223906_0_34278400_105472

    Example: 2 Drives
    gpuPlotGenerator generate direct D://3225037558013223906_0_34278400_52736 F://3225037558013223906_34278400_34278400_52736

    The 105472 stagger size is already using up 26GB of CPU RAM. I was wondering if I leave both drives to 105472 will the drives share the 105472 or will it double?


  • admin

    @Wolf it will use the configured stagger per simultaneous plot. So if plotting more than one drive at a time, split your stagger over all the drives. eg If you plotted a single drive with a 32768 stagger, and now wanted to plot 4 drives at once, use an 8192 stagger.



  • @haitch Do the drive mine better if I plot one drive at a time using the 32768 stagger vs the 8192 stagger or does it matter?


  • admin

    @Wolf It depends what your bottleneck is - if the card(s) can plot faster than a single drive can write, you'll see a performance boost by doing multiple drives, if however your card cant plot as fast as a drive can be written, you'll see no benefit - a two drive plot will just take twice as long.


Log in to reply
 

Looks like your connection to Burst - Efficient HDD Mining was lost, please wait while we try to reconnect.