gpuPlotting multiple drives



  • @administrators @Global-Moderators @haitch I finally got my gpuPlotter all configured and working--- Now I am going to be plotting multiple drives in gpuPlotter- here are my questions-

    1. my devices are set as 0 0 8096 256 8192 - but it shows I am only using 1GB of VRAM and 1GB of RAM can I leave them like that bc I have 4GB of available VRAM out of my 8GB graphics card or should I reduce it by how many drives I will be plotting?

    2. I'm confused about the nonces starting point and overlapping- the first number in the command line (after account id) should be ? xplotter used a long number that started with 6 (it was the f drive so I realize that it created a starting point of 6 bc of the 6th letter of alphabet) but now I'm direct plotting and I don't want to hit an overlap situation- I'm using 16 GB of 32 GB ram and I'm afraid through my calculations that the nonces numbers will get big and possibly overlap with that drive? should each drive be a continual increase in nonces starting point based on the last drive?

    for instance drive F is AcctID_600000001_11444480_11444480

    my new gpuplotter commands for two 4TB drives
    Drive G should be Acct ID_11444481_?32768
    Drive H should be Acct ID
    ?_?_32768



  • Here's what the plot file name means: <address>_<startNonce>_<noncesNumber>_<staggerSize> The stagger size is how may nonces your miner read at a time, your stagger size should divide equally into your nonces number, if not it will be ignored by the miner. For example: 100000÷1000=100 That means your miner will get nonces from that file 100 times and read them, the higher the stagger size will give you better read times. A bad example: 100000÷1550=64.516129032 For example: plot file 1 4566795722275569365_0_1000000_2000, plot file 2 4566795722275569365_1000000_1000000_2000 and plot file 3 4566795722275569365_2000000_1000000_2000. The start nonce number is the number you do not want to overlap, so my first file had 1000000 nonces and is starting nonce was 0, so I would add the 1000000 nonces to the start nonce number of plot 2, and plot file 3 I would take plot 1, 2 and add the number to getther and that would give me 2000000 nonces for plot 3, and so on. I use these setting for my GPUs, Tesla M2050 0 0 4096 64 8192 and my GTX 660 0 1 1024 16 8192. I hope this helps. :-)



  • @Tate-A thanks so much, that helped me! :)


Log in to reply
 

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