Parallel plotting settings

When you enable parallel plotting from the Windows GUI, are the settings for memory and threads for each of the parallel plots individually or for the entire set of parallel plots?

Individually.

I would not use GUI plotting. Why? A: You have to stop the plotters when a new update is released and you lose many hours of plotting.

Use CLI plotters.

hey @bikerdude,
do you have any advice when one sets out to learn CLI? i spent ages trying to work out how to open the farmer node on its own, rather than; chia start all. the main reason why i want to learn it, is because the gui keeps on freezing. i have since updated the gui however, moving forward ive heard the CLI is our friend. what i have found most difficult is understanding how to ascribe values, such as chia start [all] this didn’t work. for an amature such as me i wonder whether there’s a possible “paid” service one would offer to literally guide people through initial set up and not just refer one to github.

See if this tutorial will help.

You should be able to create plots using that method on the same or different machines and then farm from the GUI. Good luck

1 Like

I haven’t figured out how to use CLI yet. I tried starting power shell, but when I try to run the commands I get errors, so not sure what im doing wrong. Probably need to research further. In the meantime, I

I assume you use windows 10? I use the GUI just for farming and wallet.

You don’t need to learn CLI you just need to write a .ps1 file. Here is some help.

First run this to update powershell to version 7.
iex "& { $(irm https://aka.ms/install-powershell.ps1) } -UseMSI"

And just one line to Start one plotting process in a new shell, you can put this in a ps1 file or copy-paste into the shell. If you want to run say 4 plotters parallel you run this command 4 times with different sleep settings. Ex: 0 on first, 4000 on second, 8000, 12000.

start-process powershell { cd ~\AppData\Local\Chia-Blockchain\app-1.1.2\resources\app.asar.unpacked\daemon\ ; Start-Sleep -s 4000 ; .\chia plots create -k 32 -r 3 -n 100 -t E:\temp\1 -d F:\plots }

When it comes to planning how many plotters to run, then I use 2 plotters on 1TB nvme and 5 plotters on 2TB nvme. I don’t use SATA due to low IO (but no more then 1 plotter on 1 SATA drive).

1 Like

thanks for the information dude!!! rahhhhhh you run 10 challenges? does that increase the length of plotting time alot- obviously you’d get more proofs i assume

There is unfortunately no option to decrease the plotting time.

PS C:\Users\bikerdude\AppData\Local\Chia-Blockchain\app-1.1.2\resources\app.asar.unpacked\daemon> .\chia plots create --help
Usage: chia.exe plots create [OPTIONS]

Options:
  -k, --size INTEGER              Plot size  [default: 32]
  --override-k                    Force size smaller than 32  [default: False]
  -n, --num INTEGER               Number of plots or challenges  [default: 1]
  -b, --buffer INTEGER            Megabytes for sort/plot buffer  [default:
                                  4608]

  -r, --num_threads INTEGER       Number of threads to use  [default: 2]
  -u, --buckets INTEGER           Number of buckets  [default: 128]
  -a, --alt_fingerprint INTEGER   Enter the alternative fingerprint of the key
                                  you want to use

  -c, --pool_contract_address TEXT
                                  Address of where the pool reward will be
                                  sent to. Only used if alt_fingerprint and
                                  pool public key are None

  -f, --farmer_public_key TEXT    Hex farmer public key
  -p, --pool_public_key TEXT      Hex public key of pool
  -t, --tmp_dir PATH              Temporary directory for plotting files
                                  [default: .]

  -2, --tmp2_dir PATH             Second temporary directory for plotting
                                  files

  -d, --final_dir PATH            Final directory for plots (relative or
                                  absolute)  [default: .]

  -i, --plotid TEXT               PlotID in hex for reproducing plots
                                  (debugging only)

  -m, --memo TEXT                 Memo in hex for reproducing plots (debugging
                                  only)

  -e, --nobitfield                Disable bitfield
  -x, --exclude_final_dir         Skips adding [final dir] to harvester for
                                  farming

  -h, --help                      Show this message and exit.

Yea, my Windows machines are older and don’t have NVME capability. I do have a samsung SSD in each of the 3 windows machines, but I guess there is no way to get around the GUI trying to be a full node. So, I setup a Linux box on an old HP server and sort-of got the CLI working there. At least it’s running, although I have to do several workarounds to the published instructions. Seems they skip a few necessary details…

In that case I would run one plotter on each of your windows machines. Don’t do parallel on SATA.I would use CLI to run the plotters.

Little about my setup, I have one Win10 machine that runs GUI, many Linux machines running plotters.
They all use same key, that way a plot when finished can instantly be available to farmer. I have external USB drives connected to Linux machines as destination, shared over the network. So no time is spent coping files around.

1 Like

This is a good topic. Firstly, I am running on a Windows machine, 16GB RAM, 2TB M.2 SSD for temp files for plotting, and 32TB external USB HDDs for plot final directories.

So I firstly tried parallel plotting using CLI from a CMD window. It seemed to work, but then all the processes ran into some kind of errors. If I see that again, I will bring the errors up to this forum.

I secondly started parallel plotting with the GUI. I used to GUI to create 4 different queues with 10 plots in series. With 4 different queues running plots in series, you will have 4 plotters running in parallel, continuously until all plotters complete 10 plots.

I am going to try running parallel plots again from PowerShell, rather than CMD window.

2 Likes

Ok, I finalized on 2 HP DL380 G8 boxes with enterprise SDD in Raid-0 stripe. With this, I can plot about 30-32 plots per day on each machine under Linux. Im still confused about the farming side though. Currently, I have just short of 200 plots on a NAS to a single machine running the farming GUI. I think it’s possible to run 2 GUI’s on different boxes and have each look at 1/2 of the farm, but not sure the best way to do it yet.

What CPU and memory are you running on those DL380?

I am seeing similar on some 16-disk DL380s and while I had 16 parallel lots to SATA working in 12.5 hrs each, I went down to 12 parallel plots in 12 hours, plotting to 4 disks and farming on them also. Then rotate as disks become full so you can plot and farm on the same boxes. All JBOD to avoid seek- and write-binding disks together. E5-2650 and 128gb but RAM is overkill.

2x CPU E5-2680 v4, 256GB

Hello people , i would like some help as well,
I’m on Windows 10 with an Asus tuf x570
Ryzen 3900x
48gb of ddr4 3200 ram 2x8 gskill ripjaws 3200 and 1 32gb patriot viper gaming
And 2 nvme m2
1x Corsair mp510 920gb and
1x crucial p5 2tb both gen3 I think
I am trying out to plot 10 parallel
3x on the Corsair and
7x on Crucial with a stagger of 30 mins and settings 2 threads 4000 ram but it takes close to 18-20 hours to finish the 1st plot.
Am I missing something?
I’m plotting on GUI cause I’m a clueless noob about PowerShell and cli.
Any advice ?