Well, it’s about time a start the replotting process.
I had finished filling up my space just before the MadMax plotter was released. Initially I was just happy to be done filling space, so took a break. But now it seems like a good idea to start the replotting process, especially as more tools have released for easier plotting.
Anyone have any links to setup tutorial for MadMax plotting NFT pool plots on Linux?
Current specs on plotter machine:
Intel 10700, 46GB RAM, 2x 2TB NVMe
Curious what kinds of speeds people are getting with similar setups (single plots, and plots/day)?
Also any other suggestions/tips to help ease the replotting process (ie scripts for deleting and replacing plot files, etc) would be greatly appreciated.
For Madmax, the installation instructions for Ubuntu/Debian are here, as well as all the commands to use:
Just delete the old one and install the new one.
Only different in the command line for NFT plots is that now you use -c instead of -p command (don’t use-p at all )
You can find the pool_contract_address (-c) when you hover your mouse over the question mark next to the NFT name in the GUI. or with the CLI command
chia plotnft show
I’m replotting in windows and use:
Release v1.0.55 PSChiaPlotter 2021-07-16 · MrPig91/PSChiaPlotter · GitHub - this program automatically deletes old plots (1 at a time) when making new NFT plots. Here is a guide how to use it: PSChiaPlotter Plot Manager Update! Replotting, MadMax, Auto Plot Checking, and More! (New Features!) - YouTube (use it at your own risk, I’m not the creator).
Release madMAx43v3r/chia-plotter build for windows v0.1.5 · stotiks/chia-plotter · GitHub - In PSChiaPlotter, I choose to use the external program madmax (windows version) in order to create the new NFT plots faster. Same comment as above applies here as well, use it at your own risk.
I use two computers with 10850k processors and 16gb of ram and 2x2tb SSDs. I can create >40 plots/plotting computer with this setup. I will have replotted all of my OG plots (now @ ~3600 since I have returned some HDDs) within the next month. This process requires minimal effort in my opinion.
Thanks for the responses.
Just did my first few MadMax plots, and I guess I need to play around with the settings a bit more as my times seem a bit slow.
./chia_plot -n 1 -r 12 -u 256
and got times of ~5k seconds/plot, which would only be about 17 plots/day. Using this same setup with parallel plots w/ plotman, once ramped up I was averaging 25-30/day. I suppose I could up the threads all the way to 16, would that be recommended? Any other variables that I should play with to optimize? I’ve seen some people using the -v (changes buckets for Phase 3&4?)
Try with -u 512 and v 128.
Im getting better time with this config (4300 seconds) - r 8 16gb ram i7 9gen.
Thanks for the suggestion. I just learned about -K parameter as well, so I’ve seen suggestions about using -r8 and -K 2 for a 10700. I’m gunna keep tweaking and will try with different bucket sizes etc, will report back once I’ve dialed in.
edit: Just a quick little update. Ran some plots using these settings: -r 8 -K 2 -u 512 -v 128
and am now getting plot times of around 50 minutes. Probably can do a bit more optimizing, but that’s getting pretty close to the plots/day I was getting with Plotman, with the added ease of not having to ramp up and deal with staggering.