Thought that madMAx43v3r Gigahorse 2.0 version deserves its own thread,
1 Like
nice, so i got something todo this weekend, C18 replot 
Same here, going for C19 though.
Going to be pretty much set and forget, only thing I’ll need to do mid week is update the list of drives in the plot mover software.
Linux Mint plotting times with 512GB of ram, both cards plotting at the same time.
3080 C19 = 126 seconds
P4 C19 = 350 seconds
So if I’ve done my maths right, that should be around 56 TiB of plots a day, or around 6 days to replot about 320TiB of raw space.
Where is gui creator mmx plots. On www is old version 0.0.9
Thanks
Plots rolling off my plotter, and already being farmed by Foxy-GH-Farmer 
Old BB C7 being farmed by Foxy-Farmer, and slowly being deleted.
PS Need a faster plotter, I started plotting prior to starting to move them, and the back log is being cleared.
Is it true that the sizes of the plots lose to the Nossd version?
and also more load?
anyone have experience comparing?
yes NoSSD plots are about 5% smaller, computing power seems to be approximately the same but that’s just my guess when I look at the max farm size for the GPU exactly only test can show that for the Max GH release is too new.
For the 1-2 months that Max has worked on it is a decent performance if you consider NoSSD had about 1 year time.
But in the end, the plot technology is currently NoSSD the best.
Farm technique I personally think GH is the better solution, free pool choice & solo farming.
ultimately, everyone must decide for themselves then, because both solutions are currently very close to each other.
have a nice day
edit: yes , Remote Computing is alse possible with GH, but not with NoSSD
1 Like
the big drawback of Nossd is the lack of recompute_server
also wondering how video memory consumption will change with higher compression? because for c8 now 4 GB + 4 GB recompute_server! i only have 12gb which can be critical with higher compression
Size of plots are smudge bigger on the larger compression.
But I think Gigahorse gives you, the farmer, much more control over your farm…
-Pick your pool or go solo
-remote harvester
-CPU or GPU farming with less fees for CPU
-and more…
Having both NoSSD and Gigahorse is good for all of us. Creates a healthy competition that improves farming for all.
Maybe time to replot…again! Was wondering if someone can help me understand:
- On the GitHub page, it says GH uses the official pooling protocol, does that mean I can continue to pool with SpacePool?
- On the GitHub page, it says I can run the official Chia GUI, AFTER I launch GH. What is the reason for doing this if the official Chia GUI doesn’t support GH plots?
- I’m guessing the GH plotter runs on Ubuntu, but I see arch in the filename so wanted to confirm
- I can farm in Windows, right?
Thanks!
- You can continue to pool with any pool you want, except NoSSD’s pool.
- I think the GH farmer feeds some information back to the GUI, so you can see total amount of plots, also see the wallet etc. But the GUI version has to match the GH version, so for now that’s 1.8.2. Max is going to update to 2.1.1, but that’s a way off as there is a lot of work.
- It runs on Linux or Windows, for Ubuntu you want the x64_86.tar.gz.
- Yes you can.
I dont want to be the “negative” guy from here, but to do that – free pool choice you need full node right ?
To have on every machine full node - means you need extra 130 GiB or more
In my case, 40 machine x 130 GiB = 5.2 TiB only for nodes.
so , yes, you are right, everyone must decide for themselves the best choice.
1 Like
Flexpool was great in this case.
no, 1xFarmer & 39 x Harvesters should be no problem.
so u need only 1 Full Node.
and yes, if u want the full choice of nearly all Pools u need a full node.
in this case, farmer must have a GPU and harversters too ( another GPU) ?
40 machine is too much to compute for a farmer… i think.
it isnt like the clasick mode – I had 1 farmer + 40 harversters and no problem.
If your using the Chia software, then yes every harvester must have a GPU.
If using Gigahorse, then you can use a compute server, which not even NoSSD has, so one GPU for multiple harvesters, or two GPU’s or more shared among all the harvesters.
You can also use the Foxy Farmer software and don’t even require a full node, I’m currently using both versions as I replot from BB c7 to GH C19
That would be a remote harvester setup.
Yes thats the normal farmer / harvester setup