Optimizing farming?

With all of the extensive talk about optimizing plotting, I’m having a hard time finding any discussion of optimizing farming. Is that because it doesn’t matter? Or is it just poorly understood? Or can I just not find the great discussions?

Tell me if I’m wrong, but there is a time component related to the farming that can affect your success rate in reporting to the chain/pools. Has anyone done research to figure out what the best configuration is for farmers/harvesters? I’m pretty much tapped out on drive space for now, but I have more room to optimize the farming across multiple systems.

It doesn’t matter. If many people can farm on the Pi, it means there’s no optimization needed unless it’s not working.

Most of the issues are about the sleeping drives, farmer and harvester are not connected to each other due to some firewall configurations.

For farming, it’s more like troubleshooting than optimizing :sweat_smile:

1 Like

Yes lookup times, if your late it becomes stale.

There’s a lot of causes for this and inefficiencies in the old client which is why we made our farmer. Lookup times can easily be under a second on average with an optimized farmer and node.

Optimized farming is fairly straightforward.

  • Stay online 24/7
    • Stays synced with block-chain
      • or use flexpool’s flexfarmer (no blockchain syncing required).
  • Return proofs in a reasonable amount of time.
    • Fast look up times.
    • Hard-drives always awake and spun up.
  • Make sure your set-up is reliable.
    • Doesn’t run too hot.
    • Should last many years.
    • Good electrical supply.
    • Good internet connection.
  • Be as energy efficient as you can with your available hardware.
    • Chia is supposed to be environmentally friendly dont-cha-know.
  • Be cost-efficient with your harvester/farmer.
    • Recycle old hardware, enterprise hardware, old laptops, unused hard drives.
    • Use a Raspbery Pi if you have one.
      • Because they’re cheap and low power.
  • Join a pool if you’re impatient and want to receive rewards quickly (for a small fee).
    • Good for small farmers.

Did I miss anything? Pretty sure I got the main points. Basically a set-up you can leave in the corner somewhere to quietly farm away without much intervention. i.e easy to maintain and leave unattended.

6 Likes

I appreciate the replies, everyone!

An alternative, and the path I’ve chosen, is to make my daily use Windows PC my farmer.

Benefits include:

  • GUI farmer viewport always visible
  • Easy log access
  • Using reliable existing hardware
  • Neat, compact overall hardware footprint
  • Multiple USB/SATA power/data connections simplified
  • Single power draw
  • Easy to manage network connections (when using plotter)
1 Like

You contribute nothing. Go howl at the moon again.

Drjones, How do you go about this setup? I’m interested.

Important = yes
why you cannot find the solution = not many people are big enough and need that optimization

Managing 10 drives is very easy, but 1000+ is different story

Good luck getting a straight answer.

Sure.

Are you claiming you can win on the same plot from multiple vm’s for the same block?

Otherwise, if would seem, you have tricked the software to show 10 x capacity, with no extra chance to actually gain any extra chia from it.

I don’t think your cheating but more likely confused, of course it could not be so hence analysis needed.

This makes no sense unless you had a very low powered machine which seems to not be the case.
Each plot has certain proofs or not…

Again makes no sense for previous reason given.

That’s not winning , but passing filter, and again makes no sense for only answer I gave.

1 Like

This guy is a raving lunatic who never shared anything legitimate with anyone, if there was ability to do what he claims then it would blatantly obvious on the network it was being done. There has been several world class engineers whom have tried. Don’t bother entertaining his delusions.

3 Likes

I think if it were possible, they would be to busy adding more hdd and deploying more vm’s to be posting here.

2 Likes

If you think you have actually done this, then you have committed Federal wire fraud or something similar.

It would only be showing to them I think.
Doesn’t the network as a whole estimate size at the network level?

Yeah, lol! estimate is generous given the wildly fluctuating “estimate”.

I have no idea what security is built into the network and the estimate to detect and avoid the things that drjones is supposedly trying to do.

drjones may simply be a Troll using imaginary FUD.

Troll or not, I suggest we stop feeding him.

User profile hidden, can’t see his other activity, posts … so you could be right …

1 Like

Proof of coins earnt.
Proof of true size of farm.
Not hard.

Edit. Editing this post as it was flagged as an advertisement.
Not sure who thinks I’m advertising something here, or what exactly I’m advertising supposedly…

1 Like

I think you have managed to fool yourself.

I did find that one element that you mentioned, K8s, was very interesting.

Kubernetes, also known as K8s is a powerful container manager that just might give you some of the effects you describe.

I will be making another visit to Keybase to bring this conversation to the attention the Devs.

You are probably deluded and you are definitely attempting fraud. The fact that you publicly proclaim your fraud proves that you are also the village idiot.

1 Like