Hello everyone, I’m looking to get some advice. As i have been injured several times in my life, I cant hardly work anymore so I’m using farming as my income for the future. The advice I’m looking for is, I dont want to put all my eggs in one basket(chia) so I was thinking of purchasing a 4090 or a storage server to help with compressed plots. With a 4090 I could farm other bitcoins, but i dont know which ones I should use. I will make at least a 1 pb farm for chia, just working my way up there. I have a 10-15 years old gaming pc that I am thinking about using with TrueNas. It has a amd FX-6350 6 core cpu and 64 gb ram (probably ddr3). I have never had to money to build a decent farming rig until now. I could even buy a dedicated farming rig but I dont want spend that kind of money yet. Anyways, I need some advice for my next steps. my next immediate step is to buy 20 tb hdds one at a time every 2 weeks but I’m trying to plan beyond that.
Youll be risking even breaking even…
I would not do this to make money, but to play with the tech.
Btc mining has been an asic game for years now.
I wouldnt attempt to make money by mining btc with a gpu.
Overall i wouldnt do or advise doing what you plan.
Unless you have free electricity.
Agree w/Bones. Also, if you buy 50 x 20TB HDs for 1PB as you posit, that is maybe $200 x 50 = $10,000 … yet you are not going to build a plotter because you don’t want to spend ‘that kind of money’? What kind of money do u think a plotter costs? If u ‘now’ have the money to build a ‘decent farmer’ … but not a ‘dedicated farmer’ … then how will u either plot and/or farm anything? I’m confused … what are you actually trying to do with Chia?
I apologize for the confusion. I am trying to build a dedicated farmer and I’m trying to decide on what route I should take to do that. Yes I don’t want to buy a 10,000.00 machine but I want to make one “slowly”. Since I’m building it slowly, do I really need a nvidia gpu for a plotter? I’m making plots around 35 mins, so here’s a question, once my plots are built do I have to make more? Will I always be plotting? Does a plot get used up when you find the coin? I just don’t see the need for a nvidia gpu unless I will always be making new plots. I have couple old pcs and laptops that I could use as extra harvesters.
I’m getting a little worried, because I dont know what to do to make money anymore. I’ve had 3 left knee surgeries, 4 lower back surgeries and I have IBS and other stomach issues, so its hard for me to work in the field anymore. I’m a low voltage tech. Luckily my wife has a really good job, so thats why I will purchase the hdds slowly. I realize it will take me 2 years to get all the hdds.
Sorry for the rant, i just really was hoping chia would be my answer for income but i guess thats not going to happen.
Edit: I guess my dream of playing games all day while my machine farms for me isnt going to work. D*** it man I’m really enjoying this.
No.
Or, yes, only if you want to win more often.
You will plot only when you want more plots, to increase your chances for winning.
A plot does not get used up when you find the coin.
That winning plot has the same chance of winning, as any other plot (of the same “K” size).
It is akin to playing the same numbers in the lottery. If your ticket wins, those same numbers can be played again, and have the same chances to win again.
Also, just about any non-ancient computer can plot.
But to harvest compressed plots, you need a GPU that is fast enough to work with your compressed plots.
People use fast GPUs for plotting, but that is a “want” (not a “need”).
A fast GPU will create plots far faster than plotting only with your CPU.
Keep in mind that if you use a GPU to plot, and to also harvest, you will be sharing that GPU’s resources, possibly risking your harvesting time running long.
Perhaps a 4090 can handle both harvesting and plotting of compresses plots?
It depends on how many compressed plots you have and the compression level of those plots.
Low compression means that it will be able to harvest more plots.
Vice versa for highly compressed plots.
But then add in simultaneously plotting with that 4090, and the numbers change.
You can try plotting at the same time as harvesting, with your 4090. And if need be, you can always plot without the GPU. It will just take longer.
Or you can get a lower end GPU, that won’t break the bank, but still be fast, in a separate box, and dedicate it for plotting.
Plotting is the easy part. Speed is not required, but it is desirable.
But for harvesting, speed is required. That is the tricky part.
I think if you’re in a financial position where purchases must be carefully considered, you should not take on the immense amount of risk/hope with mining, it may not provide the steady long-term income you’re seeking.
There is an intended plot format change in the future, we don’t yet know the details or timelines, making a purchase today with certain expectations which may change tomorrow, this is why I express caution on the hardware.
This was my expectation, but if I dont have any compressed plots, I can use my cpu for plotting(amd 7800X) and I can use my old pcs and couple laptops to harvest as well, therefore giving me more chances because I’m not compressing my plots.
This is why I am here. Thank you for your advice. I am planning on going without compressed plots. So that way I will continue to go slow and buy a hdd every 2 weeks, unless theirs a major sale.
Its too bad that tape drives are too slow for chia. Hdd storage is a topic a follow so these are always popping up about how much tb they can hold. I think its above 250 tb now, but cant remember the exact number, but wouldnt that be nice, 4 tape drives for 1 pb?
Non-compressed plots have the same chances of winning as compressed plots.
The difference is that with compressed plots (depending on he compression level), you can fit far more plots on the same hard drive. If you have double the number of plots (due to using compression), then you have twice the effective storage space, and twice the chances of winning.
However, with compressed plots, you will use more electricity, for computing the missing portion of the compressed plots, on-the-fly.
No the reason I said I would have more chances of winning, is because I will have more than 2 old old pcs as harvesters and I can use them because I dont have compressed plots. IMO it would be better to not compress the plots so I can use more than 1 machine as harvesters and i dont have to buy a gpu.
So for me, buying a gpu and compressing my plots doesnt add up to non compressed plots with more than 1 harvester but that doesnt mean it wont change in the future. I can always buy 1 or 2 gpu’s (I’m thinking 2 4070 ti Supers instead of 1 4090)
Here’s a thought. Don’t buy an expensive GPU e.g., RTX4090 plus a new PS (850W minimum) plus hope it will fit in ur case (it’s huge). Instead, use your 7800X, buy a 30 series card like a 3070Ti (still an excellent GPU for plotting) and some memory - 128GB is mid-tier for plotting and a few disk and get started. Nothing like starting and trying to see how it goes to guide you when making future purchases.
With the above you can plot a few drives, then also farm them on the same machine for a while. See how it goes, and then if you want - plot a couple more drives. Rinse and repeat as you feel comfortable.
Plotting is much better w/a GPU, any decent GPU, over CPU plotting 
regardless of whether you choose to farm w/a CPU or GPU, which depends on the compression level you chose.
You can always sell a GPU, upgrade to a better one in the future, if you need to or want to.
My advice, don’t get involved in crypto full stop. You’ll highly likely lose money.
I got involved in 2021, spent a fortune and still don’t have a profit.
I got into Chia in April 2021, also bought a couple of Helium miners, also a 3080 for Nicehash.
Even if I’d sold the coins as I won them, I still wouldn’t have turned a profit.
I still run the Helium miners they use 5w so negligible power use, I still run a farmer, 321TiB raw space, with compressed plots, so 759 TiB effective. It earns me about £12 every 3 days roughly. Electricity costs have to come out of that, but I have solar so cheaper than usual.
If you keep the gains its a huge gamble on the future price, which is what I’ve done. If you sell then you’ll likely never return on investment now. Hard drives don’t last forever. Farming uncompressed plots is even worse.
With hindsight, I should have just invested the money in traditional stock funds, I’ve been doing this also from about the same time (my pension), and guess what, I’m well up.
Forgot to add, we had a halving in rewards back in March, in March 2027 there will be another halving, UNLESS the price of XCH increases substantially (the gamble) each drive will not have paid for itself by then.
Do the maths.
at this moment just buy chia coins direct without pay for the energy, taking care about the hardware and system. the price of the chia coins is low, its better buy it than farm it.
Everyone is correct. Chia is and has been a poor… no bad investment, and looks to stay that way. I’ve made more earning interest in a dumb bank account than Chia has ever produced (which on balance is negative).
What Chia does offer is entertainment and educational value. That is actually worth a lot, but it is intrinsic value, not actual monetary value. You might think of it as going to school, you pay ur $$$… learn something & have an experience, leave the school and your money is gone. What do you have then… nothing??
That schooling gets you at a store…exactly nada …but perhaps at a later point, you can call upon that knowledge & experience to produce further rewards that are monetary.
That is all.
I started farming a month before Chia even had a price. Never sold a coin and I have made about quarter of the equipment’s cost back. Could have 4x Chia if I just bough the coin now for the same money. Spent couple hundred hours of fiddling with the farmer + the electricity bills + paying the income tax for the farmed coins. I would say just don’t do it. You won’t be able to sell the farmer for a reasonable price. I made the mistake of thinking this equipment would be easy to sell. Might be able to use the drives for some AI stuff, but that is a long shot as I don’t have experience in that. If you want to invest to Chia, then just buy the coin. The coin price will have to go up at least some amount if this blockchain intends to stay alive. Nothing is ever quaranteed though…
Agree. And my two cents is that not only is it a bad investment from cost (infra/power) perspective… but I have found it to be a huge waste of time.
Learning all the optimisations, waiting for enormous files to move about your netspace, even with scripting… shopping for, ordering and maintaining the equipment, re-plotting, keeping up with the arms race etc is an ENORMOUS time cost that is totally not worth it.
Compare that to opening up Nicehash and clicking start (granted Nicehash is also not really profitable).
I have learnt some stuff, no doubt, but it’s not really reusable skills. Linux is great, but I’d rather build a Raspberry Pi / Home Assistant / NAS project that actually does something.
I also want to add that the whole green-washing thing has also been a huge disappointment, and total BS.
Buy a WeatherXM Station (also Helium IoT which it operates on). At least you learn a bit about radio on the Helium side. That’s making $2 USD a day for me currently, for a $600 investment. Also much less pointless tinkering. Check out other DePIN and distributed compute projects for AI cryptos and save yourself the grief manhandling unwieldy files. At least you will be being productive with your efforts.
LOL, I have a lot of that at this point…so no loss here 
In fact, call me nuts, but it’s some of the best fun time I have …the more it changes…the more of a challenge it is…not sure what that says, but anyway 
Yeah I’m kinda like that. A big puzzle to solve that scratches the OCD itch. But my point is (for OP joining) you should know it’s onerous, not profitable, and there are probably more productive/creative ways to spend your time.
Biggest benefit for me is I now have a Proxmox cluster/home lab with all the equipment I have acquired/configured.
To be honest though, I could have set the same thing up and learnt the same things with a lot less investment had I avoided Chia.