I've stuck few times in the past and it was merciless fucking pain in the neck to clean 'em up.
I did double spend, RBF, viaBTC, and waiting long days to get out of it, and, sometimes, I feel like I'm getting nowhere.
From then on, I've sought the ways to specifically assist low fee transactions be confirmed for others like me, and, to the least of my surprise, there isn't much to find out.
All the articles say "unsustainable", "no point of doing it", "you should put enough fee", and list goes on.
I've read enough about those naysay, and I don't give a fuck about how much I would make from cleaning the transactions, nor how I should be careful.
From time to time, there is nothing you can do to a situation happening before your eyes.
If you're to say anything similar, please refrain yourself and go read some good posts.
All I care is to help one more of those poor soul to get out of helpless situation.
Old mining gears such as S5 ASIC seems to be cheap enough.
I can pay elec-bill for one or two oldies as long as the gears don't burn my house.
Any article, github, or whatever I can read to get some help would be highly appreciated.
Thanks in advance.
In order to choose the transactions that get confirmed, you need to solve an entire block yourself (or at least run a pool where one of the miners in your pool solves a complete block by themselves).
If you want to be able to solve (on average) 1 block every month at the current difficulty (meaning
once per month you get to confirm some low-fee transactions), then you (or the pool you operate) will need at least 886432559434177 hash/second.
That's a bit more than
886.4 terahash per second.
An Antminer S9 will cost you about $2100 and will produce 14 THash/sec. Therefore, you'll need:
886.4 THash per second needed / 14 THash per S9 = 64 Antminer S9 mining ASICs.
At a cost of $2100 each, that will be:
$2100 per unit * 64 units = $134400
Each unit will draw about 1.372 kilowatts of power. Since there are on average 730.5 hours per month, if you run all 64 of them continuously all month, you'll use about:
730.5 hours * 1.372 kW = 1002.246 kWh of electricity.
I don't know where you are located, so I don't know what your electricity costs are. Lets assume for now that you can get your electricity at a rate of $0.11 per kWh after all fees and taxes are included.
1002.246 kWh * $0.11 per kWh = $110.25 per month in electricity.
So, (assuming you live in a VERY cold climate and won't need to pay anything for cooling the equipment, and that you don't have any storage or maintenance costs)
your total costs will be an initial purchase of $134,400, plus an additional $110.25 per month.
Of course, since you are solving a block per month, you will have some revenue (12.5 BTC per block). Assuming that the exchange rate stays stable at $1450 per BTC and that the difficulty doesn't increase, that means
you'll have revenue of $18,125 per month.
If you can keep the exchange rate and mining difficulty stable and avoid any storage, cooling, or maintenance costs, it will only take you:
$134,400 initial purchase cost / ( $18,125 monthly revenue - $110.25 monthly cost ) =
7.5 months to earn back the $134,400 that you spent to get the mining equipment.
Of course, the bitcoin exchange rate is never stable (and could collapse to less than $1000/BTC at any time (or could increase), and the mining difficulty is always increasing (meaning that over time you'll solve a block less and less often). Therefore,
the actual number of months until you've recovered your costs is likely to be very different from that estimate.
Given the costs associated with mining successfully, and the associated risks, perhaps you can begin to see why miners need to maximize their revenue? If there are two transactions waiting to be confirmed, you only have enough room remaining in your block for one of them, and one pays a higher fee than the other, then you increase your chances of recovering the $134,400 that you spent on equipment by choosing the transaction that voluntarily offered you a higher fee.
Additionally, if you don't confirm the higher fee transaction, then some other miner will. That will make it easier for that other miner to be able to afford more hash power sooner than you can. Their additional hash power will cause the mining difficulty to increase, and since you can't afford the additional hash power that the other miners could (since you haven't been confirming the high-fee transactions), you'll solve blocks less frequently and your revenue will decrease. Eventually, the difficulty will get too high, and you won't be able to solve a block even once a year (or once a decade).
As you can hopefully see, mining is a very competitive business. The bitcoin system is currently designed to force miners to confirm the transactions that voluntarily offer the highest fee per byte. Any miners that don't are eventually driven into bankruptcy if they don't quit soon enough.
I wish you luck in your charitable endeavor. To keep up with the difficulty increases and guarantee a block of free transactions every month, you're probably going to need to collect at least 1 BTC per month in donations (from others or yourself) to your mining charity in addition to your mining revenue. This will hopefully make up for the lost transaction fees and allow you to purchase enough new hash power to keep up with difficulty increases over time.
Given that, perhaps you finally see, that when a miner chooses to confirm a low fee transaction the miner effectively pays the fee for that user (since the miner will need to either reduce his own profit by that amount or contribute that amount of his own money to keep up with difficulty adjustments). Do you really want to pay transaction fees for a bunch of stingy cheapskates that aren't willing to pay their own fair share?
