Going down the crypto mining rabbit hole – #1 cpu mining

By Leonardo de MattosUncategorized

I don’t know why, but these days I had this urge to mine bitcoin. Not sure why. I’m usually interested in bitcoin from the investment point of view, when it is cheap I buy, when I understand that it is stretched, I sell. Mostly through ETF instead of trading the coins. I could manage to make few bucks operating like this over the years.

I’m by no means a “crypto bro” and/or a believer that bitcoin is some sort of savior and/or the solution for every problem on earth’s surface. Nonetheless, I find it an interesting concept, and kinda resonates with my rebel soul.

That said, I started to dig into the crypto mining rabbit hole. The first and obvious and conclusion is that I am very late for this party. The second conclusion, is that the learning curve is steep. The information is sparse and there are a lot of snake oil stuff. It is no easy feat to differentiate legitimate information/software from scamming stuff. Moreover, it’s been a while that I don’t write about anything, and I miss doing it, thus I’ve decided to share this experience from hobbyist perspective. The ideia is to write a series of small posts with the small advances, discoveries, and ideas that will come up along the way.

From software perspective, my first attempt to actually mine some bitcoin was with cpuminer [1], using the docker version [2] of it. First of all, it works. I’ve manage to set up to four computers to run cpuminer all at once, reaching around 150 MH/s. I’ve left only two cores from the VPS that host this website crunching bitcoin hashes. Follows the last 24h of historic:

The docker version of cpuminer was a nice first attempt because it granted me a first approach to concepts such as:

  • servers;
  • bitcoin wallet address;
  • workers;
  • decisions around hashrate and reasoning around it;
  • solo mining x pool mining;
  • sites that are interesting;

All within a well contained environment without exposing the whole computer to potentially sketchy software.

cpu mining leads to very low performance in comparison with the current network hashrate. For this reason the chances to find a block alone with this approach is nearly zero. With the average hashrate of 150 MH/s the odds would be:

This odds are too low. 1 in 60 trillion of finding a whole block per day, in the best case with all my computers crunching hard with cpuminer. Compounding the heat, the noise, the chance to damage the computers keeping them under continuous stress have shown that this is not the path to be followed.

At this moment, I’d reckon there are two big paths to follow:

  • a large bitcoin mining farm which would lead to a whole business around it. Here I’d be discussing CAPEX, OPEX, IRR, breakeven, etc. For now, this is not where I’d like to go.
  • and there is solo mining (or lottery mining). Well, almost everything may fit in this category if you don’t go to a large scale mining setup. However, I feel that this may be a game of understanding your odds and tryings to maximize your chances. Here, I may be more interested at this moment.

Follows my curated link list regarding bitcoin mining:

Those were my first step. The next post will verse around the first steps with ASIC miners.

Leave a Reply