Oh Bitcoin is still a thing? Should I start mining then?

In short… probably not.

Why do folks ask this question? More often than not, the question pops up in times when Bitcoin’s price is rising, and they may see mining as the cheapest way to get their hands on some. i.e. “Why buy bitcoin at $1,000/BTC when I could just buy a graphics card and get some for free, right?”

It is not as simple as this though, and in most cases not at all feasible. Here’s a breakdown of why that is, and of the very specific cases where mining may still be a feasible venture.

The Mining Premise

The basic premise of mining is that persons from around the world can contribute their processing cycles (CPU or otherwise) to the bitcoin network. In return they get paid for the number of cycles they contribute. These cycles are used to secure the entire Bitcoin network.

Payment is made in proportion to the number of “effective CPUs” a person can contribute. These contributions are relative to the current size of the mining network. The payment, better known as the ‘block reward’, is the thing that we’re interested in looking at.

Understanding the hardware

Bitcoin’s original vision as laid out in its whitepaper was ‘one CPU one vote’. Persons would contribute their CPU cycles to vie for the chance to ‘add the next block of transactions to the network’. In turn they would be paid for each block that they get to add.

In the early days folks did go the route of CPU mining as was the original design. This however, quickly accelerated through the ever evolving and creative use of more advanced pieces of hardware.

The progression of mining in Bitcoin went as follows:

• 
  

  CPU mining eventually gave way to GPU mining(where 1 GPU = ~28 CPUs)¹

• 
  

  GPU mining eventually gave way to FPGA/ASIC mining(1 ASIC first-gen = ~33 GPUs)¹

• 
  

  ASIC mining rapidly improved until running headlong into **Moore’s Law**(1 ASIC today = ~212 ASICs first-gen)

A comment from Reddit on different hardware types

Size of the network today

It is important to note that the absolute rate at which bitcoin is paid out globally does not change based on the number of miners. It is actually determined by a fixed supply schedule. In the early days it was 50 BTC every 10 mins and today it is 12.5 BTC every 10 mins. This equates to about 1800 BTC per day, and this is constant whether there is just 1 person mining or 10 million people mining.

Today, the bitcoin network has grown to the equivalent of 2,460 PHash/s which is just shy of 1 billion effective GPUs (if you’re a gamer) or 17.6 billion effective CPUs (if you’ve built a high end desktop computer).

A Bitcoin mining farm operation

For perspective on just how large these numbers are, let’s consider the odds of winning the US$149 million Powerball. Winning is roughly a chance.

You are 5 times more likely to win the Powerball than to find a Bitcoin block using GPUs.

You are 100 times more likely to win the Powerball than to find a Bitcoin block using CPUs.

This translates into average payouts that look something like the following:

• 1 CPU running at full capacity nets $0.000107 per day (or $0.025 per year) at current prices
• 1 GPU running at full capacity nets $0.001970 per day (or $0.47 per year) at current prices

I’ve seen this myself with a little mining rig that I set up just about a year ago. My rig was as powerful as using 2 of the highest end GPUs you could probably find today. Even then, daily payouts looked like what you see in the following image.

Results from my tiny 4 GH/s ASIC rig

Feasible bitcoin mining

That’s not to say mining is entirely dead though. Instead, it has evolved into a global energy arbitrage game of sorts, where feasibility is determined by two factors:

• Access to the latest and most efficient ASICs (at a reasonable cost)
• Access to the cheap electricity, wherever it might be

Fortunately for us in Trinidad, we have some of the lowest power costs in the world. A friend of mine and I have taken advantage of this and we’ve been experimenting with some Antminer ASICs to see just how feasible mining is locally.

A heat gun pointed at two Antminer S7s we’ve been toying with

What we’ve found is that if you can access industrial rates (US$0.02 - $0.03 per kWh) and you can avoid having to liquidate the mined BTC regularly to meet daily costs, then mining may be feasible. This of course assumes that you’ve bought into the overall Bitcoin value proposition and expect its value to continue growing over time.

With our own operation, and based on 200 days of data, we’ve found that we can expect to eventually come out with ~70% cheaper coins after 2.5 years than if we would have bought those coins outright. This is a very optimistic projection though, based on a number of uncertain assumptions.

The altcoin route: GPU mining lives!

Mining with non-ASIC equipment may also be feasible if you get a little creative. Altcoins (alternative blockchains) usually have much smaller mining networks and different mining algorithms. These properties make them far more attractive for CPU or GPU mining. One can also take the proceeds from mining on these alternative blockchains and convert them back to bitcoin on a regular basis. This is a more technical route though and the trick with this method is keeping a close eye on which altcoins to mine. This is so because the specific altcoin mining network sizes and exchange rates can fluctuate very rapidly.

Fortunately, there are sites like the following that take the various factors into account and provide comprehensive tables on the altcoins that are worth mining at any point in time.

A sample from coinwarz.com

What this means

To get back to the original question of ‘should I mine?’, people usually ask because they’re thinking of what’s the best way to get their hands on some bitcoin. In most cases the answer is usually that they should simply buy them outright on an exchange as this will afford you all the benefits of any future price increases without the headaches of having to manage a small mining operation.

These headaches come from:

• having to maintain equipment and cover power/cooling costs
• having to house them in a suitable location (climate-controlled, noise-tolerant, fire-safe)
• taking on the risk of hardware issues or sudden network growth that could significantly cut your proportion of payouts

What you end up gaining by taking on this task of mining is a route to maybe marginally cheaper bitcoins.

That being said, there are cases where mining does make sense even after taking the above factors into account. It might be preferable if you have:

• Spare capital (in multiples of ~US$1,800 for recommended Antminer S9 ASICs)
• Cheap power costs
• A space where you can run noisy, high load devices in a relatively climate controlled environment

It is also preferable in countries where accessing bitcoin is difficult because of lack of exchanges or currency controls. Miners are a great way to convert local power to a liquid asset and they effectively serve as direct foreign exchange earners for the owners of the operation in these sorts of locations.

> Hit me up on Twitter

> Share some BTC love: 1JruB7NxGuduDLE2rRbd7c561Mg8Aa3RcQ

1. ^ Statistic taken from the Digital Gold book