Should I Start Mining Bitcoin?
I talk to a lot of people about Bitcoin. Most of the time I’m explaining to people what it is and how it works. It seems that the part of the conversation that almost always captures the imaginations of people is the process of mining. I can completely understand why, though. When I first heard about Bitcoin it was the itch to mine that hooked me. It’s not hard to see why, either. As I mentioned in my Educational post, What Is Bitcoin Mining?, I was a big fan of SETI@Home back in college. But that was me donating my computer to a cause while I wasn’t using it and, at best, getting a neat looking screensaver. The idea of setting my computer to do a mundane, routine task and getting paid for it flashed Bitcoin signs in my eyes.
Most people, when they hear about Bitcoin mining, think of it the same way I did when I first heard about it: free money. As with most things Bitcoin, however, there’s a lot more to it than that. Bitcoin mining is, on an extremely basic level, verifying recent transactions and receiving Bitcoin as a reward. This is an absolutely necessary part of Bitcoin. Without it the public ledger, the Blockchain, wouldn’t be a secure and accurate log of all the Bitcoin transactions. But mining is not just writing things down and minting Bitcoin. Mining may have started out a hobby, but it has since become big business. Let’s look at the evolution of Bitcoin mining to get an idea of what we’re dealing with.
In 01/2009 when the Bitcoin protocol was first started up you simply had to have a computer and the software. Mining was done solely with the CPU of a computer. You can still go into the settings of a software like Bitcoin Core and turn mining on. Your computer will, as expected, start to hash. Back in 2009 there was even a really good chance that you would find Bitcoin in your wallet when you checked. It wasn’t hard to solo mine back then because so few people were doing it. The hashing power of the network was low. In fact, for the first year the difficulty was 1. It wasn’t until the middle of December 2009 that it even budged. CPU mining was rated in Megahash per Second, or MH/s. Over the CPU mining era, chips ranged from Atom and ARM processors in the 0.25 MH/s range to Intel and AMD processors in the 2 to 140 MH/s range.
It was during this time period that Satoshi asked for a gentlemen’s agreement to postpone the use of computer graphics cards, or GPUs (Graphics Processing Units), delaying that particular arms race until as late as possible. It seems that Satoshi saw the rise of GPU mining coming and asked everyone to wait until they had more people joining the network. This allowed the overall hash rate to grow and spread out over more and more people so that a few users with GPU miners would not be able to overwhelm the network.
The hash rate of Bitcoin remained low as requested and the difficulty only rose into the low 20s. Then, in 07/2010, it suddenly quadrupled in the first two weeks of the month. This is when Art Forz began mining with his custom Bitcoin mining software and GPU rig. Art went on to command roughly 25% of the network power for a time. Whereas CPUs were typically under the 125 MH/s mark and power hungry, AMD GPUs could often more than double or quadruple this rate with much more efficiency. People specifically built mining rigs with multiple GPUs working together to push a single mining computer into the Gigahash per second, or GH/s range. Others soon jumped on board and the age of GPU mining truly began. The difficulty skyrocketed, eventually reaching over 100,000, but it didn’t end there.
In 05/2011 there was another technological leap in Bitcoin mining: The FPGA, or Field Programmable Gate Array. The FPGA is an integrated circuit that doesn’t really have a purpose when it leaves the factory, but is programmable by the purchaser, thus the name. It wasn’t long before someone created an open source FPGA miner for Bitcoin. Some FPGAs were slower than GPUs, but many were not. The power consumption was typically lower. This unleashed single mining units in the GH/s range. And, again, people began linking multiple units to one mining computer. People gaped over the next two years as the difficulty rate exploded from 200,000 to over 3,000,000. But what happened next really blew minds.
In early 2013 the first chip built specifically for Bitcoin mining was released. This chip was an integrated circuit, but instead of being programmable like the FPGA, it came out of the factory hard wired to do one thing: mine Bitcoin. The age of ASIC, or Application Specific Integrated Circuit, mining had begun. These specialized devices were ultra low on the power consumption scale and they were tiny. Single mining units that could plug into a USB port on the back of your computer could typically beat out any GPU. People began buying up to 50 port USB hubs just to plug them all in. But ASICs kept getting faster and faster until single units were multiple GH/s themselves. Over the past three years the difficulty has gone from over 3,000,000 to over 120,000,000,000. That’s 120 billion if you don’t want to count the zeroes.
Right now, as of writing, the power of the Bitcoin network is more powerful than the fastest supercomputer in the world. It is so powerful you’d have to have over 300,000 of that computer to beat the Bitcoin network. In fact, if you took the top 500 supercomputers in the world and combined them into one network, you’d still have to multiply that network by 43,000 to rival the Bitcoin network. We have ceased rating anything in Megahash. Gigahash is on its way out. The network surpassed Terahash some time ago and recently left Petahash. In 01/2016 the Bitcoin network officially crossed over into Exahash. That means that the network can do over one quintillion hashes per second. For a lovely representation of all this data in an easy to understand graph, go here and click on the “All” at the top.
Now that you get an idea of the scale we are dealing with, let’s talk about mining as a money making operation. There are several factors you need to consider when deciding whether or not you’re going to mine Bitcoin:
- Hardware specs and price.
- Electricity cost including cooling.
- Difficulty and how fast it’s going to go up.
- Price of Bitcoin and its stability.
There is a lot of mining hardware out there to be had. Much of it is used, but the older it is the less viable it is. So, instead of wasting our time looking at old hardware you may be able to pick up at a flea market or garage sale, let’s look at three examples of miners that are currently viable as of time of writing and run them through the checklist above.
Spondoolies-Tech SP-20 Jackson is a good entry level Bitcoin miner at this point. It runs at between 1.3 and 1.7 TH/s, which is respectable for the price tag of around $360. I will have to supply it with a continuous 1152 Watts of power.
Bitmain Antminer S7 is a big step up on the power scale, hashing at a steady 4.73 TH/s, but it’s also a big step up in price at roughly $1500. Again, I’ll have to supply it with a continuous 1293 Watts of power.
Bitmain Antminer S5+ weighs in with a hashing power of 7.72 TH/s, and a price tag of about $3300. And then there’s the staggering 3436 Watts of power consumption to take into account.
So, now that I have my units and important stats, let’s take a look at what it will take to fire them up. First, I’ll need to grab a power bill to see how much I pay for power in my area. Then I need to pull up a handy Bitcoin Mining Calculator. I’ve chosen the one at CoinWarz, but you’re free to grab whichever one you think is best. Also, I’m going to try and make this as comparable as possible, so I’m going to even the playing field as best I can by trying to level the hash rates. This means I’m going to say that I need to buy 5 of the SP-20 (4.5 SP-20s would be 1 S5+) and 2 of the S7 (1.6 S7s would be 1 S5+).
I’ll start with the Spondoolies-Tech SP-20. Buying 5 of them would bring the hash rate to 8.5 TH/s, so I put 8500 in the GH/s box. In the power box I need to enter 5760. I do believe I pay $0.14 for power. I’m going to leave pool fees at 0% for now because that’ll vary between whichever pool I choose to join, if I join one at all. The difficulty is current, as is the block reward and the price, so I’ll leave those alone. And in price I’ll need to put in $1800 for the 5 units plus 5 power supplies that can handle 1152 Watts each, so I’ll add another $900 for power supplies. We’ll assume that I have all the cables I’m going to need for simplicity sake. That’s $2700 just to get up and running. Now I click calculate.
I scroll down to the Estimated Expected Bitcoin Earnings. If I run this unit for a year I’m estimated to gain roughly 12 Bitcoin. At the time of writing that’s roughly $4900. That sounds great, but then I have to realize that it’s going to cost me $7000 in electricity at the rate I’m paying. I would be paying $2100 just to run these miners over the course of a year. Another hidden cost is that I’d also have to run my air conditioner to cool whichever room in which I place that miner. So I’m looking at losing $4800.
Boy, that doesn’t sound like a very good deal. The other miners can’t be that way, right? They’re newer! The 2 S7s would be 9.46 TH/s or 9460 GH/s. I need to put in 2586 for power. The price would be $3000 plus $360 for power supplies. Hitting calculate shows that, over the course of a year, I’m likely to end up with 14.5 Bitcoin, or about $5400. Because these are more efficient I’m only going to pay out about $3200 in power. That means, if I take out the $3360 I paid for the equipment, I’m only out about $1200.
The single S5+ would be 7.72 TH/s or 7720 GH/s. The price would be $3300 plus $360 for power supplies. Hitting calculate shows that, over the course of a year, I’m likely to end up with 11.8 Bitcoin, or about $4400. But I’d also be paying out $4200 in power. Which means, minus the cost of the hardware, I’d be out about $3200.
Any miner I choose above is going to lose me money in the long run, whether it’s cost or power. So, if mining is a losing proposition, why in the world do people do it? Well, remember that Bitcoin is global. It has also become big business to mine. There are some areas of the world that get free electricity. There are some areas of the world that don’t have to pay to cool large data centers. There are some areas of the world that manufacture the hardware and thus don’t pay duties, tariffs, taxes, shipping, and other costs to receive the hardware. There are some companies that can pile together capital, presale order hardware, and do so in bulk. This means they can often receive a significant discount. And being first in line to get the hardware means you can get them fired up first causing the big difficulty jump, not reacting to it. The more of these advantages you can pile together at once, the more profitable your mining operation will become.
But, if you absolutely want to mine. If it’s an itch you just can’t get past, if you’re feeling altruistic and want to spend that money helping to secure the Bitcoin network, if it’s something on your bucket list, and you simply cannot do without mining in your lifetime, let’s look at what you should do to make the best of it.
Go ahead and spend the money buying any of the mining packages I listed above. One thing you can do is solo mine. You probably won’t even be a blip on the radar, considering 9 TH/s is about 1/100,000 of the network, or 0.00001. That’s tiny, but you’ll be keeping that power out of the hands of centralized mining pools. Or, speaking of mining pools, you could join one of the smaller mining pools to give them a bigger stake against the larger pools. Again, this is keeping power out of the hands of bigger pools and balancing the network. Either way, you’re supporting decentralization, but there is a cost to that.
The short answer to the question of “Should I Mine Bitcoin?” is no. You, as a small scale, home hobby miner, with limited capital, an electricity bill, should not invest in Bitcoin mining. My advice to you is to take that $4000 and buy the roughly 10.6 Bitcoin you could get with it at the time of writing. You should then secure it properly and sit on it for the foreseeable future.