The Behemoth Behind Bitcoin Mining
(Originally posted December 02, 2017 on Blogger)
I posted this on my Google+ page a few days ago regarding the ~900% rise in the price of a Bitcoin:
"A year ago a single Bitcoin was worth under $800. Today it's valued over $10,000. There is most definitely a future for blockchain.. in fact I believe it will be a paradigm shift for financial markets worldwide. There is a future for cryptocurrencies, but this rise in Bitcoin's 'value' is markedly a bubble that will eventually burst to one extent or another. I'm not one to give investment advice (I'm only a thousandaire), though did suggest years ago to someone we should invest in Bitcoin. My suggestion was shot down by a third person. Be that as it may, this rapid rise in Bitcoin's value presents a golden opportunity for me to pose a consideration. Bitcoin, like a $100 bill, has no intrinsic value. Worse still (or better depending on your POV), it isn't technically even a fiat money system as it has no government regulation. If a thing has no intrinsic value, and exists in cyberspace unregulated, what then can stabilize it if not trust, mutual agreement, and perhaps some international cooperation? With emphasis on the former. Few can argue a healthy dose of greed with a pinch of risk and a dash of gray-area law bending make a good fuel mix to drive "sustained" high-earning investment income.... But at what point does fear and realization envelope a market's short-sighted hoarde of investors playing before the facade of trust, mutuality, and cooperation cause a thing of no intrinsic value to be exposed for what it isn't? If intelligence has anything to do with it, then it will be sooner rather than later. Fortunately, or not, intellect isn't generally needed or good to have in the world of high-risk, short-term investments. One might think obvious Ponzis like http://bitconnect.com would wake people up to this... but.. well... hard to wake from a greed-induced coma I'd imagine. At any rate, this is an obvious bubble... but that isn't to say the future won't be in blockchain technology. It most certainly will in my humble opinion. Yet I think it's easy to say, 'investors beware' in this current BitBubble. But what do I know."
It's possible the value of a Bitcoin hits $20,000 or more; I have no idea. But I think it's obvious this rapid rise isn't sustainable, nor will the peak be stable wherever that peak will be. This of course is uneducated opinion based on experience with human nature (something we all have experience with obviously).
But this blog isn't about the future of Blockchain, or the rise of any of the cryptocurrencies out there. This blog aims to largely regurgitate things I've been reading since July regarding cryptocurrency mining.
A Bit About Bitcoin
- A Force to be Reckoned With
Consider Bitcoin; a cryptocurrency that must be "mined" to be realized. Mining involves using a computer with special software to solve mathematical problems (cryptographic hash functions) in order to earn Bitcoins, or a single Bitcoin, or a fraction of a Bitcoin.. how much one earns proportionately depends on how much solving their computer accomplished. A simplified explanation of these cryptographic hash functions and why they're important is available here).
With a single Bitcoin worth $10,922.39 at the time of this writing, we might be wondering why everyone isn't mining Bitcoin! The reason is because as more people jump in on the action to try mining Bitcoins for themselves, the Bitcoin network responds by increasing the complexity of the math problems. As the rapidity of math equation solving increases, so too does the equation complexity. It's like a braking system of sorts.
I'm simply paraphrasing other bloggers here, but a home computer that was once able to solve these math problems fairly easily and quickly before, now have a heck of a time trying to solve anything. A home computer can run all year at full capacity and not earn a fraction of a fraction of a Bitcoin for failure to solve a fraction of a fraction of an algorithm.
The computer churns away as the overworked processor heats up at full capacity, and eventually either the computer crashes or catches on fire. Darn you second law of thermodynamics! So to get around this, miners upgrade their computers with fancy gaming graphic cards that run more efficiently. Those graphics cards worked well for some time, but the network responded with more complex algorithms.
Savvy miners began reprogramming computer chips to specifically mine for Bitcoins in order to increase the efficiency (solve faster with less energy) of their mining. That is to say, mine faster with less energy consumption. Of course we all know greed dictates that "faster with less energy consumption" ultimately means "run at full power".
More problems may be solved, but eventually the network increases the complexity of the problems yet again. So we'll need more power! (More efficiency too, so we can still run at full power Scotty!)
Application-Specific Integrated Circuit Chips (ASIC) were made to increase mining efficiency, and this of course encouraged more and more people to jump in on the action, and for those already in on the action to use more and more ASICs to build bigger and bigger computing systems.
The Bitcoin network responded automatically to the increased rate of math-problem solving, just like it's supposed to, by increasing complexity still more... and more... and more... and it will continue to do so.
But we humans are persistent creatures. We respond in kind by building massive warehouse-sized computer systems. Some owned and operated by millionaires, some by regular schmoes who have banded together in large Joe Schmo mining pools (pooled mining). These mega mining monsters are located in mostly undisclosed locations around the world. Here's one which I believe is a pooled mining operation:
...and here's another one:
...and here's another..
The days of getting rich with a laptop are long gone (as far as Bitcoing mining goes).
The photos above hint at just a handful of 'mining farms' being built around the globe. And more are coming, while existing ones continue to expand. With mining farms like these, one can only imagine how complicated the cryptographic hash functions have gotten, and how much more complicated they will continue to become. Added complexity will require bigger and bigger mining farms (we'll get to increasing efficiency shortly).
All this mining is good for the network. It increases the security of the Bitcoin network in that each miner is required to approve Bitcoin transactions. The more miners, the more secure the network. It's a peer-to-peer network, so everything is interconnected.
Unlike cash money used by countries all over the world (except maybe Sweden who appears to be nearly cashless from what I saw when I was there a few months ago), where governments are responsible for printing and distributing paper money as they see economically fit... The Bitcoin network distributes Bitcoin (not a physical thing) via this mining process which is 'regulated' by these increasingly-difficult algorithms.
It's almost like an impending case of an unstoppable force (human greed, er, nature) meets an immovable object (that yet-to-be spawned 'final' cryptographic hash function).
The Dark Side of the Force
Bitcoin popularity in 2017 is going bananas, and this is driving technology to further improve processing efficiency such that more Bitcoins can be earned by miners. Efficiency is a good thing, but when combined with human nature, it begins to self undermine.
Consider Jevons paradox; It "occurs when technological progress increases the efficiency with which a resource is used (reducing the amount necessary for any one use), but the rate of consumption of that resource rises because of increasing demand" (Wikipedia).
The resource being used for mining of course, is electricity. We can continue to improve efficiency, but as usage goes up, so too does consumption despite the improved efficiency. This is has happened before, is happening, and will continue to happen. This is intrinsic to a blog I've been meaning to write for a year now, that I intend to title "There's No Such Thing as Green Energy". Spoiler alert: that blog, if I ever get around to it, will end by saying insofar as energy consumption is concerned, it boils down to individuals using less of it, not blind technological optimism that has faith that ever-improving efficiency will somehow save us. It won't. Not in the long run. Of course, I'm getting slightly off topic, and that last statement really requires a dedicated blog to explain fully.
At any rate, according to bitcoinenergyconsumption.com, as of Friday, December 1, 2017, the amount of electricty being used specifically to mine Bitcoin worldwide is 30.99 TWh (terrawatt hours). For perspective, that's about as much electricity as the country of Bulgaria uses in a year. It's more electricity than is annually used by 159 countries worldwide. We ought to seriously think about that for a moment. Particularly in a world that is becoming more and more environmentally conscious (save for the POTUS). For those interested in learning about climate change and the partial truths surrounding it, I invite you to read my blog on the subject here. It is by far my most read blog.
Anyhoo... here is a graph depicting Bitcoin energy consumption over the past month:
30.99 TWh is mind-boggling. Consider too, that there are many other cryptocurrencies being mined out there. One of them is Ethereum, the worldwide mining of which consumes 10.96 TWh as of December 1, 2017. This alone is as much as or more than 135 countries worldwide.
Together, Bitcoin and Ethereum mining are consuming more energy annually than New Zealand, and about as much as Hong Kong. As these cryptocurrency values increase, so too will their popularity, which in turn drives their 'value', which drives their popularity, and the cascade continues at the benefit of security, but at what environmental cost? I don't have an answer to that, but it's one I am sure will be addressed soon by some aspiring grad student somewhere.
As can be seen in the graph above, popularity and energy consumption appear to be directly proportional. We can expect energy consumption to continue to increase in the coming years, despite improvements in processor efficiency (see Jevons paradox). In addition to this paradox, we should also consider the fact algorithms will continue to become more complex, thereby requiring more time and power to solve... and when it comes to greed, who has time for time?
This energy-consumption monster is growing, and in a world that still heavily relies on burning fossil fuels to access useable energy (no one actually produces energy unless they've found a way around the first law of thermodynamics), cryptocurrency mining might start setting us back. If you're a climate skeptic because people like Bill Nye, or Al Gore suck at explaining it (I agree, they do), and are known to use apocalyptic scare tactics to make points that would have a much bigger impact if they were explained genuinely and in context without claims of 70-foot sea level rises as if such were possible in a human lifetime, then at least consider the pollution costs and the impact they can have on human and environmental health insofar as they might affect our children (or wallets)... because yes, many have become that arrogant to believe the environment is somehow a separate entity from us. But reality is, we're as much a part of the environment as a Bitcoin is a (number) part in the blockchain.
As always, thanks for reading this stuff.