PayPal Announces Bitcoin Support

35 Comments

View All Comments

• JarredWalton - Wednesday, September 24, 2014 - link

  Securing the coins and preventing people from stealing others' coins is certainly a "useful purpose" if you happen to use Bitcoins. How much energy does the backing of the Bitcoin network use in a day? Let's call it roughly 6000 MWh (though it might be up to twice that I suppose). According to this page (http://www.eia.gov/electricity/), the US uses about sixty thousand times that much power in a day! And the world as a whole uses about 250,000 times that much power in a day.
  
  How much of that power is going to the millions of desktops sitting idle at corporations during the night? How much of that goes to powering servers used for financial institutions? What about all the air conditioning, radio, lights, etc. that we don't really need?
  
  There are so many things that use power, and how "useful" they are is a highly subjective question. The reason things use power is that someone, somewhere has the money pay for the power and they feel it's in their best interest to do so. If we were to shut down the entire Bitcoin network the total power use of the world would drop by 0.0004%.
  
  Put another way, the "average" American home uses about 30 kWh per day of electricity, or roughly 1250 W power draw 24/7. To achieve the same 0.0004% savings a home would need to cut their power use by 0.005W. So yeah, let's tax the hell out of it so that we can save almost no power in the grand scheme of things.

• JarredWalton - Wednesday, September 24, 2014 - link

  FYI: math error. I think I'm off by a factor of 30 on the US power comparison. Even then, we're still looking at a relatively small percentage of total power use.

• garadante - Wednesday, September 24, 2014 - link

  You have to remember that even making it look small by throwing around seemingly large numbers, that's still an exorbitant amount of energy wasted, for what? Bitcoin does what tangible good? It's effectively a hype machine that's currently a scam to get people to buy into it so original miners can profit. Whether there's any long term future for it is completely in the air and until there -is- any certainty, resources invested into it are wasted. Think of how many petaflops in supercomputing that energy could have otherwise been put to use for? What sort of scientific advances could have been made that -will- have tangible benefits? Simulations for ongoing research into TWRs that could give us cheap, plentiful, clean, safe energy for centuries to come, for instance. And that's one example of infinite examples. If environmental arguments sway you, think of how many gallons of fuel/tons of coal/trees that much energy could've preserved? Again, for what? An intangible bubble driven purely by hype and empty promises.

• garadante - Wednesday, September 24, 2014 - link

  Example: According to Wikipedia, the Titan supercomputer uses 8.2 MW for 17.8 petaFLOPS in LINPACK. That means it uses about 200 MWh a day. Your estimate of 6000 MWh would run 30 Titan supercomputers. That is an astounding number. I just found a year old article for Forbes saying combined computing power for Bitcoin has surpassed 256 times that of the top 500 supercomputers -combined-... Even if those numbers have shifted in the last 10 months, which no doubt they have, we're still talking likely 2 orders of magnitude more computing power for Bitcoin than all supercomputers combined. Can you imagine how ecstatic the scientific community would be to have access to 2 orders of magnitude more computational power for their research? Can you imagine the field advancing breakthroughs that would come from that? But nope. Instead, we have Bitcoin! Woo!

• garadante - Wednesday, September 24, 2014 - link

  Just checked the bitcoinwatch website and I see the global petaflop rate for Bitcoins has increased 5 fold since that Forbes article. Which means It's more like 3 orders of magnitude more computing power for Bitcoins than the world's supercomputers combined. I just have such an immense disgust for society right now. A market cap of only $5.6 billion in Bitcoin value -or- over 1000 times more computational power for science. That's a no brainer if there ever was one.

• yefi - Wednesday, September 24, 2014 - link

  Hashrate is 250,000,000 GH/s. An ASIC miner takes about 1-2W per GH/s, so that would be 6000-12000 MWh / day.
  
  "Can you imagine the field advancing breakthroughs that would come from that?"
  
  Not much unless it involves inverting a hash. All of this computational power is built on ASICS which can only perform one specific task. Comparing them to supercomputers, they're not really equivalent. It's like comparing the flops of a GPU to a CPU and surmising that your something-GTX is 1000x faster than your i-something. They do different things.
  
  Incidentally, proof-of-stake involves a less resource-intensive way for processing transactions, though I guess as you see cryptocurrency as nothing more than a scam, an intangible bubble and a source of disgust, any form of power consumption above 0 watts would be unjustified. :)

• garadante - Thursday, September 25, 2014 - link

  Still, 6000-12000 MWh is 300-600 Titan supercomputers which is a good chunk of the world's sueprcomputing power. Titan is 17.8 petaFLOPS, according to Forbes the combined petaFLOPS for the TOP 500 supercomputers was 0.250 exaFLOPS in late November 2013 (which I believe would be the TOP 500 list Titan was on) then Titan is effectively at most 7% of the world's supercomputing power, at least probably somewhere close to that. Even a highly conservative estimate of 3% (likely higher as most of that 0.250 exaFLOPS comes from the top ranking supercomputers which means anything outside of the TOP 500 contributes negligibly) means 300 addition Titans would be a 9 fold increase in the world's available supercomputing. Using 600 additional Titans as well as 7% would be 42 fold supercomputing power.
  
  While your argument regarding ASICs is completely true I'd considered that and I wonder if there aren't common calculations run by supercomputers that could be offloaded onto ASICs. Certain types of calculations common in certain fields or perhaps many fields. Bitcoins having a market cap of $5.6 billion means companies developing ASICs for Bitcoin mining likely have a relatively low realistic revenue selling their ASICs in the millions of dollars range. Whether that's 10s or 100s I'd have no idea, this is all guessing on my part. But they decided that was profitable enough to spend the time developing ASICs for Bitcoin mining. If you could put a monetary value on the technological advancements coming from supercomputer results, considering how field changing and life changing they can realistically be (considering they're answering questions for developing cleaner energy, understanding of the very nature of our universe, weather modeling, more efficient cars/planes/etc, pursuing cures for diseases, etc, etc, etc so many times) in the long run it'd be worth a whole lot more than that $5.6 billion. But that's the whole point about this. Long run. Bitcoin was huge because it was seen as a way to make a quick buck, back when mining was easier. Mining got harder, ASICs companies thought they could profit off the swift transition to ASICs. Key point being swift. No waiting for cures for cancer to pan out, even if the tangible benefits are infinitely more useful.
  
  So at the end of the day all I see is a huge amount of waste electricity, computing power, and ASIC developing potential gone to waste to pursue what? A virtual currency independent of any government? A noble cause, if naive in many ways, and almost impossible to execute in the real world. Bitcoin is worthless without having the exchanges for most people. The only reason it became so popular is because of the exchange rate into real, physically ownable currency and how that exchange rate ballooned as more and more people saw the surge in exchange rates. Bitcoin derives its value from standard, government backed currencies. I suppose you could argue that some people would still value it without that exchange rate but that's the whole point. That group would be a very, very small subset of the whole population giving Bitcoin no real ability to succeed long term. But discussing the real and imagined benefits and problems with Bitcoins is an entirely separate discussion. My disgust for the wasted power still stands. Any virtual currency that promotes such a blatant waste of precious resources sits ill with me. Jarred mentioned something like Curecoin. That sounds as if its coin mining + some sort of folding and that sounds interesting. People getting paid for all those folding points! That seems far more interesting than Bitcoin with its otherwise completely worthless hashing computations.

• JarredWalton - Thursday, September 25, 2014 - link

  The issue with the work that Bitcoin -- or any other cryptocurrency -- uses is that you need something fundamentally secure if you're going to use it as a means of preventing duplicate transfers, among other things. Basically, the solution to every single block needs to be effectively random, and that's definitely not the case with most scientific calculations.
  
  There may be ways to somehow get that sort of distribution of "solutions" with real-world problems, but take protein folding as an example: the computations are far more general in nature than SHA256 hashing, which means making ASICs is more difficult, which means everything moves back to CPUs and GPUs, which means we're effectively back to where we started with Folding@Home.
  
  Yes, Bitcoin "wastes" energy that could potentially be used for noble and good causes. But even if Bitcoin stopped wasting energy (i.e. disappeared), the energy saved doesn't inherently get put into doing something useful. I suspect if we look at the world as a whole, you could argue that upwards of half of the energy we as humans consume is "wasted" -- lights and electronics left on when no one is around using them, inefficient power conversions, etc.

• whyso - Thursday, September 25, 2014 - link

  Thats no reason to create new ways to waste power. The world needs to work on ways to reduce its use of resources. I'm not a left wing environmentalist but conservation will be key moving into a future where the entire world (not just the small proportion of 1st world countries) can live a comfortable life.

• The Von Matrices - Friday, September 26, 2014 - link

  As much as I like Bitcoin, it brings to the forefront about a sad truth about the world. The only reason that such computational power and the resulting waste of energy is needed is because humans are inherently untrustworthy. The large computational power is only there to present a barrier to anyone wanting to falsify transactions. There would be no need for a proof of work system or the wasted computational power (and energy) if we all were able to trust each other to not falsify records.