That was years ago before the rise of mobile and power consumption became such a priority, but I expect that just shifts the curve a bit without really changing the general shape of it.
For the high-end consumer desktop, that's mostly just resulted in idle power becoming negligible while full power remains the same. Well, I guess AMD doesn't have >100 W TDPs anymore, but the 95 watt category for the top offerings has been standard for quite some time now.
Too many people see value in it for it to be going anywhere. You guys heard about the Bitcoin futures market that started this week right?
The one that's entirely priced and settled in USD based upon the auction price on Gemini? http://cfe.cboe.com/cfe-products/xbt-cb ... ifications
See, the only way that touches bitcoin demand is if someone starts deliberately gaming the Gemini auction price to manipulate the Cboe: This isn't far-fetched, it was done to chicken prices and even fricking LIBOR! Gemini, a very low volume exchange, makes it trivial (it was chosen only because it's the only US-based one basically)
Captain Ned wrote:
The whole concept of the distributed ledger, where every market participant holds a definitive and unalterable copy of the entire system record (however that may be defined by the market), is just too technologically "sweet" to not become a part of the fintech landscape. Operationally, it eliminates a vast swath of back-office workers whose sole task today is to ensure that the single definitive record is properly maintained.
But that's only part of the "crypto-currency" concept, right?
I mean, we have "blockchain" filesystems already: ZFS. And we have all sorts of competing distributed databases/filesystems too.
Bitcoin's novel insight was solving the double-spend problem, it's what Satoshi even said:
The Holy Whitepaper wrote:
In this paper, we propose a solution to the double-spending problem using a peer-to-peer distributed timestamp server to generate computational proof of the chronological order of transactions
You're more interested in a provably unaltered document trail, which a mere tree/chain of hashes can provide. You don't need an escalating proof-of-work contest to confer the privilege of adding another record.
Captain Ned wrote:
Woe unto someone who ends up having to take delivery on that futures contract in a market that can't sufficiently process the underlying transactions that give rise to the derivative pseudo-asset (the futures contract). A futures market in an illiquid asset is a mug's game, and I don't want to be Clarence Beeks (or the Duke brothers for that matter).
Nothing to deliver:" It's purely a USD denominated bet on the current price.
If it is true that most mining is occurring in China, then yeah it is probably mostly powered by coal. But I'm not sure we know that for certain.
It is difficult to say, because while we know the mining pool that mines a block (and the majority of mining pools are chinese), we don't know the geographical location of the participants.
There have been credible attempts to guess it: https://www.jbs.cam.ac.uk/fileadmin/use ... -study.pdf
They say it's 50%, but it very well might be more.
Even then, it is known that there are large mining operations powered by (currently) under-capacity hydroelectricity projects, so it's clearly not all coal in China either.
I still take issue with the assertion that blockchain is somehow inherently "too complex" though; it's certainly less complex than many other things that power our digitally connected lives, and the underlying mathematics/cryptography is well-defined and well-understood.
Yeah, it's a really simple data structure and the basic concept is used (with more actually complexity, not chains but trees) in plenty of commonly used already existing things: ZFS completely pre-dates bitcoin, for one example.