Inside one of the world’s largest Bitcoin mines
qz.com
qz.com
However, I always find it sad to see how mining is depicted in most media: as a way to "generate" bitcoins. This view is inaccurate for two reasons:
1. The purpose of this whole computation power is to secure the Bitcoin transactions. The mined Bitcoins are "just" an incentive (i.e. a desired side effect).
2. In addition to the mined Bitcoins, the miners receive the transaction costs. In the future, the mined Bitcoins will become a lesser and lesser part of the incentive, gradually replaced by the transaction costs.
Luckily, this one seems to be an exception. Although the title is misleading, the process is stated mostly correct (except for not mentioning transaction fees). From the related article:
"The lives of bitcoin miners digging for digital gold in Inner Mongolia"
https://qz.com/1054805/what-its-like-working-at-a-sprawling-...
Today, Ordos (population 2 million) has emerged as a center of bitcoin mining, the process of approving transactions and creating new coins in the digital currency’s system
Essentially, for an article that is aimed to such a wide audience you need to find an appropriate simplification. Such a simplification will never be completely accurate. Therefore, I consider anything appropriate that is technically not wrong and does not create more confusion. For example, describing mining as a process of generating bitcoins would probably satisfy those requirements (ignoring transactions fees).
I agree that it would be nice to be able to explain what the real purpose of mining is, but I don't see a really good way to do this.
All from coal-fired plants.
Surely there's a better form of "currency" that's not so wasteful?
https://cointelegraph.com/news/why-blockchain-needs-proof-of...
[1] https://qzprod.files.wordpress.com/2017/08/bitmain_127.jpg
"Bitcoin Boiler"
I've been working on a design to cool my cpu with one. The best part about peltier modules is also the worst part: the cold side gets really cold. So cold you get below the humidity threshold, and end up with water all over your system. I'm trying to hook up a bunch of humidity, and temperature sensors to an arduino that will regulate the power to the peltier module to try to only let it get as cool as the relative humidity will safely allow.
What I'm getting at is you could do it, but a peltier cooler would need to be manufactured for it to be reasonable at the scale they're looking for.
Separate from the condensation problem they use a LOT of power. They're not terribly efficient, which would likely defeat the purpose. ( https://en.wikipedia.org/wiki/Thermoelectric_cooling#Perform... )
Nowadays people in that department seem to have moved to the elegant solution of manually pouring LN2 on to a cup that sits on top of the CPU (or GPU).
How about water cooling?