Hacker News Reader

Comment by makomk | Hacker News Reader

The part about stolen electricity is not actually true, though, at least the way it's justified there. The paper argues that botnets will drive out honest mining, but at this point they can't can't because custom-made hardware is so much faster at mining than general PCs. The current global hashrate is currently about 10 petahashes/sec and your average computer can manage around 10 megahashes/sec. Doing the maths, in order to achieve that through a botnet and drive out all honest mining you'd need a billion computers - that's roughly every PC in the world, running 24/7!

(Computers with high-end AMD cards can do better - up to 1.2 gigahashes/sec on AMD's top-of-the-range card but they're also a lot rarer so shouldn't drive up the average much.)

View on HN

You don't need to drive out all honest mining. All you need is not to have to pay for any costs.

The issue he's pointing out is that the rational, amoral actor will have to decide between a saturated market of ASICs with extremely long time to ROI, and a quick buck off of grandma's draining laptop battery.

The problem is that the difficult rate increases based on the hashing power put into the network, and the reward for mining is cut in half as more coins are mined until there are none left.

http://www.bitcoindifficulty.com/

You'd need a LOT of regular PCs to match even one simple asic. A butterfly lab's jalapeno does ~5 gigahash/sec. You'd have to infest 500 decent PCs to achieve that, and then wait about a month for about a tenth of a bitcoin.

So to get 1 btc in a month, you'd need to infect 5k PCs without them noticing that "damn, my machine is really pegged" and "golly, this laptop is awfully hot and my battery dies really fast...".

It has happened, but realistically you could just buy a miner and be done with it. https://krebsonsecurity.com/2013/07/botcoin-bitcoin-mining-b...

They were able to get ~240 machines, at a time when CPU mining was making them ~$350/day. Decent, but worth all the time to write the software and distribute it? Eh, probably not.

A pegged CPU or GPU is pretty noticeable. Illicit mining has to be the best use of the access (instead of say, stealing $0.01 from an ad network or whatever).

> The issue he's pointing out is that the rational, amoral actor will have to decide between a saturated market of ASICs with extremely long time to ROI, and a quick buck off of grandma's draining laptop battery.

The rational, amoral actor with access to a botnet of general-purpose machines will have to decide between using them to mine for BTC and using them for another profitable, illegal purpose, such as relaying spam, performing DDOS attacks etc.

As the difficulty of mining increases, the profitability of mining on a botnet decreases relative to the other potential uses for the botnet.

Currently, I would be surprised if mining for BTC on home computers would yield enough coin to pay even for the cost of building the botnet in the first place.

This is correct. Last friday one of our LAMP servers was running a litecoin miner via a security vulernability. Everyone laughed at how inefficient that would be for the hacker. Well, the hacker wrote a script and probably got tens of thousands of high powered VMs to do his bidding. Free mining for him. Even if its crazy inefficient, 10k inefficient machines are better than spending $5,000 on a single mining rig. Whatever he makes via this approach is pure profit.

Note: if you run Pydio, update to the newest version. Everything below 5.04 has a serious security vulnerability. That's how they got into our machine.

Litecoin != Bitcoin. My argument doesn't apply to Litecoin, because by design it doesn't have the same vast disparity between the performance of the best available mining hardware and your average PC. In fact, if anything Litecoin is another reason why botnet miners are unlikely to drive out professional Bitcoin mining - it's much more profitable to mine Litecoin.

They're both nearly identical cryptocurrencies with a lot of the very same problems.

> it's much more profitable to mine Litecoin.

Only because its younger than bitcoin. Making bitcoins was trivial a couple years ago. Considering both have a maximum number of coins, this dicussion is academic. Once that max is hit then its just traded as any ugly market commodity.

Litecoin uses scrypt, a memory-intensive hashing function, whereas BtC uses SHA, which is merely computationally expensive. It was specifically designed to be difficult to mine with ASICs, so as to lessen the advantages of professional ASIC mining farms and give relative power back to all the guys who built GPU mining rigs back when that was cutting-edge.

The unfortunate side-effect of this is that LtC botnets are far more viable relative to the total network hashrate than BtC botnets, as the difference in power between commodity computers and ASICs as far as hashing is concerned is far less for LtC than BtC.

No, litecoins algo (http://www.tarsnap.com/scrypt.html) is easier to mine on a general cpu (see https://litecoin.info/Comparison_between_Litecoin_and_Bitcoi...)

Litecoin is about to be overthrown by dogecoin.

Shows how little this site knows about how normal users perceive currency.

Keep your butts, keep your LTC, and your peer cons. You guys missed the moonboat and now real users are going to dictate the future of currency.

Doge coin has arrived to crash the alt coin market and rise above to save the people from all the greed and scum that has overtaken the bitcoin inindustry. Yes I'm looking at you hacker news. Why do you all just care about money so much?

Yeah why care about your currency? Care about the one I espouse, that is not about money!

I really would like to know how much of the current bitcoin supply was mined by stealing electricity from a public university.

And I don't mean hacked boxens but people (mostly PhD students) working there, running some unused machine for mining under their desk.

> stealing electricity from a public university

If they give you free electricity, is it really "stealing"?

Why is this a prediction for the future, or in other words, why does this appear not to be happening now? We've seen three generations of Bitcoin mining (CPU, GPU, and ASIC), and at all times it appears that parasitic mining has been an extremely small portion of the mining network. And ASICs, capital intense though they may be, have been developed and purchased en masse by your "rational, amoral actors."

Reply on news.ycombinator.com