66 comments

[ 3.3 ms ] story [ 132 ms ] thread
Does anyone have an idea of the ROI on this? I'm thinking you'd need a massive number of 10 second to mine any meaningful number of BTC on consumer laptops.
I wonder if sbux even knows. Im sure theyve outsourced their wifi by now...
Probably $0 after the backlash, perhaps even net negative.
Not sure about the ROI, but I do know that Coin-Hive (which is being used in this "trick") mines Monero, not Bitcoin.
Even if they did this in all Starbucks locations, I can't see it really making any money. Ten seconds on a laptop CPU rounds to zero. Coinhive has just never made any sense to me in practice. Maybe 5+ years ago, but not today. What a mismatch of effort and timing.
Coinhive seems pretty upfront with this though (to answer "Will this work on my site?":

"Technically yes, economically probably not. If you run a blog that gets 10 visits/day, the payout will be miniscule. For the captcha and shortlinks with a sensible hash goal (1024–16384) you'll need to have a whole lot of users to make this worthwhile."

I do like it as an accessibility friendly captcha system, though.

Coinhash mines monero and it is a business aimed towards hackers who want to monetize websites they have access to.
I've wondered what the ROI would be for me to leave my home office computer mining bitcoin while I'm at work. I know the likelihood of me mining something is tiny, but so is the expected payout when I buy a lotto ticket.

I spent about $20 / year on lotto tickets (basically whenever Powerball gets near a record jackpot). That same amount of money will keep my i5 ThinkPad running for a long time. Am I more likely to mine some Bitcoin or win the lottery?

I am by no means an expert, but my understanding is that a personal computer isn't energy/heat efficient enough to make a profit from mining. It costs more in energy than you would make from the bitcoin. My friends who do some mining have told me the same.

You're competing with computers that are made exclusively for bitcoin and people who are heavily investing their time into it so it makes sense.

It would probably be fine just for trying the system out and seeing how it works before you invest though.

Jeez, you're completely missing the point.

The GP was asking for a comparison the odds/cost of mining a coin with the odds/cost of winning the lottery. Not about "investing," and not about "making a profit."

Basically, which form of gambling is better from a cost/rewards perspective.

My understanding is you are always outspending on energy costs because of the energy inefficiency of personal computers for mining vs. a dedicated machine. So the chance of return would be more or less 0%. More like, "If I pay $5 for a lottery ticket where the max I can win is $4, what are my chances of profiting?"

But again, not an expert. Just the articles I've read and people I know who mine bitcoin say you can't profit that way.

In my understanding, there is continuously a chance (when mining solo) that you mine a Bitcoin block. If that happens (unlikely with consumer hardware) you make a lot of money at once since you get a whole block.
> "If I pay $5 for a lottery ticket where the max I can win is $4, what are my chances of profiting?"

If you use a mining pool then you won't make a profit on a laptop. However if you were solo mining on your laptop, then there is a very small chance that you could win seriously big moola.

Very false. You appear to not understand how mining works.

The reason people join pools is to make money even if you never personally mine a coin. In a pool you just get paid for your effort. This is why joining a pool with consumer hardware will always lose.

In this case we aren't talking about pools!

There is two things that can happen when you solo mine - you either "lose it all" (all money spent on electricity down the drain) or "hit the jackpot," (1 full Bitcoin all for you!)

At that level with a small hash rate its gambling, you're throwing some amount of money at it at the very, very tiny chance of a big payout.

Say, for example, there's a 1/10000 chance any one person will mine a coin on consumer hardware per every $100 in electricity. So you're buying a $100 lottery ticket where the payout is $17,000.

Of course you "can't make money like that," the vast majority of people will lose money, just like the majority of lottery players will lose money.

With the lottery, you are at least partially funding public education (35% of the revenue if I remember correctly). Are there any social benefits to Bitcoin mining?
I was mistaken then. What I've read about it was somewhat incomplete from what people are saying here.

That said there is absolutely no need to be such a jerk about it.

I think that the current reward for mining a block is 12.5 Bitcoin (+ transaction fees of course)
(comment deleted)
>Just the articles I've read

I strongly recommend reading the Bitcoin white paper: https://bitcoin.org/bitcoin.pdf It is not as nearly as long and complicated as you would think, if you know the basics (what's a hash, how do public keys and digital signatures work). I found it way clearer than all the explanation articles I've seen on the internet.

Thank you, that's quite useful! I was doing some light reading on it so I just started with some simple articles, but perhaps they were over simplifying.
(comment deleted)
I don't think anyone mines for bitcoins by themselves anymore - they join a pool so you get a small guaranteed payout.
I have no interest in a small guaranteed payout.

I'm asking if I have $20 to buy Powerball tickets (with a chance to win $1 billion) or $20 to buy electricity for my i5-based mining machine, how do the odds compare?

I'm not sure what the exact house edge is on Powerball. But let's say it's 20%.

Due to dynamic difficulty adjustments, the profit for miners is just a bit more than the cost of electricity. However, this is only true for those who have the best specialized mining equipment and the cheapest electricity in the world. For an average person living in a typical area, even with dedicated equipment, you'd probably be far under breaking even. Most likely much less than 20%.

And that's with specialized equipment (costing thousands of dollars). With your home computer, it'll be hundreds of thousands of times less efficient per watt of electricity to mine bitcoin. Even for other cryptocurrencies that might be resistant to specialized equipment, you'd still most likely be over 10x less efficient.

Go with the powerball tickets.

You can still mine yourself and that might appeal to someone who likes buying lottery tickets. Mining yourself at least has the possibility of positive ROI as opposed to a guarantee of losing money.
I didn't say you can't do it - I said people don't really do it anymore. You can easily get a positive roi if you do it correctly, why do you think there are literal crypto factories built to mine coins?
not very good odds on your thinkpad..

but if you set up a handful (3-5) of high end ASIC miners and mine BCH (10x more difficulty/hashrate on BTC right now) without a pool for a year, there is a realistic chance you'll get lucky and mine a block (worth about $17k).

The numbers are changing all the time, but when I last did that calculation I was surprised.

It makes no economic sense at all.

Mining Bitcoin using CPUs is unprofitable. They're basically gaining cents worth of bitcoins using dollars worth of electricity. See https://bitcoin.stackexchange.com/questions/41276/in-the-asi...

Are they paying for that electricity?
Sort of, I'm assuming a lot of people bring their chargers with them if they're using the WiFi at Starbucks
I suppose the cost benefit depends on whether they stay in the store charging their laptops longer because of that 10 second delay. If the customer plugs in their laptop at 1pm and plans to leave at 1:30pm, then at least they'd be making money 10 seconds longer while spending the same amount of money on electricity. If the customer leaves at 1:30:10, then they'd be paying extra for the time. But in any case, they're not paying for any extra hardware.
Coinhive estimates that a user with a mid-range laptop would have a hashrate of 30 h/s.

10 minutes of mining = 30 * 10 * 60 = 18,000 hashes

Since Coinhive pays out 0.000090 XMR per million hashes (current rate as of Dec 11. 2017), that's $0.0004250394 per user (18000/1000000 * 0.000090 * 262.37).

In order to make $1, you would need 2,353 visitors with laptops mining for 10 minutes each.

I’m guessing this would work extremely well if it were set up as a honey pot at University? Tempted to test it.
(comment deleted)
The Coinhive script mines Monero: https://coinhive.com/ . It's been a while since JavaScript Bitcoin mining was feasible :)
The unwashed masses can't understand the difference between different coins, so you might as well call everything Bitcoin.
Most of these crypto currency exchange are derived from bitcoin's codebase (or at least from bitcoin's white paper) so from that perspective I don't consider calling these other crypto's as "bitcoin" to be incorrect, just not specific. I could imagine that maybe could use the distinction of capital Bitcoin to refer to the specific blockchain while lowercase bitcoin to refer to any Bitcoin forks.
I really like the idea of this as a captcha alternative, especially with the "train our neural nets" approach google has been taking recently with them.
How would coinhive work as a captcha alternative? There is no human involvement in mining, is there?
It works as a rate-limiter in a way. If you're interested look into HashCash as proposed to combat email spam: https://en.m.wikipedia.org/wiki/Hashcash
rate-limiting is not what captcha does, that's for DDoS protection and even in that scenario I feel HashCash and crypto mining is a very terrible idea.
Hence "captcha alternative" and not "alternative captcha".
Wouldn't "captcha alternative" imply that the said alternative is a replacement for captcha and has similar functionality?
Not necessarily. If the goal is to reduce bot activity, captchas and hashcash are two ways to do that. Captchas are hard for bots to pass, whereas hashcash wastes a fixed amount of CPU time. The result is that robot operators have to invest more effort to abuse your service at scale than they're willing to (hopefully). In this sense, hashcash is a captcha alternative—it's not a captcha, but it helps with the same problem. An alternative captcha might be using something other than reCaptcha. At least, that's how I'd interpret it.
CPU for services seems legit as long as they give an affirmative checkoff.
Are you suggesting it'll become part of the click-through license agreement in future?
I'm picturing going into a cafe, using the wifi, and my phone turning into a hot coal or my laptop fan blowing me back out of my seat.
This is smart...or maybe not. Has anybody done the calculation here? I am very curious to know into how much $s the 10s translate to
Not to mention that some (perhaps many) people will be running their devices off of the store's own power outlets. So the store will actually be paying the electrical bill for at least some of that mining.
One of the big problems I have with this (and with mining, in general) is that it's attempting to create value by simply wasting electricity. Come on, we should not be spending CPU cycles just for the sake of spending CPU cycles...
You may find GridCoin interesting. GridCoin rewards miners for running scientific computations for a variety of projects.

CureCoin & Golem are two other coins that try to use the cpu cyles for something other than heat.

Using this logic, dollars are creating value by wasting paper and metal. What you're ignoring is the value of having a publicly auditable ledger of transactions and a decentralized currency. That _isn't_ a waste of electricity.

There are plenty of reasons to be skeptical of cryptocurrency; this just isn't a good one.

This script is mining monero which uses Cryptonight algorithm. This algorithm kinda famous for being CPU mine-able unlike other cryptocurrencies.

Now, the coinhive page is completely blocked by all the plugins so I cannot find the exact story. But, if memory serves correctly, during the launch devs have presented a story where they use the script on a forum. The results were pretty impressive with payout reaching ~ 1k USD.

Though there is a section which clearly says this is not economical, my take on this was that they wanted to entice enough people to mine using their script.

The devs also promised to look into this kind of abuse but I guess that is not happening.

I had a run-in with CoinHive this weekend so I did a bit of research.

Most modern computers can do about 30/h a second. Coinhive currently pays out 0.00009030 XMR ($0.02 USD) per 1M hashes.

For a 10 second pause, they'd mine 300 hashes (about $0.000006 USD). To make $1 USD, they'd need to have ~166,666.66 people connect to their in store WIFI.

That's plausible given that Fibertel is the #1 provider of free wifi in Argentina. It could well be that some IT guy had the idea, figured it was possible to make a few hundred bucks a month this way, under the radar. Completely not worth it for the company.
30 hashes a second seems extremely low and I'm wondering where they got that number. Are they using unoptimized JavaScript I nstead of ASM.js?
30 hashes a second is what you'd likely see on mid-range laptops. You can test your machine on CoinHive (https://coinhive.com/)

My fully maxed out 2017 rMBP can only do 50-60h/s at full speed with all threads. Most people don't have a machine nearly as powerful as mine.

How much Monero can a company reasonably mine using this tactic?
I bet this was done by a contractor without Starbucks permission.