Ask HN: does bitcoin mining pay off?

[agranig]

I'm curious whether there are people who do (semi-)professional bitcoin mining, for example with hardware like https://en.bitcoin.it/wiki/Mining_hardware_comparison, and whether this really pays off?

In the worst case, the costs for powering these devices is higher than the value of bitcoins mined, plus there is the risk of dropping exchange rates.

If it is somewhat profitable though, then I'm wondering why not more people are doing it, as it seems easy enough? If more people did it, would there be a risk of cannibalizing each others efforts? It might sound like naive questions as I don't know much about the mechanisms of bitcoins.

[chanced]

The reason more people don't do it is because:

A) it's difficult. Since the vast majority of miners didn't get in early, they weren't able to get in on the "gold rush" that enabled them to ramp up hardware in accordance with their output. When mining first started, you could find blocks with a decent CPU. Since then, dedicated rigs have become almost a necessity if you plan to make any kind of money at all.

B) Electrical efficiency. Since the recent decrease (reduction from 50 bitcoin per block down to 25 per block) and the rise in popularity of bitcoin (thus rise in difficulty of finding a block.. it scales naturally), the "cost" of finding blocks has gone up. By that I mean it requires more raw computational power to mine, raising the bar on both the hardware needed to be relevant AND the amount of electricity consumed during the process.

There are a lot of calculators out there. They will give you some insight into how much you could make (or very likely lose) depending on your local electricity pricing.

[jyu]

A year ago, I entertained the idea of mining bitcoin, but with high electricity rates locally, the numbers and breakeven point didn't make sense. Long term edges in mining are based on specialized hardware and low electrical costs.

For me, it only made sense if bitcoin kept increasing in value, at which point it's much easier to just buy and speculate.

[adityaathalye]

I was wondering about BitCoin and global power price arbitrage just today... The market may not be big enough to make it a very lucrative thing, but I still think it'll be a cool experiment to try. Especially since the computing requirements will only get more rigourous. Longer post here: http://adityaathalye.wordpress.com/2013/02/04/bitcoin-and-gl... [deleted previous comment - mistakenly added reply to comment one level higher]

[agracey]

Could you use something like AWS or EC2 to reduce the overhead?

[dustcoin]

No. CPUs are useless for mining, and the only GPU instance AWS offers uses Nvidia GPUs, which are about 5x worse than the equivalent AMD/ATI GPUs. Even if AWS had AMD/ATI GPU instances, the costs would probably be too high.

[chanced]

I honestly don't know. My guess would be no, for two reasons.

GPU vs. CPU; GPUs are much better at the number crunching needed in producing the hashes ("mining"). I doubt very seriously that you can get the kind horsepower out of a vm. I could be wrong; they may have machines dedicated to scientific numeric calcs but then you would likely need to produce a driver for harnessing it, if it even exists.

Computational pricing would likely dwarf any kind of return. Electricity costs alone can be enough to offset profits. I can only imagine what adding AWS fees on top of it would do too.

[steven_h]

It will pay off, but it's more a long term investment and you should an opportunity cost analysis on the cost of hardware and expected returns.

Use this to calculate your profit. http://www.bitcoinx.com/profit/

Disclaimer: I mined bitcoins at the start and cashed out during the last bubble at ~$30USD per coin. I sold the hardware I had (ATI 58xx series graphics card mining rigs). The new ASIC hardware looks too great if you could gt your hands on it.

Remember though that you are at a disadvantage to more established miners who can afford these $X0,000 mining rigs that increase the total pool speed which brings down your contributions and your profit.

[gesman]

If you can run your mining rig somewhere where you don't pay for electricity (like under your employer's desk :) ) - then you'll get some benefits. Otherwise electric bill will eat your profits.

I think it's a bullshit business where opportunity is guaranteed to decline.

Let others mine and I'll develop products and sell them for bitcoins (and dollars).

[cfaulkingham]

I have been looking into it. From what I can tell the only way it would be profitable is with a custom ASIC. Something like http://www.butterflylabs.com/products/

[chanced]

As of right now, I'm still considering BFL vaporware. They've moved that goalpost far too many times for them to be taken seriously.

[redegg]

I've been mining since July 2011 with a small operation of 17 ATI 5830s.

With the amount of ASIC preorders currently speculated, I don't think its worth my time to upgrade. I'll be selling my mining hardware off this year.

[undrcvr]

Don't forget that according to the (non existant) spec there's a finite number of btc. So even if it was profitable (which i doubt), it's guaranteed to stop being viable when every block has been mined.

[speeder]

Except that this will take a while, and then fees will be collected by the miners, and as Bitcoin price rises, fees don't need to be in the integer amounts to be profitable.