# How does one calculate the profitability of GPU mining?

As GPU mining made CPU mining non-profitable, at some point in time more specialized mining hardwares (FPGA and ASIC) would make GPU mining non-profitable.

When is GPU mining considered not-profitable?

Considering \$1000 rig, \$0.2 per kWh and one year break-even period, at which difficulty does GPU mining become non-profitable?

Or, what is the formula to find out "max non-profitable Mhash" for given parameters like, rig cost, power consumption, investment break-even period, difficulty, exchange rate, etc. (please ignore redundant parameters, and include more parameters if required.)

Or, whats the current "max non-profitable Mhash"?

Or, what is the required current energy efficiency (Mhash/J = millions hashes per joule) for mining to be profitable?

(energy efficiency; 1 joule of energy is 1 watt during 1 second: 1 J = 1 W*s)

For given parameters, Max. Non-profitable Hash Rate, is breakeven point for hash rate after which mining becomes profitable.

Using the Bitcoinx calculator, I found that a \$1,000 rig with a 300 W power draw would need to have 750 MHash/s just to break even on power at \$0.20/kWh at the current difficulty (3,438,909), exchange rate (13.13), and block reward (25).

You'd never break even on the rig itself. Thus why GPU mining is on its way out.

For a more absurd example, I priced out a reasonable rig for ~\$1,405 with three Radeon HD 7970s, the best mining card combo - 2050 MHash/s - on the market at the time of this writing according to the Mining Hardware Comparison chart. Its power draw would be around 850 W, according the MHC and a quick review of the parts. This rig would never be profitable at \$0.20/kWh. Hell, at my electric rate of \$0.07/kWh, it'd take nearly three years just for a return on investment. Prior to the recent halving, this same rig would make a return after just ~260 days.

More proof that GPU mining is on its way out.

• Interestingly enough, having not calcuated my profitability since the halving, I just realized that my own rig isn't profitable anymore :-( – Colin Dean Dec 5 '12 at 15:24
• This is no longer correct, with the new exchange rate of 91USD/btc – Sameer Alibhai Mar 31 '13 at 2:29
• This is true. The parameters from the first example produce a still-profitable system, likewise the second. Perhaps this price spike is breathing some extra life into GPU mining! – Colin Dean Mar 31 '13 at 11:47

In general, GPU uses a lot of power and at the current difficulty doesn't generate that much bitcoins. Even if you have a return of investment of 1 year, look at the current difficulty graph. It is rising rapidly. Also, note that the ASIC devices are still not out, the difficulty will increase a couple times that.

When calculating whether it is profitable to mine, you need to calculate how many coins you can mine in a given period of time and how much money you will spend on electricity. Be pessimistic with your calculations - assume that the difficulty will rise in the future. If the number is still positive, see how long would it take to earn back the cost of your GPU and other supplies. If it is more than a couple months, you are entering the uncertainty waters of the future market.

In short, unless you have very cheap electricity and low startup costs (second hand hardware at a discount price perhaps), chances are you answer will be that the GPU mining won't ever be profitable again, at least when you are talking about block mining. There are alternative uses for a mining hardware, for example vanity address mining which can still be profitable. But remember, always do your math before making an investment.

It is deemed not profitable when you cannot recoup the costs you make with the energy spend (and possibly the costs of the video card). It very much depends on how much the card costs, your energy price, and how you think the difficulty will change in the future. If you can provide

• Can you finish your – Colin Dean Dec 5 '12 at 15:25

I don't really see GPU-mining continuing much after the ASICs "finally" arrive. However, there is one major factor that none of the online calculators properly project. That's the rise of the BTC->USD ratio. This becomes rather speculative (kind of like the stock market), but given that the next having won't be for another 4 years, a significant rise in valuation could once again make GPU-mining profitable for the casual miner. Take a look at some of these charts.

• Its OK 'online calculators' just doing their job. After all they are not speculation or technical analysis tools. :) – vi.su. Dec 7 '12 at 8:00
• "rise in valuation could once again make GPU-mining profitable for the casual miner" - wouldn't it be better for the casual miner to then just speculate in Bitcoins and buy them. Less hassle and same profit/loss exposure. – Guy Apr 14 '14 at 18:38

GPU mining will still be in demand as long as there are scrypt coins like Litecoin that of any value. Scrypt coins are ASIC resistant at this time. However the price of LTC/USD has dropped significantly recently down to just \$2.40 as of this post. There are many alternative crypto coins to mine now a days though...

Here is a great crypto coin comparison table

And here is an even better Mining Hardware Comparison table