> Of course the value of Bitcoins could raise dramatically to make it profitable again
If the value didn't go up people would stop mining. If people stopped mining the difficulty level would go down.
So basically the value of bitcoins will always stabilize at the cost of electricity for the most efficient rig.
But the efficiency of rigs changes constantly, but that is offset by the difficulty changing. So I'm not quite sure how the balance will stay longterm.
But in the short term the value is pegged to the cost of energy.
It's not the value of bitcoins that stabilises but the difficulty of producing bitcoins (measured in average number of trial hashes per BTC). Given:
D = difficulty (HASH/BTC)
E = marginal efficiency (HASH/kWH)
V = value of bitcoins (USD/BTC)
C = cost of electricity (USD/kWH)
then in equlibrium, the difficulty will be equal such that the electricity you can purchase with a generated bitcoin will just balance out the electricity required to produce it (neglecting depreciation of equipment, time value of money etc..):
D ~= (E * V) / C
So this predicts that equilibrium difficulty will increase when marginal mining efficiency increases (ie. more power-efficient GPUs are available), when the value of bitcoins increases, and when the price of electricity decreases.
(Note that there is no guarantee that the system will reach equilibrium - the relatively long time lag taken for the difficulty to adjust in the protocol may mean it doesn't).
It is also notable that you can rearrange to solve for `V`, allowing you to determine an estimate of the underlying theoretical value of bitcoins based on the current difficulty level, the efficiency of the best GPU now available, and the price of electricity.
Some people (eg, college students) have access to unmetered electricity, but we don't see such people dominating current mining efforts (AFAIK). I'm guessing this is because of the cost of hardware, so I think your model should also take into account the amortized cost of buying GPUs - though I don't know how significant that would be compared to the costs of electricity.
That is, of course, left as an exercise for the reader ;)
(As long as the number of students with unmetered power partipating is small compared to the number of miners, it shouldn't significantly affect the analysis. It's the marginal cost faced by a new entrant that's of interest - a relatively small number of students mining simply adds a background base level of difficulty).
It is the dominating cost for students with unmetered electricity, so it's probably not negligible. (Well, maybe it's only the second most important cost for them, after sleep loss due to fan noise in the dorm room, but that is difficult to put a monetary value on.)
Well, to take it one step further, price differences in electricity costs could play a role too. Also people subsidizing electrical costs from various sources (office for example).
Yea, due to the way my apartment complex charges for electricity I end up paying less than .2% of my personal energy bills. (It's total energy for the split among ~500 units). Because, I also have fairly nice graphics card so I could mine bit coins for basically zero cost to me other than setup time. However assuming the return is basically the same as your energy costs ~.3kw * .08c/kwh * 24 * 365 ~= 200$/year which is worth me setting it up and having slightly more fan noise for a year.
PS: I would also feel somewhat guilty wasting that much energy for so little net gain.
So theoretically, being in the midwest with super cheap electricity, I should always be able to make money because my cost of power is about three times cheaper than a place like California?
Depends on whether you are trying to run a mining rig out of your home, or a data center that you have pass-through rates on electricity. Nobody in their right mind runs a set of mining rigs out of California out of their home at scale, particularly out of their home where they are usually on an E-1 Rate schedule at $0.34/kWh pretty quickly. http://www.pge.com/tariffs/tm2/pdf/ELEC_SCHEDS_E-1.pdf
So, running a 42U Rack of mining gear out of your data center would cost you 500 Watts * 42 * 24*30 =15,120 kwH or $1512/electricity month. It would cost you $4500/month to do the same thing out of your home.
i dont totally agree since there are people that get electricity for free and others who are just in it for the fun. Think of folding@home or seti@home which both were very popular without ANY financial incentive thus i reckon mining will be on the edge of profitability but more on the unprofitable side.
Adding to that that in germany for example i pay .3USD per kWh so i am out of the game much sooner than others.
If the value didn't go up people would stop mining. If people stopped mining the difficulty level would go down.
So basically the value of bitcoins will always stabilize at the cost of electricity for the most efficient rig.
But the efficiency of rigs changes constantly, but that is offset by the difficulty changing. So I'm not quite sure how the balance will stay longterm.
But in the short term the value is pegged to the cost of energy.