spiral_mind (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2013, 11:07:49 PM Last edit: August 30, 2013, 11:19:09 PM by spiral_mind |
|
In my other thread someone suggested that since mining is a way many people first get into Bitcoin that an increase in the price of mining each coin could increase Bitcoin market price. I think its definitely a plausible theory. The steady flow of new investors who would normally buy equipment are instead going to buy Bitcoins directly as long as the market price is less than the price per Bitcoin from hardware manufacturers. The real question is whether most miners can afford to hold out for these high break even prices or whether they will sell early (or potentially hold for a long time to double their profits?). The price per BTC represents what price point a miner would have to sell at to make his money back (no profit).
So how much does it cost to mine one Bitcoin? Assuming a 76% increase in difficulty per month, .15$/kWh, and a 2% pool fee:
(I used mining.thegenesisblock.com for these estimates)
BFL Single SC
Power cost over 7 months: ~$280 Initial cost = ~$1300 Total cost over 7 months: ~$1580
For people who get their Single SC 60 GH/s tomorrow: They will mine about 18 coins over the next 7 months. Cost: ~$88 per Bitcoin. For people who get their Single SC 60 GH/s at the beginning of October: They will mine about 10 Bitcoins over the next 7 months. Cost: ~$158 per Bitcoin. For people who get their Single SC at the beginning of November: 5.6 Bitcoins mined over 7 months. Cost: ~$282 per Bitcoin.
Bitfury - 400GH/s
Power cost over 7 months = ~$280 Initial cost = ~$8000 Total cost over 7 months = ~$8,280
Delivery day October 1: 66 Bitcoins mined over next 7 months. Cost: ~$126 per Bitcoin. Delivery day Nov 1 = 37.5 coins generated over 7 months. Cost: ~$220 per Bitcoin
KnCminer - 400GH/s Jupiter
Power cost over 7 months = ~$490 Initial cost = ~$7000 Total cost over 7 months = ~$7490
Delivery October 1: 66 coins mined over 7 months. Cost: ~$114 per Bitcoin Delivery Nov 1: 37.5 coins mined over 7 months. Cost: ~$200 per Bitcoin
HashFast - BabyJet
Power cost over 7 months: ~$259 Initial cost = ~$5600 Total cost over 7 months = ~$5859
Delivery Nov 1: 37.5 coins mined. Cost: ~$156.24 per Bitcoin Delivery Dec 1: 21.3 coins mined over 7 months. Cost: ~$275 per Bitcoin
|
|
|
|
gmaxwell
Staff
Legendary
Offline
Activity: 4242
Merit: 8654
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2013, 11:27:24 PM |
|
So how much does it cost to mine one Bitcoin? Assuming a 76% increase in difficulty per month, .15$/kWh, and a 2% pool fee:
So. 76% per month. Thats 88238% per year. So in 5 years the hashrate will be ~3,765,807,437,718,645,270,000,000,000,000 hashes per second, and with the current 110MH/j miners thats a power consumption of 34,234,613,070,169,502,454,545 watts. This is the energetic equivalent of exploding 8182 gigatonns of TNT every second. But since the sun produces 4e26 watts, I can't see how this isn't totally feasible... we just need to build a solar array with a radius of roughly the same radius of the earth to trail us in orbit and catch light from the sun.
|
|
|
|
spiral_mind (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2013, 11:31:41 PM |
|
So how much does it cost to mine one Bitcoin? Assuming a 76% increase in difficulty per month, .15$/kWh, and a 2% pool fee:
So. 76% per month. Thats 88238% per year. So in 5 years the hashrate will be ~3,765,807,437,718,645,270,000,000,000,000 hashes per second, and with the current 110MH/j miners thats a power consumption of 34,234,613,070,169,502,454,545 watts. This is the energetic equivalent of exploding 8182 gigatonns of TNT every second. But since the sun produces 4e26 watts, I can't see how this isn't totally feasible... we just need to build a solar array with a radius of roughly the same radius of the earth to trail us in orbit and catch light from the sun. Obviously it's not going to keep up forever, and likely won't keep up for a whole year either. For the next half a year or so though? That certainly seems possible at least. There's an insane amount of network power that's already been pre-ordered and just needs to go into production already. Great job wasting your time on straw man calculations though. At least mine were somewhat related to reality (that % is predicted based upon previous difficulty changes observed, I didn't just invent it to make a point like your numbers).
|
|
|
|
gmaxwell
Staff
Legendary
Offline
Activity: 4242
Merit: 8654
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2013, 11:36:11 PM |
|
Obviously it's not going to keep up forever. For the next half a year or so though? That certainly seems possible at least. There's an insane amount of network power that's already been pre-ordered and just needs to go into production already.
Of course, but once you recognize whats creating the growth you realize that it's not actually an exponential process— it's not like more hashpower begats more hashpower like breeding rabbits. Hashpower comes on line in fits and starts as chips come off assembly lines. In the long term I like to think about mining as a process that decays words low profitability for the most power efficient miners... but right now supply is manufacturing side limited, not competition limited (relative to power costs mining is insanely profitable still)... you can pick any result you want just by picking parameters and just about any result is equally (in)sane. Trying to predict price from difficulty has failed for everyone historically— no shocker because it makes no sense to do so. Difficulty has closed loop control from the system, it'll be whatever it needs to be to control the blockrate. When difficulty makes mining not profitable enough people stop mining and the difficulty drops. More than half the coin that will ever exist is already in circulation, miners are in little position to dictate market prices to improve their profitability, but in a lot of position to turn on and off.
|
|
|
|
spiral_mind (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2013, 11:55:14 PM Last edit: August 31, 2013, 06:14:58 AM by spiral_mind |
|
Obviously it's not going to keep up forever. For the next half a year or so though? That certainly seems possible at least. There's an insane amount of network power that's already been pre-ordered and just needs to go into production already.
Trying to predict price from difficulty has failed for everyone historically— no shocker because it makes no sense to do so. Difficulty has closed loop control from the system, it'll be whatever it needs to be to control the blockrate. When difficulty makes mining not profitable enough people stop mining and the difficulty drops. More than half the coin that will ever exist is already in circulation, miners are in little position to dictate market prices to improve their profitability, but in a lot of position to turn on and off. Why would people who pre-ordered miners and waited months ever stop mining considering how quickly their device's lifetime coin generation will drop off as other's asics are delivered though? You can't ignore how massively oversold every company's pre-orders are. People with pre-ordered asics are much more likely to mine 24/7 without turning off no matter the price (they have to count on the price going back up to ever hope to have a return on investment now). We're not going to see a drop in difficulty until mining is totally dominated by the most efficient possible units. My point for this post was simply that miners want to make a profit, and they will have to wait to sell until a certain price point to get one. Buying a miner is basically betting on a minimum price of Bitcoin in the future (since difficulty is all but guaranteed to go up for months due to the stack of pre-orders). If this decreases supply it's possible it will be balanced out by the increased supply from faster block confirmation though.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2013, 05:11:08 AM |
|
In my other thread someone suggested that since mining is a way many people first get into Bitcoin that an increase in the price of mining each coin could increase Bitcoin market price. I think its definitely a plausible theory. The steady flow of new investors who would normally buy equipment are instead going to buy Bitcoins directly as long as the market price is less than the price per Bitcoin from hardware manufacturers. The real question is whether most miners can afford to hold out for these high break even prices or whether they will sell early (or potentially hold for a long time to double their profits?). The price per BTC represents what price point a miner would have to sell at to make his money back (no profit).
So how much does it cost to mine one Bitcoin? Assuming a 76% increase in difficulty per month, .15$/kWh, and a 2% pool fee:
(I used mining.thegenesisblock.com for these estimates)
BFL Single SC
Power cost over 7 months: ~$280 Initial cost = ~$1300 Total cost over 7 months: ~$1580
For people who get their Single SC 60 GH/s tomorrow: They will mine about 18 coins over the next 7 months. Cost: ~$88 per Bitcoin. For people who get their Single SC 60 GH/s at the beginning of October: They will mine about 10 Bitcoins over the next 7 months. Cost: ~$158 per Bitcoin. For people who get their Single SC at the beginning of November: 5.6 Bitcoins mined over 7 months. Cost: ~$282 per Bitcoin.
Bitfury - 400GH/s
Power cost over 7 months = ~$280 Initial cost = ~$8000 Total cost over 7 months = ~$8,280
Delivery day October 1: 66 Bitcoins mined over next 7 months. Cost: ~$126 per Bitcoin. Delivery day Nov 1 = 37.5 coins generated over 7 months. Cost: ~$220 per Bitcoin
KnCminer - 400GH/s Jupiter
Power cost over 7 months = ~$490 Initial cost = ~$7000 Total cost over 7 months = ~$7490
Delivery October 1: 66 coins mined over 7 months. Cost: ~$114 per Bitcoin Delivery Nov 1: 37.5 coins mined over 7 months. Cost: ~$200 per Bitcoin
HashFast - BabyJet
Power cost over 7 months: ~$259 Initial cost = ~$5600 Total cost over 7 months = ~$5859
Delivery Nov 1: 37.5 coins mined. Cost: ~$156.24 per Bitcoin Delivery Dec 1: 21.3 coins mined over 7 months. Cost: ~$275 per Bitcoin
Wow, that's a great calculation, and the price range is also reasonable! This basically laid out a price range for the next 6 months I suppose that miners are enthusiast bitcoin investors, and mining usually is the lowest cost to acquire coins. When mining electricity cost become higher than coin price, miners will shut down rigs and buy coins directly, thus re-balance the price-difficulty But it is difficult to evaluate the effect of the equipment cost. Higher equipment cost will only stop new investors from investing in mining rigs, but existing miners who have already invested in mining rigs might never get their investment back if the mining income quickly dropped close to electricity cost. I don't know what kind of choice they have, but very likely they won't sell their coins (If they do, their loss will be realized and never have a chance of ROI). And they might bring the rigs offline, which will create a huge offline hash power reserve, is this good or bad for the network???
|
|
|
|
spiral_mind (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2013, 06:34:36 AM |
|
they might bring the rigs offline, which will create a huge offline hash power reserve, is this good or bad for the network???
This likely won't happen until it costs about as much in electricity to mine a Bitcoin as it does to buy it. As long as mining appears to be cheaper people will pump money into mining equipment (and to pre-orderers not using a calculator with high enough difficulty adjustments it will look very profitable until most people's asics get delivered). When mining finally reaches the point of being barely profitable at all even with already owned asics it'll be a very good thing for the network. We'll have reached a point that will make it extremely expensive to 51% the Bitcoin network so it will be much more secure.
|
|
|
|
gmaxwell
Staff
Legendary
Offline
Activity: 4242
Merit: 8654
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2013, 07:29:03 AM |
|
Why would people who pre-ordered miners and waited months ever stop mining considering how quickly their device's lifetime coin generation will drop off as other's asics are delivered though?
Why would they run them when they cost more in power to run than they yield in coin? No matter how many months you waited for it, it's not going to be super fun to run when every moment you have it running costs you money. Might was well turn it off and buy some coins for less. Mining is only a small part of the market supply, we've _never_ seen evidence of price following difficulty, only the opposite. As time goes on less and less of coin supply is being freshly mined— things may turn out differently than they were in the past, but I wouldn't expect it.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2013, 11:48:04 AM |
|
On further thoughts, I think this huge offline hash power reserve will work the same as GPUs last year: All the GPUs were available for mining coins but many of them were not used. At the end of 2011, electricity cost was higher than coins mined, that pushed difficulty down for some time but exchange rate never went too low There is another question of the energy consumption for bitcoin: If bitcoin's total market value were really going to reach trillions of dollars, it means the electricity consumed will also be worth trillions of dollars. If you compare with today's fiat money system, they only spend billions of dollars to maintain the infrastructure The heat generated will cause global warming World's energy consumption: 180PWH for the year, 493TWH consumed for each day, and each day there are 3600 coins, 137GWH/coin, 137,000,000KWH/coin, at a price of $0.1/KWH, using 1/10 of world's power to mine coins, $1.3 million/coin is the cap ![Tongue](https://bitcointalk.org/Smileys/default/tongue.gif)
|
|
|
|
Rival
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 12:32:13 AM |
|
LOL! Al Gore will lead the charge against bitcoins because it contributes to global warming. I can see the headlines already.
|
|
|
|
prophetx
Legendary
Offline
Activity: 1666
Merit: 1010
he who has the gold makes the rules
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 12:51:13 AM |
|
LOL! Al Gore will lead the charge against bitcoins because it contributes to global warming. I can see the headlines already.
umm Al Gore invented the Bitcoin
|
|
|
|
xxjs
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 01:10:24 AM |
|
It will cost, but only for the remaining coins.
|
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 01:20:30 AM Last edit: September 01, 2013, 03:02:00 AM by DeathAndTaxes |
|
There is another question of the energy consumption for bitcoin: If bitcoin's total market value were really going to reach trillions of dollars, it means the electricity consumed will also be worth trillions of dollars.
Not exactly correct. The total value of money supply is irrelevant. What matters is annual miner rewards. The money supply is 11.5M BTC worth ~ $1.65B USD; however miners collect only ~1.31M BTC worth ~$150M USD per year in mining rewards. Annual electrical consumption isn't going to be more than that. Would you spend $2 in electricity to mine $1 in BTC even if you had free hardware?
|
|
|
|
xxjs
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 01:41:34 AM |
|
And by the way, the correlation between mining cost and price is that price comes first, out of the minds of the market actors. The cost of the coins will match that price.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 06:55:52 PM |
|
There is another question of the energy consumption for bitcoin: If bitcoin's total market value were really going to reach trillions of dollars, it means the electricity consumed will also be worth trillions of dollars.
Not exactly correct. The total value of money supply is irrelevant. What matters is annual miner rewards. The money supply is 11.5M BTC worth ~ $1.65B USD; however miners collect only ~1.31M BTC worth ~$150M USD per year in mining rewards. Annual electrical consumption isn't going to be more than that. Would you spend $2 in electricity to mine $1 in BTC even if you had free hardware? Thanks for pointing out this! It is much greener than I actually thought. So the actual figure could be much less once we pass certain stage, I think the reward in transaction fee might rise to 5-10 coins in 10 years and that will decide the power consumption
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 06:59:19 PM |
|
And by the way, the correlation between mining cost and price is that price comes first, out of the minds of the market actors. The cost of the coins will match that price.
Price come first because the market always looking forward 6 months, the price for bitcoin is always a projection of the mining cost per coin 6 months later, by smart investors
|
|
|
|
xxjs
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 07:19:05 PM |
|
And by the way, the correlation between mining cost and price is that price comes first, out of the minds of the market actors. The cost of the coins will match that price.
Price come first because the market always looking forward 6 months, the price for bitcoin is always a projection of the mining cost per coin 6 months later, by smart investors Let me try again. The value of a bitcoin originates in the minds of all actors, including buyers, sellers, sellers of goods, hoarders, asic manufactures, bitcoin pole dancers..., taking into account everything, including the present and the future. If some actors's values cross each other, there will be a trade and hence a current price. The cost of mining is derived from that price. Your theory is the labour value theory, capital being condensed labour. It has been refuted for centuries.
|
|
|
|
odolvlobo
Legendary
Offline
Activity: 4424
Merit: 3324
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2013, 09:25:55 PM Last edit: September 01, 2013, 09:42:19 PM by odolvlobo |
|
In my other thread someone suggested that since mining is a way many people first get into Bitcoin that an increase in the price of mining each coin could increase Bitcoin market price. I think its definitely a plausible theory. The steady flow of new investors who would normally buy equipment are instead going to buy Bitcoins directly as long as the market price is less than the price per Bitcoin from hardware manufacturers.
There are two problems with this idea. - First, the demand for bitcoins is the same whether the person chooses to buy them or mine them.
- Second, the cost of mining will never exceed the value of the mined bitcoins (at least in a rational world). This discontinuity breaks any dependency of price on cost of production.
The idea that the cost of mining determines the value of the bitcoins is essentially the Labor Theory of Value. This theory doesn't apply to bitcoins because the supply of new bitcoins is unaffected by the cost of mining. For example, the relative values of apples and oranges depend on the cost to grow them because if increasing costs cut into profits of one, then farmers will grow the other instead. The varying supplies create an equilibrium around the cost of production. However, it doesn't matter how much or how little it costs to mine bitcoins. The supply of new bitcoins is predetermined and unaffected by production costs. Finally, you can't overlook the fact that Bitcoins aren't consumed. The supply of bitcoins at this point is overwhelmingly determined by the bitcoins that are already in circulation.
|
Join an anti-signature campaign: Click ignore on the members of signature campaigns. PGP Fingerprint: 6B6BC26599EC24EF7E29A405EAF050539D0B2925 Signing address: 13GAVJo8YaAuenj6keiEykwxWUZ7jMoSLt
|
|
|
marcus_of_augustus
Legendary
Offline
Activity: 3920
Merit: 2349
Eadem mutata resurgo
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2013, 01:29:52 AM |
|
The cost of mining does not determine the price of the coins at any time, they are probably not correlated at all ... but it can act as a floor (elastic lower bound) to the price during down swings. This is an important dynamic to keep in mind, so doing the kind of calcs as in the OP are useful (and worthwhile), although doing them properly is becoming ever more complex with hardware evolution (ASICs), decreasing supply/total btc ratio amongst other factors.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2013, 03:28:38 AM |
|
And by the way, the correlation between mining cost and price is that price comes first, out of the minds of the market actors. The cost of the coins will match that price.
Price come first because the market always looking forward 6 months, the price for bitcoin is always a projection of the mining cost per coin 6 months later, by smart investors Let me try again. The value of a bitcoin originates in the minds of all actors, including buyers, sellers, sellers of goods, hoarders, asic manufactures, bitcoin pole dancers..., taking into account everything, including the present and the future. If some actors's values cross each other, there will be a trade and hence a current price. The cost of mining is derived from that price. Your theory is the labour value theory, capital being condensed labour. It has been refuted for centuries. There are smart investors who know what is really happening behind the scene and they could project the future price with high precision, they buy coin before it is too late. Their purchase will push up the price inevitably, so when you see a steady rise in exchange rate, you know that difficulty will rise soon It is easy to project the sharp rise in difficulty if you know BFL is going to put hundreds of thousands of ASIC devices into market in the second half of the year. But this only becomes clear when Avalon first delivered their ASIC at the end of January. Following Avalon batch 1 delivery in February, all the uncertainty about ASIC devices were gone, exchange rate started to explode And just like marcus_of_augustus said, the cost of mining act as a reference of the coins exchange rate (if the cost is higher than price of the coin, miners will shutdown rigs and buy coin directly, their buy will raise the price) We suppose that people always want to get coins, and they can get coin through either mining or buying, mining decided the lowest possible cost for a coin
|
|
|
|
|