Correlation between mining costs and Bitcoin value and ecological nightmare

[flug]

This must have been covered before, but I cant find where.

Even tho there isn't a direct causal relationship between mining costs and bitcoin price, there is a market relationship which ensures that mining costs do tend to correlate with bitcoin prices. Correct?

If so, does that mean that when Bitcoin takes over the world and bitcoin prices increase a million fold, mining power will increase from the current 5MW to 5TW?

[Synaesthesia]

This must have been covered before, but I cant find where.

Even tho there isn't a direct causal relationship between mining costs and bitcoin price, there is a market relationship which ensures that mining costs do tend to correlate with bitcoin prices. Correct?

Well there may be a trend like that - but bitcoins are worth a lot more than it costs to mine, in electricity currrently. So no.

If so, does that mean that when Bitcoin takes over the world and bitcoin prices increase a million fold, mining power will increase from the current 5MW to 5TW?

That's a big "if". And the power usage of all PC's around the world is a mere fraction of total energy consumption.

[wareen]

You also have to consider that the block reward is decreasing - today you'll get 50 BTC for every mined block which halves roughly every 4 years. So the price of one BTC could theoretically double every 4 years without changing mining profitability.

In the end it will be only the transaction fees covering the miners cost but if we assume, say 1 BTC total transaction fees per block, you could still have a Bitcoin worth 1000 USD with current mining profitability levels (ie. roughly the same power usage).

[Gabi]

Well, a lot of dollars are printed each years, same for gold and what else

[BombaUcigasa]

mining power will increase from the current 5MW to 5TW?

DID YOU READ THE SATOSHI PAPER?!!?!?

It clearly states that by his estimations he had to size the total volume of bitcoin and release of new coins in such a way that mining operations had an impact on the world by increasing electricity costs noticeably, but not in a disruptive manner.

[cloud9]

Energy costs in producing all the world's secure bitcoins should be lower than energy costs mining the same value of gold.  Energy cost is not the only production cost in manufacturing bitcoin - capital cost and human effort should not be neglected in the calculations.

At some stage when the bitcoin network reach more than half the total processing hashing power in the world - maybe the threat of a hostile 50%+ attack will be less and this fact will become a new influence in the network growth as well?

[flug]

Quick calculation:

In a few years, assume BTC price rises by a factor of 1000 and total bitcoins = 10M

BTC Market Cap = 10M BTC @ $20 x 1000 = $200B (not a huge %age of the world's current currency)
Power = 5MW x 1000 = 5GW

So, market cap of $200B = 5 Nuclear power stations (@ 1GW each)

I might be out by factors of 2, but not factors of 10.

Where is this calculation wrong?

[JoelKatz]

Even tho there isn't a direct causal relationship between mining costs and bitcoin price, there is a market relationship which ensures that mining costs do tend to correlate with bitcoin prices. Correct?

Nope. Mining costs don't affect the supply and they don't affect the demand.

[flug]

Even tho there isn't a direct causal relationship between mining costs and bitcoin price, there is a market relationship which ensures that mining costs do tend to correlate with bitcoin prices. Correct?

Nope. Mining costs don't affect the supply and they don't affect the demand.

Mining costs don't affect supply/demand, but the bitcoin price affects how profitable mining is.

Price doubles -> it's worth spending twice as much to mine the same amount of BTC

[NO_SLAVE]

look in the mining thread, there was several discussions there.

[wareen]

So, market cap of $200B = 5 Nuclear power stations (@ 1GW each)

What does market capitalization have to do with the electricity used for mining?
Seriously, did you even read the posts in this thread?

The incentive for miners to mine does not depend on market capitalization. Difficulty (~hashrate, hence energy usage) loosely follows price, but difficulty will roughly be halved about every 4 years due to decreased block reward. Therefore if in 20 years we have 1 BTC = 1000 USD we would have about the same energy usage as today (if energy prices stay the same).

[bitcoinBull]

This must have been covered before, but I cant find where.

Even tho there isn't a direct causal relationship between mining costs and bitcoin price, there is a market relationship which ensures that mining costs do tend to correlate with bitcoin prices. Correct?

I, for one, agree here.

If so, does that mean that when Bitcoin takes over the world and bitcoin prices increase a million fold, mining power will increase from the current 5MW to 5TW?

The hash rate will increase, but total power consumption may not.  Chips will get more power efficient, in an exponential manner of Moore's Law.  So no, mining power consumption could stay roughly the same.  But the money spent on chips, and the hash rate, will increase dramatically.

[flug]

The hash rate will increase, but total power consumption may not.  Chips will get more power efficient, in an exponential manner of Moore's Law.  So no, mining power consumption could stay roughly the same.  But the money spent on chips, and the hash rate, will increase dramatically.

Not sure about this. I think it steady states when the electricity-cost/hour equals bitcoin-value-mined/hour. Increase in efficiency just produces increased hash rate as a side effect, which is good for the integrity of the network, but won't save power.

[flug]

So, market cap of $200B = 5 Nuclear power stations (@ 1GW each)

What does market capitalization have to do with the electricity used for mining?
Seriously, did you even read the posts in this thread?

The incentive for miners to mine does not depend on market capitalization. Difficulty (~hashrate, hence energy usage) loosely follows price, but difficulty will roughly be halved about every 4 years due to decreased block reward. Therefore if in 20 years we have 1 BTC = 1000 USD we would have about the same energy usage as today (if energy prices stay the same).

OK, I'll simplify things somewhat.

If in 18 months time, when the reward is still 50BTC, the price has gone to $2000 ($20x100), the power used in hashing will be 500MW (5MWx100), which is about half a nuclear reactor. Yes?

[PatrickHarnett]

The power might be 50MW or 500MW (doesn't really matter how much) because that determines the cost of production, but not the price someone is prepared to sell at.  Normally people like to make a profit, but that's not always the case, especially when there is a fixed/variable relationship.  So for the short-run cost (SRMC), if you can sell with a margin, you should, if not, don't.  You might choose to stock-pile or cease production.  At some point the transaction fee per block is supposed to become material when the coin reward is not.

Those that can produce cheaply will, those that cannot won't.

[wareen]

If in 18 months time, when the reward is still 50BTC, the price has gone to $2000 ($20x100), the power used in hashing will be 500MW (5MWx100), which is about half a nuclear reactor. Yes?

Ok sorry, I misinterpreted your previous timeframe ("in a few years") to mean something longer than 18 months.
If we really would see a 100 fold price increase in 18 months then energy usage will probably be very high as well but there are a few factors which you might not have considered:

1. If difficulty increases steadily at the required ~29% per month then we will reach block 210.000 already in about 13 months.
2. New mining hardware will not be bought anymore some time before block 210.000 because there is not enough time left to recoup the costs.
3. Peak energy usage will probably only be held for a relatively short period of time.
4. Increasing energy prices (which are probably to be expected in the next years) will further cut down on energy usage.
5. Intermediate price decreases or even crashes due to speculation might scare off some potential big mining investors.

There could also be other factors which would limit the power requirements if it really gets out of hand - for example, miners could agree that new blocks are only accepted within a 5-minute time window (granted - this is highly speculative and such changes would have to be very carefully designed to not change the basic dynamics of the system).

Having said that, I very much doubt that we would see a 100 fold increase in price within the next year, but if it really happens, then it is obvious that society deems Bitcoin as a very valuable service and therefore it is probably justified to use up some energy (which is hopefully saved in part with other financial services becoming less important then).

Last, I really hope that the percentage of energy coming from renewable sources increases in the next years, which would also mitigate some of the ecological impact.

[flug]

If in 18 months time, when the reward is still 50BTC, the price has gone to $2000 ($20x100), the power used in hashing will be 500MW (5MWx100), which is about half a nuclear reactor. Yes?

Ok sorry, I misinterpreted your previous timeframe ("in a few years") to mean something longer than 18 months.
If we really would see a 100 fold price increase in 18 months then energy usage will probably be very high as well but there are a few factors which you might not have considered:

1. If difficulty increases steadily at the required ~29% per month then we will reach block 210.000 already in about 13 months.
2. New mining hardware will not be bought anymore some time before block 210.000 because there is not enough time left to recoup the costs.
3. Peak energy usage will probably only be held for a relatively short period of time.
4. Increasing energy prices (which are probably to be expected in the next years) will further cut down on energy usage.
5. Intermediate price decreases or even crashes due to speculation might scare off some potential big mining investors.

There could also be other factors which would limit the power requirements if it really gets out of hand - for example, miners could agree that new blocks are only accepted within a 5-minute time window (granted - this is highly speculative and such changes would have to be very carefully designed to not change the basic dynamics of the system).

Having said that, I very much doubt that we would see a 100 fold increase in price within the next year, but if it really happens, then it is obvious that society deems Bitcoin as a very valuable service and therefore it is probably justified to use up some energy (which is hopefully saved in part with other financial services becoming less important then).

Last, I really hope that the percentage of energy coming from renewable sources increases in the next years, which would also mitigate some of the ecological impact.

No worries, I only had the question half formulated when I first typed it in. It'd be interesting to factor in those points above, and graph out the corresponding max value of BTC over time assuming a max of 1GW mining power.

The main point I'm trying to make is that with the current setup, bitcoins going to $10,000 or taking any significant percentage of world GDP or the world black market over the next few years isn't going to happen. It's too severely limited by the power available in the world.

[benjamindees]

The main point I'm trying to make is that with the current setup, bitcoins going to $10,000 or taking any significant percentage of world GDP or the world black market over the next few years isn't going to happen. It's too severely limited by the power available in the world.

No no this is all wrong, ugh.  If Bitcoin demand rises then Bitcoin prices rise then mining demand rises then power demand rises and power prices rise and the price of
Bitcoins in dollars has nothing to do with the amount of power available anyways.

[Timo Y]

If Bitcoin ever becomes that big, mining will be so competitive that it will use electricity that can can be used for little else but mining.

[iya]

If Bitcoin ever becomes that big, mining will be so competitive that it will use electricity that can can be used for little else but mining.

A new kind of electricity?