total bitcoin energy consumption

[mihai]

Hi there!

I've been lurking quite some time, finally decided to register because I wanted to answer in the thread about energy consumption (http://forum.bitcoin.org/index.php?topic=32755.0). Unfortunately as a newbie I can't post there, so maybe someone could just copy the following there? Thanks a lot!


Estimation using current difficulty (d = 1690896) and a crude guess of the current mean hardware efficiency (m = 1.5 MHash/J):

Hashrate: h = d * 2^32 / 600 = 12.1 THash / s
total power = h / m = 8.1 MJ/s = 8.1 MW.

Comparing that to a typical (german) nuclear power plant of say 1300 MW, that would make about 0.6 percent. Or, 80 medium-sized wind turbines (1 MW each). So it's not really neglible, but also not a big deal globally. But who knows how difficulty rises in the future.

Regarding the "free energy" guy: yeah, dream on...

[1714601064]

1714601064

[1714601064]

1714601064

[1714601064]

1714601064

[1714601064]

1714601064

[gellimac]

Hi there!

I've been lurking quite some time, finally decided to register because I wanted to answer in the thread about energy consumption (http://forum.bitcoin.org/index.php?topic=32755.0). Unfortunately as a newbie I can't post there, so maybe someone could just copy the following there? Thanks a lot!


Estimation using current difficulty (d = 1690896) and a crude guess of the current mean hardware efficiency (m = 1.5 MHash/J):

Hashrate: h = d * 2^32 / 600 = 12.1 THash / s
total power = h / m = 8.1 MJ/s = 8.1 MW.

Comparing that to a typical (german) nuclear power plant of say 1300 MW, that would make about 0.6 percent. Or, 80 medium-sized wind turbines (1 MW each). So it's not really neglible, but also not a big deal globally. But who knows how difficulty rises in the future.

Regarding the "free energy" guy: yeah, dream on...

false (I think with the calcul I made in the Bitcoin discussion topic I found 1/3 or a nuclear plant...

[mihai]

I'm afraid you will have to prove that by showing your "calculus" and proving the above wrong (good luck with that). It's true, though, that the estimated hashrate is too low (12.1 instead of ~15 THash/s as can be found in the other thread), because that was true for the last 2016 blocks.

Also, you're confusing physical units in your post, because you don't refer to a time period (like x KWh for a year). Power is energy per time units.

[gellimac]

Yeah sorry I tank wrong data for wikipedia I took kWh instead of kW, I found 1% with the good date 

[mihai]

Oh I see. Never underestimate the importance of units

[deepDown]

I find that comparing it to cars gives much nicer perspective than the comparison with nuclear power plants.

Lets round it up to 10MW.
In horse power (hp) it is:

10MW = 13 410hp

now let say an average US car has ~150hp

So the total power of the network is under
13400 / 150 ~= 90 average US cars

or less than 35 MACK trucks, assuming avg power of these is 400hp (http://www.macktrucks.com/default.aspx?pageid=1359)

Now it seems much less ecologically terrifying, no?

[alemaaltevinden]

Hey,

Free energy may be a little bit too (though in social country's like Belgium, The Netherlands, Most other European country's it may be possible (not the USA  ) but I expect energy to become very cheap. And maybe if they find a way to get a lot of energy that needs to be consumed asap there maybe in some moments of the day (think at night when there isn't much energyusage) free electricity. Though you would still pay for the cable and probably some other shit they gonna ask you money for, but it may be like internet. You pay 50 and you get unlimited access.

Regards,
alemaaltevinden

[kommykiller]

What would be the turning point of profitbale in MHash/s vs power input?

[Ringmasta]

Hey,

Free energy may be a little bit too (though in social country's like Belgium, The Netherlands, Most other European country's it may be possible (not the USA  ) but I expect energy to become very cheap. And maybe if they find a way to get a lot of energy that needs to be consumed asap there maybe in some moments of the day (think at night when there isn't much energyusage) free electricity. Though you would still pay for the cable and probably some other shit they gonna ask you money for, but it may be like internet. You pay 50 and you get unlimited access.

Regards,
alemaaltevinden

A lot of places in the US are starting to vary energy cost by time of day to account for demand, so it may be possible to take advantage of that to do off-peak mining.  However, it's still pretty tough to get back your GPU investment on a dedicated rig if you can only mine for ~8 hours/day or so.

[zveda2000]

I have a possibly silly question:

So the total computing power must remain high at all times in the system in order to protect the block chain from attack, right? Could this trigger an energy / computing power arms race in the future, where a very large amount of energy is needed to be applied constantly to maintain the high computing power to protect the block chain?

Or is this amount of computing power always going to be limited by the amount of hardware etc., rather than the energy required?

[Cellarius]

I am surprised that I see so little discussion on energy consumption.  I love the concept of bitcoin but hate to see so much energy spent to maintain it.  I don't think bitcoin can survive unless a way is found to consume less.

[notlist3d]

I am surprised that I see so little discussion on energy consumption.  I love the concept of bitcoin but hate to see so much energy spent to maintain it.  I don't think bitcoin can survive unless a way is found to consume less.

It is not energy consumption that makes it hard to mine in most cases, but energy prices.  I think you will find the big mega mines have very low cost on power, also they are mostly privately owned and you just will not find the numbers OP is looking for.   With being privately owned most don't open up their books and say how much electricity they use as they don't have to.

One thing to is some places produce more electricity then they have use for.  One of the few big one's that opened up is over here - https://bitcointalk.org/index.php?topic=1072474.0 . eric@haobtc did a amazing job sharing living at a mine.  But in their case they teamed up with a hyrdo dam, so they were using electricity that would have wasted if they were not there.  So not all electricity is from "bad" sources.

[helloeverybody]

I find that comparing it to cars gives much nicer perspective than the comparison with nuclear power plants.

Lets round it up to 10MW.
In horse power (hp) it is:

10MW = 13 410hp

now let say an average US car has ~150hp

So the total power of the network is under
13400 / 150 ~= 90 average US cars

or less than 35 MACK trucks, assuming avg power of these is 400hp (http://www.macktrucks.com/default.aspx?pageid=1359)

Now it seems much less ecologically terrifying, no?

This seems to be a better way to look at it. One of my concerns for the growth of bitcoin was the ever growing consumption of energy making bitcoin one of the lest eco friendly things out their. In a world where everyone seems to be trying to reduce ttheir energy consumption some people would look at bitcoin as a mammoth waste of energy. But using the current system there isnt really an easy way to reduce this and over the yeras its just going to get worse as difficulty increases, unless the miner makers can create more economical rigs.

[CoinBreader]

I am surprised that I see so little discussion on energy consumption.  I love the concept of bitcoin but hate to see so much energy spent to maintain it.  I don't think bitcoin can survive unless a way is found to consume less.

Well if that happent anytime soon, ill go again all into mining industry! but i dont see it coming any time soon.. and i dont think in the near future btc mining will become eco friendly ...