Debunking the waste of energy argument against Bitcoin

[Bitcotalk]

You make great points, and I completely agree, but the true problem lies within the general public who will consume the news piece -- they're not going to care.

This is yet another issue that could be sensationalized, and those aren't good for Bitcoin as a whole. Green groups are going to fixate on the cons, and with enough traction, could cause government regulators to step in. It's not a problem now, but it could be in the future.

Yeah, really great and valid ones. Isn’t the media's main target the public viewers and readers who really do not have any knowledge and are as ignorant as them? It is just deliberate and even if it is not, it is just a way of showing how dumb they are and will always be (referring to the media).

Gone are those days when you can ever trust anything that comes out of the media as these days, even the media is corrupt. At the end though, those who are smart will always not be deceived.

[1715673598]

1715673598

[1715673598]

1715673598

[1715673598]

1715673598

[1715673598]

1715673598

[stompix]

...
Second.
As long as somebody in this world consumes coal, you burning free green energy doesn't make things right.
Because if you would stop, then the energy consumed by you would be redirected to the one burning coal energy, and the coal power plant will be shut down....

I concede that my comparison of the energy consumption of the US military to the
energy consumption of Bitcoin mining is somewhat flawed, because they are not directly comparable. However,
I simply wanted to show that there are still many things that are destructive instead of constructive that consume
more energy than Bitcoin mining.

I strongly disagree with the statement in the quote. E.g. if a BTC mining company uses renewable energy
from hydro power in Iceland there is no way that the energy would be redirected to a country that is burning
coal energy if the mining company would stop mining. This is simply completely uneconomical due to the location of Iceland.
On the other hand Bitcoin mining is possible from any place on Earth with an internet connection and therefore it
is a great use of renewable energy that is abundant in countries like Iceland or Canada (countries like Norway would
be great, too, but energy prices are too high there).

You see, you're again missing the point, and you treat bitcoin related activities in a special way.

Miners can move from China to Iceland to mine with green energy but other industries are not allowed!!!!
There are 4 aluminium smelters in Iceland (big ones), and 20x more in China burning coal.
Why not move those there?

You just need an internet connection?
Oh, let's have datacenters there.

And you gave one example, probably the only in a world where we have an island that is too far away to connect via a cable..........for now
https://www.landsvirkjun.com/researchdevelopment/research/submarinecabletoeurope

Bitcoin is not perfect, nor is the PoW.

[pawel7777]

I´m tired of hearing the argument "but Bitcoin wastes more energy than
a whole country..." all over the mainstream media.

Here are a few reasons why the energy that is consumed by Bitcoin mining is not wasted:
1. The energy that is consumed by Bitcoin mining is the foundation of the security of the network
2. Proof-of-work is superior to other consensus algorithms like Proof-of-Stake and offers less attack vectors
3. Fiat currencies are supposedly backed by the power of the state... meanwhile the US military consumes more energy than Bitcoin mining per year
(this is a comparison of the military of a single country compared to a truly global currency)
4. No one is talking about the enormous costs of transporting fiat money, securing it and printing it in the first place
5. The mining of other commodities is extremely energy-intensive as well

If you take all these factors into account, Bitcoin may actually be less energy-intensive than traditional
fiat currencies. Besides, many mining companies have set up their operations in countries with an abundance
of renewable energies like Canada or Iceland. The days where Bitcoin was solely mined in rural Mongolia and China
may finally be behind us.

1. Yes. But that doesn't mean it's not a waste. You could change the foundation to something non-wasteful. PoW is quite literally a waste of energy by design. This energy is not going towards producing anything. You waste more energy - you earn more rewards.

2. Is it? Have you considered centralisation as a safety factor (be it mining pool or geographical centralisation)? Are you not aware that there was a time when ghash.io pool briefly exceeded 50% of total hashrate, making mining centralised for a moment. In free market conditions PoW is inherently trending towards centralisation of mining and/or manufacturing mining hardware. No electricity cost will be the same in every country, one hardware will be more efficient than others (see bitmain), one mining farm will be more efficient than competition etc.

3-5. Not an arguments. What you're saying is "X is not a waste of energy because Y is using a lot of energy". I'm not fat because some other guy is also fat/fatter. You get the gist.

You mentioned Iceland's renewable energy. Any thoughts on Iceland's proposal to tax bitcoin mining, because they're concern it can drain too much electricity? https://www.coindesk.com/icelandic-lawmaker-floats-bitcoin-mining-tax/

This thread is also worth reading, it explains how rising Bitcoin price could fuck up energy supply on a global level: https://bitcointalk.org/index.php?topic=2847680.0

[stompix]

This thread is also worth reading, it explains how rising Bitcoin price could fuck up energy supply on a global level: https://bitcointalk.org/index.php?topic=2847680.0

There was also an older thread, back in 2014 ringing a few alarms.

https://bitcointalk.org/index.php?topic=694401.msg7841368#msg7841368

At the current price ($600 per BTC) it means that the network consumes something like 600MW. That is one third of the electric capacity of the plant of Fessenheim that made the news today (in France).

We're 10 times above that (price is almost x20) but also the rewards halved.

The question is, what if BTC goes to 100 000$?
The entire untapped energy in Iceland won't be enough.
With that much money , I hate to say it but "printed" each day, we will have millions of miners added to the network.

Should we do the numbers for 1 million /BTC?
At 1 million, and even at a 10c/kwh price we are going to see 300 GW , as it's still profitable to mine and you will still get a ROI.
And even Itapu is a joke  to that number.

I really think that people dismiss to easy the impact mining has, and it's not just bitcoin.
Tried to buy a GPU lately ?

[drays]

So they say - "Bitcoin wastes more energy than a whole country..." ??  

Well, yes. So what? I would say - what Bitcoin brings to the world is more important than most countries will ever bring. As to me, Bitcoin is more important to humanity than most of the countries are. So its ok for it to spend (waste is a wrong term here) so much electricity.

Except that, mining-produced heat won't be wasted left unused for long. I am pretty sure in the very near future it will be used to heat up houses or whole buildings/suburbs, as well as for other, not so obvious needs.

Even now, with proper watercooling, you can use the mining heat to keep your house warm at winter. Just needs a bit of engineering effort. As soon as someone produces a standard commercial solution for that (an actual product kit with optional installation service), it will be adopted by many.

Imagine the world where big buildings/cities are using mining farms at boiling houses instead of regular "dumb" boilers

[mriansa]

yes you are right miners bitcoin now is too over they use energy that is very much and they should make environmentally friendly energy such as solar panel energy for mining it is very good or water power and wind power is pretty good because not too dredge oil that exist on this earth for the energy they use.

[severaldetails]

To me bitcoin is an economic structure where value is created. And nearly every economic structure that creats value needs energy. I do not see any difference to the automobile industry. They create cars, we create bitcoins. And I see even the tendency that we handle energy responsible. Miners use the newest energy saving hardwar to create bitcoins.

[d5000]

I was curious about what could be the final consumption of Bitcoin mining** if it was a major world currency. It is a rough calculation, we would need much more details, but I think the approximate order of magnitude could be correct.

For this calculation, I took two sources:
1) Electricity consumption of Bitcoin: a market-based and technical analysis (March 2017; his estimation is between 300-800 MW)
2) What is the fair value of a cryptocurrency? - pablomp's theory about a dependency of a "fair" price of a cryptocurrency on the transaction volume, based on the Quantity Theory of Money.

Let's base the calculation on the following assumptions:

1) What transaction rate would a "World Currency Bitcoin" have?
- World population: 8 billion (8000000000)
- Family/household size: 4 persons
- Every household should be able to do one Bitcoin transaction per day (It is NOT important if it's on-chain or off-chain, it can be using LN, sidechains, pegged smartcoins etc. - but the value must be denominated in Bitcoin).
- Necessary transaction rate: ~25000 tx/sec (2,16 billion/day)

2) What "steady" price would such a "world currency" have?
- pablomp's theory roughly assumes that with the 3,5 tx/s the Bitcoin network reached in mid-2017, we had approximately a "fair price" of $250 (based on QTM). At the moment of his study the price was about 10 times higher, but that is not important as most of today's price is based on speculation on future usage.
- That means that 1 tx/sec would mean a price of $71,4.
- With 25000 tx/s the price of 1 BTC becomes 1785000 USD (Moon! ).

3) How much would electricity would a 1.8 million BTC consume?
This is the hardest part, because we could say that "it depends" on various factors, like the mining reward, transaction fees, efficiency of mining hardware, and so on. Let's do it the simple & lazy way and say that miners' income in BTC and miner hardware efficiency would be similar to 2017 (this is a forum post, not an academic study ).*

OK, let's go:
- In March 2017, according to Marc Bevand, the electricity consumption was between 325 and 774 MW. The price of 1 BTC, in this period, was about $1000 (it was the era of the "ETF rally and crash").
- So we can calculate the approximate consumption of a 25K tx/s and 1.785 million BTC and get the following results:

Lower bound: 325 * 1785 =580125 MW (~580 GW) - about 500 nuclear plants of 1 GW
Upper bound:  774 * 1785 = 1381590 MW (~1,38 TW) - about 1300 nuclear plants of 1 GW


*There is actually a better approach: calculate the consumption based on the past electricity cost instead of the past electricity consumption because miner hardware efficiency then would not be relevant anymore. But this would be much more complicated.
** Clarified that it's only taking into account mining, not other power consumptions (like users' devices).

[sjefdeklerk]

- Necessary transaction rate: ~25000 tx/sec (2,16 billion/day)

You do realize bitcoin has a maximum of 7 tx/sec ? Well ok in all fairness, 14 tx/sec after segwit.

[d5000]

- Necessary transaction rate: ~25000 tx/sec (2,16 billion/day)

You do realize bitcoin has a maximum of 7 tx/sec ? Well 14 tx/sec after segwit.

Yes, but the 3,5 tx/s (depending on the day, between 3 and 4 tx/s, actually) was the real transaction rate at the time of Marc Bevand's study in March 2017 (see Blockchain.info stats). That is the relevant number for this kind of calculation, not the maximum transaction rate in tx/s.

We can discuss if the transactions rate is the correct measure, maybe it's also the "number of new UTXOs" (because one Bitcoin TX can in theory generate hundreds of UTXOs), but I don't know stats about that.

PS: Maybe you meant "Bitcoin could never reach 25K  tx/s" but read the phrase before this assumption - these 25 tx/s can be carried out on-chain, off-chain, on sidechains and even on pegged coins that are not sidechains (like the BitBTC on the Bitshares platform).

[sjefdeklerk]

PS: Maybe you meant "Bitcoin could never reach 25K  tx/s"

Exactly. BTC can, since segwit, do a maximum of 14 transactions per sec, nowhere near 25.000.

but read the phrase before this assumption - these 25 tx/s can be carried out on-chain, off-chain, on sidechains and even on pegged coins that are not sidechains (like the BitBTC on the Bitshares platform).

Ok but off-chain is a whole different beast. You're interpolating on-chain but who knows how much electricity LN transactions will costs. As a device NEEDS to be online in order to be part of a LN payment route and tons of info need to continually pass through the network (for routing purposes), this is really a whole different calculation. Where people now mostly power down their pc's at night, with LN they'll probably keep them running 24 hours/day. IF LN would ever to become a success (which I highly doubt for reasons I've laid out already: https://bitcointalk.org/index.php?topic=2792933.0)

Besides, BTC is currently doing 51 Tw/year already: https://digiconomist.net/bitcoin-energy-consumption

[stompix]

3) How much would electricity would a 1.8 million BTC consume?
This is the hardest part, because we could say that "it depends" on various factors, like the mining reward, transaction fees, efficiency of mining hardware, and so on. Let's do it the simple & lazy way and say that miners' income in BTC and miner hardware efficiency would be similar to 2017 (this is a forum post, not an academic study ).*

OK, let's go:
- In March 2017, according to Marc Bevand, the electricity consumption was between 325 and 774 MW. The price of 1 BTC, in this period, was about $1000 (it was the era of the "ETF rally and crash").
- So we can calculate the approximate consumption of a 25K tx/s and 1.785 million BTC and get the following results:

Lower bound: 325 * 1785 =580125 MW (~580 GW) - about 500 nuclear plants of 1 GWp
Upper bound:  774 * 1785 = 1381590 MW (~1,38 TW) - about 1300 nuclear plants of 1 GWp


*There is actually a better approach: calculate the consumption based on the past electricity cost instead of the past electricity consumption because miner hardware efficiency then would not be relevant anymore. But this would be much more complicated.

It's quite scary, the maximum installed capacity of all the nuclear plants (capable of producing energy, not online) is around 400GW and 750 GW when it comes to hydro (again maximum installed)

Fortunately there two things that will limit this:

a) before we reach the 1.8 million mark we are going to have at least a halving for sure, that is going to drop the energy consumption by half based on the model.

b) and one that is not that good overall
Once there is need for electric power prices go up, at some point it won't be that profitable to mine and the consumption will go down.But in this case, there is one little catch, if energy prices go up, then everybody is affected.

Probably BTC has grown a little to fast.
Or the distribution model should have been a bit shorter.
If we would have dropped already to 3 coins per block we probably wound't have had this discussion.

[drays]

Folks, all your calculations and all those fears are based on the wrong assumption the energy spent on mining will be immediately lost - evaporated into nothingness.

Yes it is that way right now, but it won't continue for long. It would be too stupid to continue losing the heat produced by miners. That heat will be reused in boillers, and the mining will become practically free to those who'll implement that. Hence - no waste anymore and no problem, mining cryptocurrency will be just a side-job of heaters and boilers!
Cities will host heating/mining farms, persons will use mining heaters... Why not?

I know I am dropping a million dollar idea here , but I am also sure I am definitely not the first one who thought of actually implementing this. With mining popularity rising in such tempo recently, I wouldn't be surprised to see such offers soon, on a regular consumer-level first.
Think of consequences of such a "merger" - they are going to be huge, for both Bitcoin and energy market.

What about that scenario? Wouldn't you buy an electric heater with Ethernet connection which also mines..?

[d5000]

Ok but off-chain is a whole different beast. You're interpolating on-chain but who knows how much electricity LN transactions will costs. As a device NEEDS to be online in order to be part of a LN payment route and tons of info need to continually pass through the network (for routing purposes), this is really a whole different calculation.

Obviously it's a rough calculation and takes into account only mining, but I don't believe the "continuously online LN devices" would make much of a difference. And no, not all users using LN have to be online all the time, only those that want to be part of the infrastructure. LN could also run on a smartphone or on another low-powered device that is always online (routers!).

But I won't get into details because it's outside of the scope of my post. To avoid misunderstandings, I'll clarify that I talk only about mining, not about other energy consumptions.

Besides, BTC is currently doing 51 Tw/year already: https://digiconomist.net/bitcoin-energy-consumption

The source I used, Marc Bevand, believes the Digiconomist calculation to be false - but even then: it uses TWh/year, not TWp. Values are not so different between both sources if you take this "little fact" into account: The 500 MWp that Bevand assumes to be the consumption of BTC mining mean it consumed 4,38 TWh/year in February/March 2017. Digiconomist calculated 9 TWh/year for this point in time - about doubling Bevand's average calculation, and only about 1,33 times above his upper bound.

[sjefdeklerk]

And no, not all users using LN have to be online all the time, only those that want to be part of the infrastructure. LN could also run on a smartphone or on another low-powered device that is always online (routers!).

But in order for the network to become a success it NEEDS users online. So it can only become a success with high energy costs. Sure, people could start selling custom LN devices but I think it's reasonable to assume a lot of people would just use their PC. Smartphone would be possible too indeed, but it would drain a lot of battery. Most phones die within 5-6 hours of activity already.

The source I used, Marc Bevand, believes the Digiconomist calculation to be false - but even then: it uses TWh/year, not TWp. Values are not so different between both sources if you take this "little fact" into account: The 500 MWp that Bevand assumes to be the consumption of BTC mining mean it consumed 4,38 TWh/year in February/March 2017. Digiconomist calculated 9 TWh/year for this point in time - about doubling Bevand's average calculation, and only about 1,33 times above his upper bound.

Obviously it's hard to really calculate it correctly. But fact of the matter is that it's a HUGE waste of energy, just because someone invented a strategy that pays more if you spend more energy. I really hope that, and assume, that the winning coin will use something like PoS instead of PoW. This is really unsustainable.

[d5000]

It's quite scary, the maximum installed capacity of all the nuclear plants (capable of producing energy, not online) is around 400GW and 750 GW when it comes to hydro (again maximum installed)

Yep. However, we can use also a more "ecologic" source for the comparison: solar energy. And then it becomes less scary.

Let's take Andasol, a solar concentration plant in Spain. It generates 150 MW or 110 GWh/year, on 56 ha. To get the 1,38 TW of the most pessimistic assumption, solar plants of the same efficiency in the same climate (and Andasol-1 is already 10 years old) would have to cover 56 * (1380000/150) = 515200 ha or 5152 square km. This is less than the size of the US state of Delaware.

before we reach the 1.8 million mark we are going to have at least a halving for sure, that is going to drop the energy consumption by half based on the model.

This is true, but it is likely that transaction fees also would go up, as in a 25K tx/s Bitcoin on-chain transactions would only be used for very large operations and "settlements" (open/close payment channels, sidechain creation ...) where the users (mostly organizations and user groups) could afford large fees. With merged-mined sidechains the miners can also collect fees on these chains.

But in order for the network to become a success it NEEDS users online. So it can only become a success with high energy costs. Sure, people could start selling custom LN devices but I think it's reasonable to assume a lot of people would just use their PC. Smartphone would be possible too indeed, but it would drain a lot of battery. Most phones die withint 5-6 hours of activity already.

Most likely they would use their smartphone during the night, when it's connected to a power source, and their PC during the day. Or, as I already said, their router - in a "Bitcoin-as-a-world-currency-world" there would be sure router manufacturers that included LN and Bitcoin into the router's firmware, as Bitcoin would be one of the main Internet applications.

Obviously it's hard to really calculate it correctly. But fact of the matter is that it's a HUGE waste of energy, just because someone invented a strategy that pays more if you spend more energy. I really hope that, and assume, that the winning coin will use something like PoS instead of PoW. This is really unsustainable.

I also follow PoS and other possible competitors to PoW, like PoC (proof-of-capacity). PoC is, in reality, a form of PoW, but the cost structure of "validators" and the economics are different, and I assume it's more energy-efficient as hardware costs in PoC are more important than electricity consumption costs to determine its security.

However, we still don't know for sure if the Nothing-at-stake problem (PoS) and memory-time-tradeoffs (PoC) could not make these algorithms too insecure for a big cryptocurrency - although I'm cautiously optimist here and that's why I also support some PoS and PoC coins (well, afaik there's only one PoC coin as of now ...).

[drays]

Obviously it's hard to really calculate it correctly. But fact of the matter is that it's a HUGE waste of energy, just because someone invented a strategy that pays more if you spend more energy. I really hope that, and assume, that the winning coin will use something like PoS instead of PoW. This is really unsustainable.

Its not. PoW is perfectly sustainable, as soon as mining integrates into power infrastructure, and is performed with zero energy waste. In the long run, it could be done by municipalities as a public service - to support the world financial network - without mining profit being the driving factor.

We are not there yet, but if we are speculating about the future, lets be a bit more creative and don't assume everything in the future will be exactly as it is now, with only the numbers being bigger.

[deisik]

I´m tired of hearing the argument "but Bitcoin wastes more energy than
a whole country..." all over the mainstream media.

Here are a few reasons why the energy that is consumed by Bitcoin mining is not wasted:
1. The energy that is consumed by Bitcoin mining is the foundation of the security of the network
2. Proof-of-work is superior to other consensus algorithms like Proof-of-Stake and offers less attack vectors
3. Fiat currencies are supposedly backed by the power of the state... meanwhile the US military consumes more energy than Bitcoin mining per year
(this is a comparison of the military of a single country compared to a truly global currency)
4. No one is talking about the enormous costs of transporting fiat money, securing it and printing it in the first place
5. The mining of other commodities is extremely energy-intensive as well

If you take all these factors into account, Bitcoin may actually be less energy-intensive than traditional
fiat currencies. Besides, many mining companies have set up their operations in countries with an abundance
of renewable energies like Canada or Iceland. The days where Bitcoin was solely mined in rural Mongolia and China
may finally be behind us

Debunking the debunking

1. This point doesn't tell us anything. The energy consumed could be twice or twenty times as much and you could still claim the same
2. The superiority of the PoW model is not evident. PoW may be better than some PoS implementations but it is by far more energy-hungry
3. You can compare only comparable things. Comparing the energy consumption of the US military with Bitcoin mining is meaningless
4. Most money today exists in cashless form anyway, so this argument doesn't hold either
5. See point 3

In a nutshell, you have picked up some as random as irrelevant points and then proceeded to conclude that Bitcoin "may actually be less energy-intensive than traditional fiat currencies". Your only argument which can theoretically make sense is about security advantage of PoW over PoS, but, first, that remains to be seen, and, second, that has nothing to do with fiat currencies. Try better next time

[cryptokwuk]

And then there's modern Proof-of-Stake.

All Proof-of-Stake arguments and hate is from bitcoin maximalists that keep bringing up 5 year old arguments for issues with PoS that have long been solved.
It can actually be argued that modern PoS with decent distribution is more secure and more expensive and difficult to attack than PoW.

It does not matter if the energy is used or wasted. Superior alternatives are available and we should make use of that and not hold on to the past.

[pinkflower]

And then there's modern Proof-of-Stake.

All Proof-of-Stake arguments and hate is from bitcoin maximalists that keep bringing up 5 year old arguments for issues with PoS that have long been solved.
It can actually be argued that modern PoS with decent distribution is more secure and more expensive and difficult to attack than PoW.

It does not matter if the energy is used or wasted. Superior alternatives are available and we should make use of that and not hold on to the past.

You are wrong. The problem with pure PoS blockchains is the initial distribution of coins. How do we know the coins were distributed fairly and how do we know no one owns 51% or more of the coins.

A good Proof of Capacity implementation is a better solution than both PoW and PoS. Its cheaper, more energy efficient and doesnt cause environmental pollution as much.

Educate yourself.

A PoC and PoS comparison

Why Proof of Capacity should be taken seriously part 1

Why Proof of Capacity should be taken seriously part 2

Why Proof of Capacity should be taken seriously part 3