11 ms·

No, Bitcoin won't raise global average temperatures by 2°C. This Nature article was debunked by other researchers. I too was asked to review it (recently, post-

by mrb 8y ago

No, Bitcoin won't raise global average temperatures by 2°C. This Nature article was debunked by other researchers. I too was asked to review it (recently, post-publication) and the gist of what they did wrong in their analysis is: * They overestimate (probably by 3x-5x) the present power consumption of miners because they assume the distribution of types of mining machine is homogenous in this table: https://github.com/moracamilo/Bitcoin/blob/master/Randi_TableS1.csv https://github.com/moracamilo/Bitcoin/blob/master/Randi_Tabl... (when in fact 60-70% of the mining power comes from a single row in this table: Antminer S9). This error causes their calculated average efficiency (J/GH) to be much overestimated as their dataset contains mostly inefficient obsolete ASICs * They overestimate future power consumption by ~50x by assuming it grows linearly with the transaction rate (in reality it doesn't). Power consumption grows with miners revenues, proportional to {block rewards + transaction fees}. Fees account for currently ~2% of revenues (average of last 60 days). Rewards decline over time, so that fees will eventually account for close to 100% of revenues, which will happen on the authors timeline of 100 years. So if fees per transaction remaimed constant, we could see a 50x tx rate growth with no increase of power consumption. * They assume the proportion of CO2 emissions per kWh never improves over the next 100 years (great progress of renewables coming to a sudden stop?), and that Bitcoin consumes more fossil fuels than what the entire world currently consumes (infinite fossil fuel reserves?). It is not unrealistic to imagine the proportion of CO2 per kWh could be in a century 20% of what it is today. Overall, these errors combined mean their scenario overestimates CO2 emissions by a factor of about a thousand: (3 to 5) x 50 ÷ 20% = 750x to 1250x Edit: simplified bullet point #2

ForHackernews 8y ago
You're quibbling over details: The point is that bitcoin is deliberately designed to waste as much energy as possible. It's a constant arms race where you have to keep escalating your hash rate, or you risk somebody else being able to execute a 51% attack. It's an absolutely atrocious example of the Red Queen effect ("running just to stand still") and I can't believe that technologists who in other contexts value elegance and efficiency can defend bitcoin with a straight face. I can only assume they have a lot of money tied up in bitcoin, because otherwise it makes no sense.
- atrus 8y ago
  Well, you need to spend some sort of resource to ensure the integrity of your money supply. The dollar is backed by a few million troops. Bitcoin is backed by a few million tons of coal.
  - soVeryTired 8y ago
    That's a pretty roundabout way of thinking. Plenty of countries have troops but an unstable currency (e.g. Venezuala). Plenty have a stable currency but no significant troops (e.g. Iceland).
    - PycptN2MH 8y ago
      Militarily robust sovereignty is necessary but not sufficient for the integrity of a fiat currency. Governments such as Iceland borrow military power through international agreements.
      - soVeryTired 8y ago
        Robust against what?
        PycptN2MH 8y ago
        Threats to national sovereignty such as invasion.
        soVeryTired 8y ago
        I don't really see how protection against invasion helps you maintain a fiat currency. I don't think an army is either necessary or sufficient. These just seem like two different issues.
        PycptN2MH 8y ago
        Do you have an example of a nation failing to repel an invasion yet maintaining the integrity of its currency?
moonbug 8y ago
Scratch a libertarian, invariably uncover a clime change denier.
mrpopo 8y ago
* They assume the proportion of CO2 emissions per kWh never improves over the next 100 years (great progress of renewables coming to a sudden stop?) I mostly agreed with the first two bullet points, but please show me a significant country that managed to reduce their CO2 emissions with renewables. For instance, in China, where a lot of Bitcoin mining happens, the share of renewables increases in many countries, but not even as fast as the actual consumption by the population, leading them to open more coal power plants as well. I'll wait until I see it.
- ThomPete 8y ago
  France who uses nuclear.
  - kasey_junk 8y ago
    Nuclear is not generally seen as a renewable is it? It’s a really efficient extraction based source.
    - ThomPete 8y ago
      Its plentiful and its greener than both wind and solar. Furthermore i am assuming the point is reduction of co2, if thats the case then does it really matter in this context?
      - kasey_junk 8y ago
        I think it matters only because there is a strong contingent of alternative energy supporters that do not see nuclear as part of the mix when talking about ‘renewables’ and for those folks the math is really hard. Conversely the math is easy with a strong nuclear mix.
        ThomPete 8y ago
        Oh I get that but it also illuminates the shallowness of their position given that the reason they are against wind and solar and hate fossil is because of CO2. Nuclear is literally the best combination of the best features from each of the other forms of energy.
Retric 8y ago
> by assuming it grows linearly with the transaction rate That’s a reasonable approximation. Over time mining is paid for by transaction fees in a competitive market. If you get 2x the fees and the cost to mine stays the same, you get 2x as much mining. Granted that assumes transaction fees are ~1:1 with transaction count. However, protection from a 51% attack eventually requires 1:1 increase in mining as the value of a 51% attack increase, so the lower bound is there.
- burgerboy 8y ago
  Yes I'm sure you know better than the researcher who put more than five minutes into looking into it
- nybble41 8y ago
  Mining would scale with the transaction fees, but there is no reason to assume that transaction fees alone would match the current block reward. We don't need the amount of mining we have now just to defend against 51% attacks.
  - X6S1x6Okd1st 8y ago
    That depends entirely upon how motivated an attacker is and how many blocks organizations wait for.
- mrb 8y ago
  "That’s a reasonable approximation" No, because the authors don't understand how miners gain revenue. For every dollar of miners revenues, currently $0.02 come from tx fees and $0.98 come from the block reward (12.5 BTC today.) Therefore with everything remaining equal (same average fee per tx, which is the authors' assumption), if the tx rate grew 50x it would increase miners revenues to $1.98, hence only doubling revenues and power consumption.
  - Retric 8y ago
    It’s got a fair amount of volatility. The Reward (last 24h) 1,688 + 115.2 BTC https://bitinfocharts.com/bitcoin/ https://bitinfocharts.com/bitcoin/ So, 115.2 / ( 1,688 + 115.2) = 6.4% not 2%. However, looking at historical data the current fees are well below the cap. https://bitinfocharts.com/comparison/bitcoin-transactionfees.html https://bitinfocharts.com/comparison/bitcoin-transactionfees... Now, currently Bitcoin can’t scale transactions so extrapolation is kind of up in the air. But, in practice if it’s going to survive it needs to create value to offset the cost of mining.
lvs 8y ago
You're debating numbers you don't like by making up other numbers. Whatever. The point is that it's a massive waste of energy and emissions, and that's undeniable.
- frontloadpro 8y ago
  Who says? People want money that isn't controlled by the government. Is moving gold lower cost?
modzu 8y ago
can you please cite your source? was the refutation also published in a peer reviewed journal?
- modzu 8y ago
  didnt think so. hundreds of comments, some lengthy, and nobody knows whats really going on? also, i mean, maybe some cynicism ought to be tolerated if misinformation is? its hard work to do real, rigorous science, which is what journals like nature represent. but then we have this echo chamber in here where its been "refuted" by internet experts in their parents basements? (allegorical). i know there are/used to be educated people on hn, not just businesspersons. where did they go? hn is ostensibly about intellectual curiosity. but you cant actually satisfy intellectual curiosity with makebelieve. someone has to stand up for the truth, and for the blood and sweat that makes discovering truth possible: the scientific method. starting from doubt or skepticism, form a hypothesis, test it: observe and predict. theorize. replicate. repeat.
- once_inc 8y ago
  The refutation (https://medium.com/coinmonks/the-reports-of-bitcoin-environmental-damage-are-garbage-5a93d32c2d7 https://medium.com/coinmonks/the-reports-of-bitcoin-environm...) is a medium article. Please try not to let that influence your opinion about the article though, because it does rightly point out the flaws in the original calculations, and the validity of the peer review of the original article.
plutonorm 8y ago
I have half a suspicion that it's the fact that it's burning through lots of fuel that is driving it's adoption. Processes in general seem to be driven by a kind of 'pressure' between low entropy and high entropy regions. This includes life itself. If life has a purpose it is to increase the flow across these low/high entropy regions. I don't think it's too outlandish to suppose some kind of hidden process that drives systems in general to optimise this flow. Bitcoin is just such a system, it's doing this transfer perfectly. Something tells me it's going to be extremely difficult to stamp out and I will be surprised if it does not continue to rise in value and energy consumption at an exponential rate, unless something systematic is done to disrupt it's progression. Buy bitcoin.
- arcticbull 8y ago
  95% of transaction data is fake. We have no idea what adoption actually looks like. [1] The bitcoin price is "based on a true story" in true Hollywood fashion. Its design makes it impossible to stop the rampant and obvious manipulation. The game is rigged. No amount of technical analysis or philosophy is going to help you predict the outcome of a rigged game. [1] https://www.forbes.com/sites/cbovaird/2019/03/22/95-of-reported-bitcoin-trading-volume-is-fake-says-bitwise/#20a1b6f66717 https://www.forbes.com/sites/cbovaird/2019/03/22/95-of-repor...
  - flatfilefan 8y ago
    We shouldn’t confuse trading for fiat money with on-chain transactions.
    - arcticbull 8y ago
      I'm curious why they should get a pass. They're a first-class part of the ecosystem, and pretending exchanges don't exist to goose the numbers feels dishonest. Further, it's the exchanges that set the "price" of bitcoin relative to fiat, which in turn sets the exchange rate for BTC to goods in on-chain transactions. Unless you're suggesting there's an "exchange" price for goods and a "black market" price for goods that each value Bitcoin differently?
thisisit 8y ago
I am really curious about the points you make in your post. > This Nature article was debunked by other researchers. Who are these other researcher? Care to link to their research on this? > when in fact 60-70% of the mining power comes from a single row in this table: Antminer S9 How do we know that? > It is not unrealistic to imagine the proportion of CO2 per kWh could be in a century 20% of what it is today. Again, how do we know that?
- davidgerard 8y ago
  > Who are these other researcher? Care to link to their research on this? When a bitcoiner says something is "debunked", what they mean is that someone wrote a post on Medium. In this case, it's an investment banker who put some money into electricity companies before he put money into Bitcoin: https://hackernoon.com/the-reports-of-bitcoin-environmental-damage-are-garbage-5a93d32c2d7 https://hackernoon.com/the-reports-of-bitcoin-environmental-... Digiconomist's response: "Verify, don't trust" - he shows his working. https://digiconomist.net/verify-dont-trust/ https://digiconomist.net/verify-dont-trust/
- mrb 8y ago
  "Who are these other researcher? Care to link to their research on this?" It takes months to write, peer-review, and publish research papers, so given that this Nature Climate Change article is quite recent at the moment there exists absolutely zero peer-reviewed studies or counterstudies following up on it. However here are some researchers who have criticized it in their own informal analysis or press interviews: see links in https://is.gd/ksyKyv https://is.gd/ksyKyv — shortened URL for https://thinkprogress-org.cdn.ampproject.org/v/s/thinkprogress.org/experts-debunk-dangerous-and-misleading-study-hyping-bitcoin-energy-use-8f8744672611/amp/?amp_js_v=0.1&usqp=mq331AQICAEoATgAWAE%3D#origin=https%3A%2F%2Fwww.google.com&prerenderSize=1&visibilityState=visible&paddingTop=54&p2r=0&horizontalScrolling=0&csi=1&aoh=15408320314945&viewerUrl=https%3A%2F%2Fwww.google.com%2Famp%2Fs%2Fthinkprogress.org%2Fexperts-debunk-dangerous-and-misleading-study-hyping-bitcoin-energy-use-8f8744672611%2Famp%2F&history=1&storage=1&cid=1&cap=swipe%2CnavigateTo%2Ccid%2Cfragment%2CreplaceUrl https://thinkprogress-org.cdn.ampproject.org/v/s/thinkprogre... The researchers in question are: Dr. Jon Koomey, who was a Lawrence Berkeley National Laboratory (LBNL) scientist for more than two decades. Professor Eric Masanet, who leads the Energy and Resource Systems Analysis Lab at Northwestern University. Arman Shehabi, research scientist at Lawrence Berkeley National Lab. And myself who has criticized it here on HN (my work's been cited in Nature Sustainability and the NYT: http://blog.zorinaq.com/bitcoin-electricity-consumption/ http://blog.zorinaq.com/bitcoin-electricity-consumption/ so I guess this makes me a researcher now) "How do we know that?" It's common knowledge to anyone in the industry that the Antminer S9 is by far the most popular Bitcoin mining machine. For some official reports on that, see market research reports and IPO filings from multiple mining hardware manufacturers. For example this has been neatly researched and presented by the team lead by Christian Stoll in https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3335781 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3335781 Read their Supplementary Data, tab "3.4 IPO filing analysis" "Again, how do we know that?" As I said elsewhere in this thread, according to https://www.hsdl.org/?abstract&did=801297 https://www.hsdl.org/?abstract&did=801297 CO2 emissions per kWh have already decreased -25% over 11 years. Therefore -80% in 100 years seems easily achievable (the trend only has to continue for 60 years, then there can be 40 years of technological stagnation.) In fact we don't even need any technological breakthrough to achieve -80%. France already emits 90% less CO2 per kWh than the average country, thanks to nuclear power.

by mrb 8y ago

ForHackernews 8y ago

You're quibbling over details: The point is that bitcoin is deliberately designed to waste as much energy as possible. It's a constant arms race where you have to keep escalating your hash rate, or you risk somebody else being able to execute a 51% attack. It's an absolutely atrocious example of the Red Queen effect ("running just to stand still") and I can't believe that technologists who in other contexts value elegance and efficiency can defend bitcoin with a straight face. I can only assume they have a lot of money tied up in bitcoin, because otherwise it makes no sense.

atrus 8y ago
Well, you need to spend some sort of resource to ensure the integrity of your money supply. The dollar is backed by a few million troops. Bitcoin is backed by a few million tons of coal.
- soVeryTired 8y ago
  That's a pretty roundabout way of thinking. Plenty of countries have troops but an unstable currency (e.g. Venezuala). Plenty have a stable currency but no significant troops (e.g. Iceland).
  - PycptN2MH 8y ago
    Militarily robust sovereignty is necessary but not sufficient for the integrity of a fiat currency. Governments such as Iceland borrow military power through international agreements.
    - soVeryTired 8y ago
      Robust against what?
      - PycptN2MH 8y ago
        Threats to national sovereignty such as invasion.
        soVeryTired 8y ago
        I don't really see how protection against invasion helps you maintain a fiat currency. I don't think an army is either necessary or sufficient. These just seem like two different issues.
        PycptN2MH 8y ago
        Do you have an example of a nation failing to repel an invasion yet maintaining the integrity of its currency?

moonbug 8y ago

Scratch a libertarian, invariably uncover a clime change denier.

mrpopo 8y ago

* They assume the proportion of CO2 emissions per kWh never improves over the next 100 years (great progress of renewables coming to a sudden stop?) I mostly agreed with the first two bullet points, but please show me a significant country that managed to reduce their CO2 emissions with renewables. For instance, in China, where a lot of Bitcoin mining happens, the share of renewables increases in many countries, but not even as fast as the actual consumption by the population, leading them to open more coal power plants as well. I'll wait until I see it.

ThomPete 8y ago
France who uses nuclear.
- kasey_junk 8y ago
  Nuclear is not generally seen as a renewable is it? It’s a really efficient extraction based source.
  - ThomPete 8y ago
    Its plentiful and its greener than both wind and solar. Furthermore i am assuming the point is reduction of co2, if thats the case then does it really matter in this context?
    - kasey_junk 8y ago
      I think it matters only because there is a strong contingent of alternative energy supporters that do not see nuclear as part of the mix when talking about ‘renewables’ and for those folks the math is really hard. Conversely the math is easy with a strong nuclear mix.
      - ThomPete 8y ago
        Oh I get that but it also illuminates the shallowness of their position given that the reason they are against wind and solar and hate fossil is because of CO2. Nuclear is literally the best combination of the best features from each of the other forms of energy.

Retric 8y ago

> by assuming it grows linearly with the transaction rate That’s a reasonable approximation. Over time mining is paid for by transaction fees in a competitive market. If you get 2x the fees and the cost to mine stays the same, you get 2x as much mining. Granted that assumes transaction fees are ~1:1 with transaction count. However, protection from a 51% attack eventually requires 1:1 increase in mining as the value of a 51% attack increase, so the lower bound is there.

burgerboy 8y ago
Yes I'm sure you know better than the researcher who put more than five minutes into looking into it
nybble41 8y ago
Mining would scale with the transaction fees, but there is no reason to assume that transaction fees alone would match the current block reward. We don't need the amount of mining we have now just to defend against 51% attacks.
- X6S1x6Okd1st 8y ago
  That depends entirely upon how motivated an attacker is and how many blocks organizations wait for.
mrb 8y ago
"That’s a reasonable approximation" No, because the authors don't understand how miners gain revenue. For every dollar of miners revenues, currently $0.02 come from tx fees and $0.98 come from the block reward (12.5 BTC today.) Therefore with everything remaining equal (same average fee per tx, which is the authors' assumption), if the tx rate grew 50x it would increase miners revenues to $1.98, hence only doubling revenues and power consumption.
- Retric 8y ago
  It’s got a fair amount of volatility. The Reward (last 24h) 1,688 + 115.2 BTC https://bitinfocharts.com/bitcoin/ https://bitinfocharts.com/bitcoin/ So, 115.2 / ( 1,688 + 115.2) = 6.4% not 2%. However, looking at historical data the current fees are well below the cap. https://bitinfocharts.com/comparison/bitcoin-transactionfees.html https://bitinfocharts.com/comparison/bitcoin-transactionfees... Now, currently Bitcoin can’t scale transactions so extrapolation is kind of up in the air. But, in practice if it’s going to survive it needs to create value to offset the cost of mining.

lvs 8y ago

You're debating numbers you don't like by making up other numbers. Whatever. The point is that it's a massive waste of energy and emissions, and that's undeniable.

frontloadpro 8y ago
Who says? People want money that isn't controlled by the government. Is moving gold lower cost?

modzu 8y ago

can you please cite your source? was the refutation also published in a peer reviewed journal?

modzu 8y ago
didnt think so. hundreds of comments, some lengthy, and nobody knows whats really going on? also, i mean, maybe some cynicism ought to be tolerated if misinformation is? its hard work to do real, rigorous science, which is what journals like nature represent. but then we have this echo chamber in here where its been "refuted" by internet experts in their parents basements? (allegorical). i know there are/used to be educated people on hn, not just businesspersons. where did they go? hn is ostensibly about intellectual curiosity. but you cant actually satisfy intellectual curiosity with makebelieve. someone has to stand up for the truth, and for the blood and sweat that makes discovering truth possible: the scientific method. starting from doubt or skepticism, form a hypothesis, test it: observe and predict. theorize. replicate. repeat.
once_inc 8y ago
The refutation (https://medium.com/coinmonks/the-reports-of-bitcoin-environmental-damage-are-garbage-5a93d32c2d7 https://medium.com/coinmonks/the-reports-of-bitcoin-environm...) is a medium article. Please try not to let that influence your opinion about the article though, because it does rightly point out the flaws in the original calculations, and the validity of the peer review of the original article.

plutonorm 8y ago

I have half a suspicion that it's the fact that it's burning through lots of fuel that is driving it's adoption. Processes in general seem to be driven by a kind of 'pressure' between low entropy and high entropy regions. This includes life itself. If life has a purpose it is to increase the flow across these low/high entropy regions. I don't think it's too outlandish to suppose some kind of hidden process that drives systems in general to optimise this flow. Bitcoin is just such a system, it's doing this transfer perfectly. Something tells me it's going to be extremely difficult to stamp out and I will be surprised if it does not continue to rise in value and energy consumption at an exponential rate, unless something systematic is done to disrupt it's progression. Buy bitcoin.

arcticbull 8y ago
95% of transaction data is fake. We have no idea what adoption actually looks like. [1] The bitcoin price is "based on a true story" in true Hollywood fashion. Its design makes it impossible to stop the rampant and obvious manipulation. The game is rigged. No amount of technical analysis or philosophy is going to help you predict the outcome of a rigged game. [1] https://www.forbes.com/sites/cbovaird/2019/03/22/95-of-reported-bitcoin-trading-volume-is-fake-says-bitwise/#20a1b6f66717 https://www.forbes.com/sites/cbovaird/2019/03/22/95-of-repor...
- flatfilefan 8y ago
  We shouldn’t confuse trading for fiat money with on-chain transactions.
  - arcticbull 8y ago
    I'm curious why they should get a pass. They're a first-class part of the ecosystem, and pretending exchanges don't exist to goose the numbers feels dishonest. Further, it's the exchanges that set the "price" of bitcoin relative to fiat, which in turn sets the exchange rate for BTC to goods in on-chain transactions. Unless you're suggesting there's an "exchange" price for goods and a "black market" price for goods that each value Bitcoin differently?

thisisit 8y ago

I am really curious about the points you make in your post. > This Nature article was debunked by other researchers. Who are these other researcher? Care to link to their research on this? > when in fact 60-70% of the mining power comes from a single row in this table: Antminer S9 How do we know that? > It is not unrealistic to imagine the proportion of CO2 per kWh could be in a century 20% of what it is today. Again, how do we know that?

davidgerard 8y ago
> Who are these other researcher? Care to link to their research on this? When a bitcoiner says something is "debunked", what they mean is that someone wrote a post on Medium. In this case, it's an investment banker who put some money into electricity companies before he put money into Bitcoin: https://hackernoon.com/the-reports-of-bitcoin-environmental-damage-are-garbage-5a93d32c2d7 https://hackernoon.com/the-reports-of-bitcoin-environmental-... Digiconomist's response: "Verify, don't trust" - he shows his working. https://digiconomist.net/verify-dont-trust/ https://digiconomist.net/verify-dont-trust/
mrb 8y ago
"Who are these other researcher? Care to link to their research on this?" It takes months to write, peer-review, and publish research papers, so given that this Nature Climate Change article is quite recent at the moment there exists absolutely zero peer-reviewed studies or counterstudies following up on it. However here are some researchers who have criticized it in their own informal analysis or press interviews: see links in https://is.gd/ksyKyv https://is.gd/ksyKyv — shortened URL for https://thinkprogress-org.cdn.ampproject.org/v/s/thinkprogress.org/experts-debunk-dangerous-and-misleading-study-hyping-bitcoin-energy-use-8f8744672611/amp/?amp_js_v=0.1&usqp=mq331AQICAEoATgAWAE%3D#origin=https%3A%2F%2Fwww.google.com&prerenderSize=1&visibilityState=visible&paddingTop=54&p2r=0&horizontalScrolling=0&csi=1&aoh=15408320314945&viewerUrl=https%3A%2F%2Fwww.google.com%2Famp%2Fs%2Fthinkprogress.org%2Fexperts-debunk-dangerous-and-misleading-study-hyping-bitcoin-energy-use-8f8744672611%2Famp%2F&history=1&storage=1&cid=1&cap=swipe%2CnavigateTo%2Ccid%2Cfragment%2CreplaceUrl https://thinkprogress-org.cdn.ampproject.org/v/s/thinkprogre... The researchers in question are: Dr. Jon Koomey, who was a Lawrence Berkeley National Laboratory (LBNL) scientist for more than two decades. Professor Eric Masanet, who leads the Energy and Resource Systems Analysis Lab at Northwestern University. Arman Shehabi, research scientist at Lawrence Berkeley National Lab. And myself who has criticized it here on HN (my work's been cited in Nature Sustainability and the NYT: http://blog.zorinaq.com/bitcoin-electricity-consumption/ http://blog.zorinaq.com/bitcoin-electricity-consumption/ so I guess this makes me a researcher now) "How do we know that?" It's common knowledge to anyone in the industry that the Antminer S9 is by far the most popular Bitcoin mining machine. For some official reports on that, see market research reports and IPO filings from multiple mining hardware manufacturers. For example this has been neatly researched and presented by the team lead by Christian Stoll in https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3335781 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3335781 Read their Supplementary Data, tab "3.4 IPO filing analysis" "Again, how do we know that?" As I said elsewhere in this thread, according to https://www.hsdl.org/?abstract&did=801297 https://www.hsdl.org/?abstract&did=801297 CO2 emissions per kWh have already decreased -25% over 11 years. Therefore -80% in 100 years seems easily achievable (the trend only has to continue for 60 years, then there can be 40 years of technological stagnation.) In fact we don't even need any technological breakthrough to achieve -80%. France already emits 90% less CO2 per kWh than the average country, thanks to nuclear power.