Bitcoin uses a hell of a lot of power!

in #bitcoin8 years ago (edited)

The amount of electricity used per Bitcoin transaction is stupidly high

While it may not cost a lot of power to broadcast a transaction to the Bitcoin network, once you take into account all the mining that occurs the cost per Bitcoin transaction is a remarkably high!

Here is a back-of-the-envelope calculation that will give us a reasonable lower bound on the amount of electricity used per transaction. One of the ASIC miners released 2016 claims that it expends 0.1J / gigahash (GH) and this is considered very efficient. Currently the hash rate is about 1,500,000,000 GH/s. Thus if everyone was using this particular machine the combined power would be 150,000,000W or 150MW. In a day (24 hours) this equates to 3600MWh

The number of transactions is currently sitting at about 225,000 per day. Dividing 3600MWh by 225,000 yields a cost per transaction of 16kWh/transaction!
This is just a lower bound for the following reasons:

  • this only holds if every miner is using this particular machine which is top of the line and considered very efficient.
  • this does not take into account electricity which is not used for mining but merely for running nodes

In fact this article puts the figure at 46kWh per transaction. This is enormous. It is about 1.57 American households worth of electricity per transaction. To put it another way, each transactions costs the same as turning on your dryer for a little over 15 hours.

I don't know about you all but I think this clearly shows that Proof of Work may ultimately be a flawed idea and that a successor algorithm (Proof of Stake?) is an imperative.

Sort:  

I disagree completely. The electricity expended on Proof of Work is a sunk cost that 'charges the value battery' of every bitcoin.

bitcoin_value = electricity ( cryptographic signatures + network trust )

Compare that to something like, a dollar, (which you use to pay for turning your drier on) ...

dollar_value = (gdp - entitlements + taxes) / corruption

I agree on the network trust value, but I would disagree on the mining costs.

The cost of mining has a direct impact on LIQUIDITY.

And I mean by liquidity:

The degree of an asset's ability to be converted to cash or to make transactions at its fair market price

With a low liquidity, a currency loses value as it becomes more and more difficult to make transactions.

What made bitcoin rise is its liquidity, and especially on some markets at the begining (silkroad...).

Hence, it you want bitcoin to rise, you must facilitate transactions. Especially if you aim mass adoption (ie. e-commerce at Amazon scale)

Those equations you have written there contain non-quantifiable variables (e.g. what is the SI unit of corruption?)

Not only that, I get the strong sense that the equations were just made up on the spot and have not been rigorously tested or even theorised about on a firm basis.

You are right. In the long term, improvements in the way the network handles transactions are to be expected. What other possible solutions have you explored including Proof of Stake?

No, Proof-of-Work will stay. Proof-of-Stake is not the same level of security, by an order of magnitude.

In theory, Proof-of-Work should eventually disappear in the ecological equation, because competition will drive mining to be worthwhile only where the waste heat can be used (and where no other sources of heat than electrical are applicable). Think of heating swimming pools with miners etc...

I wish to make something clear that I did not in the article. I understand that without Proof of Work there is no Bitcoin. Consensus cannot be reached without it. But it is proof of useless work. One simply finds a hash that is less than a given value and this becomes harder to do over time. It is not useful work like modeling protein folding or even calculating prime numbers. Short story: I get the need for it but it is useless work for anything but consensus.

But I do think that this service that miners provide comes at too high a cost. Perhaps there is a way to get the same service at a lower cost through an algorithm that has not yet been devised.

I just don't think I can get behind any argument that says 46kWh is an acceptable electricity cost per transaction. It offends me from a sheer efficiency point of view that every time I make a transaction it's the same as turning on my dryer for 15 hours.

Congratulations @seafood! You have received a personal award!

Happy Birthday - 1 Year on Steemit Happy Birthday - 1 Year on Steemit
Click on the badge to view your own Board of Honor on SteemitBoard.

For more information about this award, click here

By upvoting this notification, you can help all Steemit users. Learn how here!

Congratulations @seafood! You received a personal award!

Happy Birthday! - You are on the Steem blockchain for 3 years!

You can view your badges on your Steem Board and compare to others on the Steem Ranking

Vote for @Steemitboard as a witness to get one more award and increased upvotes!

Coin Marketplace

STEEM 0.18
TRX 0.16
JST 0.029
BTC 62555.95
ETH 2435.55
USDT 1.00
SBD 2.64