Bitcoin network cost is OK now, but may soon be hugely wasteful

[go1111111]

You've all heard the environmental/waste argument "the Bitcoin network uses too much computing power."

Right now I don't think this is a good argument. We're not spending that many resources mining if you make the comparison with the cost of securing banks and credit cards. However, if bitcoins jump in price like a lot of us hope they will, this will be a legitimate problem.

Imagine that the price of BTC reached $1,000,000 USD on January of 2016. The block reward will still be 25 BTC, meaning that every day, 3.6 billion dollars worth of bitcoins is distributed to miners. Miners would be expected to spend almost 3.6 billion dollars per day to mine these coins. That's 1.3 trillion dollars per year, or almost 2% of the wealth produced globally every year (as of right now).

Of course in this situation, the Bitcoin network will be very widely used and worth protecting from attacks. The question is: is 2% of global wealth devoted to protecting the Bitcoin network overkill? Maybe it would have enough protection with just 1% of global wealth, or even 0.2%.

The general problem is that the rewards to miners will be somewhat arbitrary and not calibrated to the security needs of the network, because they're based on parameters that Satoshi picked before he knew what the adoption/price curve would look like.

[vitalemontea]

Why do you assume block reward will be 25BTC if/when 1BTC is 1 million?

[jonanon]

Interesting topic.

I think there are just too many variables that could alter too may things to be trying to predict what will happen tomorrow let alone in 2016.

It's all a lot of suck it any see at the moment! 

[human]

Exactly this thought keeps me thinking that BTC is overvalued even right now.

[go1111111]

Why do you assume block reward will be 25BTC if/when 1BTC is 1 million?

I'm not assuming it, I'm saying that if it were to happen, it would lead to a waste of resources.

The point is that the sooner the price gets high (before the block rewards have a chance to drop off significantly), the more likely we are to be in a wasteful situation where people are providing way more security than the network requires.

[LAMarcellus]

Ignore Federal Reserve Notes for a moment. That's just a unit of account and a very poor one at that. If everything was priced in BTC then how much would it cost to secure the network?
The same amount of electricity it uses to secure the network today. Not 2% of the world electricity.

The cost in true wealth, ie. food water, oil, electricity, homes, time, labor   will remain static or will grow as more miners come online.

Don't get hung up on the Federal Reserve Note price. That's a flawed way of using the word "wealth".   Federal Reserve notes are just the currency you are forced by law to use today, as a medium of exchange for actual wealth, babes, booze, and other material goods.

Try to think of it in term of "how much wealth, how much electricity as a percentage of the world total does this system use".

Do not conflate paper notes with wealth.
Have you read about Zimbabwe or the Weimar republic?  People there had trillion dollar notes but that wouldn't buy a loaf of bread.


What percentage of the worlds air, power, and people does it take to secure the network? That is the cost.
The amount of FRNs isn't an accurate way to measure this cost because FRNs lose value thru time as more and more are counterfeited into existence. If you dont understand what I'm saying then I highly encourage you to look into Weimar Germany and other hyperinflationary currencies.

[mgio]

BTC is not going to be worth 1 million a coin in 2 years. That is where you are wrong so the rest of your argument makes no sense.

[stompix]

You've all heard the environmental/waste argument "the Bitcoin network uses too much computing power."

Right now I don't think this is a good argument. We're not spending that many resources mining if you make the comparison with the cost of securing banks and credit cards. However, if bitcoins jump in price like a lot of us hope they will, this will be a legitimate problem.

Imagine that the price of BTC reached $1,000,000 USD on January of 2016. The block reward will still be 25 BTC, meaning that every day, 3.6 billion dollars worth of bitcoins is distributed to miners. Miners would be expected to spend almost 3.6 billion dollars per day to mine these coins. That's 1.3 trillion dollars per year, or almost 2% of the wealth produced globally every year (as of right now).

Of course in this situation, the Bitcoin network will be very widely used and worth protecting from attacks. The question is: is 2% of global wealth devoted to protecting the Bitcoin network overkill? Maybe it would have enough protection with just 1% of global wealth, or even 0.2%.

The general problem is that the rewards to miners will be somewhat arbitrary and not calibrated to the security needs of the network, because they're based on parameters that Satoshi picked before he knew what the adoption/price curve would look like.

Let's assume it goes to 1 billion. Just for fun.
For your computing , that would mean 2000% of the wealth produced globally.

And now , it seems there is a problem with this , right?
Spot it and you have the answer to your question in the first place.

[Xyver]

I wrote this paper on this exact topic a few days ago:

https://docs.google.com/document/d/1Lb_iLMmH5sIiYfGeHr9tCTnaoWjK25zDlvO0nXXgEng/edit?usp=sharing

The TL,DR is that we're fine now, but as Bitcoin grows in value we may have a problem with electricity usage.  We (I) must reanalyze later!

[mgio]

BTC is not going to be worth 1 million a coin in 2 years. That is where you are wrong so the rest of your argument makes no sense.

oh, also, and if it were, the block reward would no longer be 25 BTC. It would be 12.5 BTC.

Why?

Well, a 2,000 fold increase in the price of bitcoin would cause a huge demand in mining. You agree since you believe people would spend 3.6 billion per day mining.

Right now many ASICs can achieve 1 GH/s per watt of power. Let's assume ASICs don't get more efficient (they will) and assume an average electricity of $0.15 kwh. 3.6 billion dollars a day in electricity is enough for 1 TRILLION watts of power. or a total network hashrate of 1 x 10^12 * 1 x 10^9 = 10^21 or 1 zettahash/s worldwide hash rate. The current world hashrate is roughly 10 petahashes (1 x 10^16). That means the world hashrate would have to increase by 10^21/10^16 = 10^5 = 100,000 times. This would make the difficulty be around 100 trillion.

This would mean the world hashrate would have to double roughly 17 times. The hashrate would have to increase at such a ridiculously high rate that we would got through blocks so quickly that we would hit the next block having date in far less than two years from now. Thus the block reward would only be 12.5 BTC.

I'm too lazy to actually calculate how long it would take to hit the next halving. I'll let someone else do it.

[stompix]

I wrote this paper on this exact topic a few days ago:

https://docs.google.com/document/d/1Lb_iLMmH5sIiYfGeHr9tCTnaoWjK25zDlvO0nXXgEng/edit?usp=sharing

The TL,DR is that we're fine now, but as Bitcoin grows in value we may have a problem with electricity usage.  We (I) must reanalyze later!

You have some maths wrong in your paper. 11kw average consumption of a house , it's more like 33.
And why compare it to paypal or google?
Compare that to the electricity banks consume.
In my parents hometown ~ 50k they have 21 banks , 2 of them are 3 floors high and over 1000mp2 . Let's compare how energy they waste and how much the bitcoin network does.

[stompix]

BTC is not going to be worth 1 million a coin in 2 years. That is where you are wrong so the rest of your argument makes no sense.

oh, also, and if it were, the block reward would no longer be 25 BTC. It would be 12.5 BTC.

Why?

Well, a 2,000 fold increase in the price of bitcoin would cause a huge demand in mining. You agree since you believe people would spend 3.6 billion per day mining.

Right now many ASICs can achieve 1 GH/s per watt of power. Let's assume ASICs don't get more efficient (they will) and assume an average electricity of $0.15 kwh. 3.6 billion dollars a day in electricity is enough for 1 TRILLION watts of power. or a total network hashrate of 1 x 10^12 * 1 x 10^9 = 10^21 or 1 zettahash/s worldwide hash rate. The current world hashrate is roughly 10 petahashes (1 x 10^16). That means the world hashrate would have to increase by 10^21/10^16 = 10^5 = 100,000 times. This would make the difficulty be around 100 trillion.

This would mean the world hashrate would have to double roughly 17 times. The hashrate would have to increase at such a ridiculously high rate that we would got through blocks so quickly that we would hit the next block having date in far less than two years from now. Thus the block reward would only be 12.5 BTC.

I'm too lazy to actually calculate how long it would take to hit the next halving. I'll let someone else do it.

Excellent point!

Reward-Drop ETA: 2016-10-17 03:57:01 UTC

And now

Reward-Drop ETA: 2016-09-14 06:35:24 UTC (142 weeks, 2 hours, 20 minutes)

We're gonna have a "party" a few months earlier.

[Xyver]

You have some maths wrong in your paper. 11kw average consumption of a house , it's more like 33.
And why compare it to paypal or google?
Compare that to the electricity banks consume.
In my parents hometown ~ 50k they have 21 banks , 2 of them are 3 floors high and over 1000mp2 . Let's compare how energy they waste and how much the bitcoin network does.

I'm trying to get more data, that was just a first draft.  I think I can also better estimate the average watts:gigahash ratio, based on what companies have shipped what amount of power.

I did the original comparisons to other computing giants, I really want data on how much power Visa consumes.  Banks would be nice, but I imagine they'd be hard to find....

I want the data on how much power it takes to print, store, and transfer money.  If anyone can help me out, PM me, and lets get this paper more factual!

[go1111111]

That's just a unit of account and a very poor one at that. If everything was priced in BTC then how much would it cost to secure the network?

My argument does not rely on any exchange rates, and is about waste of real resources. Substitute "1000 ounces of gold" or "200,000 blocks of cheese" as the price of a bitcoin and it applies just the same.

BTC is not going to be worth 1 million a coin in 2 years. That is where you are wrong so the rest of your argument makes no sense.

Try to look at the general point. If it goes to $50k, the same logic applies.

Let's assume it goes to 1 billion. Just for fun.
For your computing , that would mean 2000% of the wealth produced globally.

And now , it seems there is a problem with this , right?
Spot it and you have the answer to your question in the first place.

The error in your analogy is that it's impossible for us to spend 2000% of our current GDP on anything in two years, whereas it wouldn't be impossible for us to spend 2%.

I wrote this paper on this exact topic a few days ago:

https://docs.google.com/document/d/1Lb_iLMmH5sIiYfGeHr9tCTnaoWjK25zDlvO0nXXgEng/edit?usp=sharing

The TL,DR is that we're fine now, but as Bitcoin grows in value we may have a problem with electricity usage.  We (I) must reanalyze later!

Nice analysis! I think this shows that the bitcoin network is perhaps using more resources than necessary even right now. It is using more power than Facebook, but doing a lot less useful processing (look at the number of requests/transactions Facebook is processing). One could argue that the Bitcoin network needs to be extra secure in the early stages so as to avoid an attack that could kill it before it develops. On the other hand, one could argue that a 51% attack might kill Bitcoin (or might not -- it could easily recover after some manual editing of the blockchain), but that another cryptocurrency would just take its place, and that protecting Bitcoin itself is not that critical.

I also think in the future, people will spend up to almost the entire value of block rewards to get them (once ASIC manufacturers create the technology to make their latest ASICS, producing extra ones is very cheap). So IMO your arguments based on difficulty are red herrings in the long run. The cost to run the Bitcoin network will be about the value of the block rewards.

 

[tom.hashemi]

Imagine that the price of BTC reached $1,000,000 USD on January of 2016.

Yeah, but that just isn't going to happen...

[DobZombie]

Imagine that the price of BTC reached $1,000,000 USD on January of 2016.

Yeah, but that just isn't going to happen...

He said imagine, not "is this possible?"

I believe tom is giving us a thought exercise, isn't this what the forums are for.  Thinking up situations that could possibly happen, and what those consequences would be.

I for one think that the argument is a valid one.  As the price rises, the more miners jump on board.  The more miners, the difficulty increases.  Difficulty increases, miners want a price increase to cover it.  So on and so forth.  It all started about a month after the reward halving, and then when ASICs hit the scene then the price skyrocketed.

It's not an instant effect, but it's there.

[tom.hashemi]

He said imagine, not "is this possible?"

I believe tom is giving us a thought exercise, isn't this what the forums are for.  Thinking up situations that could possibly happen, and what those consequences would be.

I for one think that the argument is a valid one.  As the price rises, the more miners jump on board.  The more miners, the difficulty increases.  Difficulty increases, miners want a price increase to cover it.  So on and so forth.  It all started about a month after the reward halving, and then when ASICs hit the scene then the price skyrocketed.

It's not an instant effect, but it's there.

I assume when you said tom you meant the OP?

We can all dream up situations that could possibly happen and then concoct some kind of thought exercise, sure. But what's the point in dreaming up a totally implausible thought exercise? There is no worth to it.

What is far more useful is to imagine a scenario that could realistically happen and then explore around that point.

The reason this situation is implausible is because the time it would take to get to a $1million per btc price tag would allow for significant technological developments, making mining more efficient as well as energy infrastructure, energy generation etc.

There is simply no point in conjecturing around something which is so far into the future that you cannot even begin to comprehend how different even the most basic functions of life will be. If Bitcoin does have an effect comparable to the Internet, then the world we will be inhabiting in 15 years time (when the price per bitcoin will still not be anywhere near $1million) will be totally different, having gone through a new round of innovation and economic evolution. Then imagine how it will look if we do reach 1btc=$1million. I cannot even begin to comprehend what that world would look like, hence this debate is pointless.

Anyway, I realise I'm just being annoying now so I'll bow out and leave you to your conjecturing. Happy Christmas!

Tom

[DeepCryptoanalist3]

Bitcoiners why you ignoring the fee? It contradicts to your view on Bitcoin? Every time when bitcoins are mentioned there is a false statement that transaction is free of any fee. Sorry but there is a fee. Bitcoin isn't a free to use currency. Fee will grow through the time. It will be connected to an amount of transfered money probably. Miner simply will ignore transactions with small fees.

[go1111111]

The reason this situation is implausible is because the time it would take to get to a $1million per btc price tag would allow for significant technological developments, making mining more efficient as well as energy infrastructure, energy generation etc.

This misunderstands the argument. It doesn't matter how cheap electricity gets. If every day there are one billion USD worth of bitcoins distributed to miners, then miners will spend almost one billion dollars per day trying to get these bitcoins. This is standard economic theory, and you can see why this is true via a proof by contradiction. If miners were spending significantly less than 1 billion USD per day, then any miner that started spending more money on mining would make a profit.

There is simply no point in conjecturing around something which is so far into the future that you cannot even begin to comprehend how different even the most basic functions of life will be.

We have very good reasons to believe that Bitcoin will NOT somehow overturn the most foundational truths of economics. This argument is so general that it doesn't depend on the exact particulars of the future.  

Bitcoiners why you ignoring the fee?

The fee is irrelevant to the argument I'm making.

[Explodicle]

How much is spent protecting the fiat currency system? All those guns and drones... If bitcoins are ever worth $1M each, it means people will be wasting less energy on fiat.