Topic: Energy consumption will become an issue if bitcoin really breaks through - page 5. (Read 23448 times)

@brenzi
Unfortunately   PoB (wriiten in that article) is more like science-fictional poetry artifact , than formal proposal (

Also, i am not sure it will be 100% compatible with Bitcoin protocol,
though it can be used as the base of other crytocoin.

Add. it will be hard to sell to Joe
 Average the idea of
"burning of your money"...

Energy consumption *could* become an issue? 
Energy consumption is a huge issue.  Some key facts which prove humans are extremely stupid:


1) More than half of all hydrocarbon energy of our spaceship earth already was burned off with no attempt at use or capture ("natural gas flaring").

2) Continued pouring of energy as visible light away from Earth at night for no fucking reason at all (see pictures of the Earth from space).

3) Horrendous inefficient use of energy for transport, including millions of folks bringing 2 tons of steel huge distances every day with them.

4) So-called "defense" or "military" using energy just to destroy other energy, in efforts to demonstrate how psychotic they are so other folks will be afraid of them. 


Yes, piss-poor use of energy resources is a *huge* problem on spaceship earth right now.  No, proof of work networks don't add meaningfully to the total.  But yes, it is something we should think about, so my apologies for getting upset.  It happens whenever the topic comes up.   
     

I tried to estimate this in this previous post. You are right that one should compare it to today's banking usage. But keep in mind that banking is not just about issuing and transferring money. There are services included that bitcoin doesn't supply and never will (funds/asset management, customer services, loans....)

I would be very interested in the figures, but they might be hard to get. But please read my linked post. If you can show me with quotes to reliable sources that the ratio of (paypal transaction volume)/(paypal plus credit cards energy use) is better than in the case of bitcoin, I'll be more than glad to change my opinion. I came to the conclusion that bitcoin's ratio is worse by orders of magnitude.

Flaff made a point that I think this entire debate was missing.  Yes, mining bitcoins has a price beyond just the cost of the ASICs, in electricity.  But how much energy do all the banks in the states use, not even counting the ones elsewhere in the world?  You could count just the energy the banks use directly, but I think it'd be more fair to count the banks' energy, the energy used by the armored cars transporting money, that used by the Federal Reserve, and so on and so forth.

That's not even getting into the energy used to encrypt credit card transactions online.  Actually, with the speed of computers currently, that's the cheap part of the cost: you just have to pay for an SSL certificate.  The expensive part is making sure websites comply with PCI Compliance and so on.  That may be derailing the conversation somewhat but it's to make a point: there's a real opportunity cost in accepting credit cards. 

Granted, if you're selling things online and not accepting credit cards, for most markets you're pretty much shooting yourself in the foot.  But the bitcoin protocal handles all that security so that if all the world used bitcoin instead of credit cards, e-commerce websites wouldn't have to worry about these things.  In other words: how much energy does PayPal use?  Even if credit card transactions are being done and not PayPal transactions, they are often still using PayPal as a credit card processor.

So the question to me isn't, does the bitcoin network use energy?  But rather, does the bitcoin network use more energy than the traditional financial network?  And, would it use more if it scaled up to the volume of, say, Visa and became a replacement for Visa in many cases.

As if banks don't use electricity.

The question is, less efficient than WHAT? Are they less efficient than using electricity to heat the solution to create the heat absorbtion effect? Yes. That, and the fact that they're not as compact as electric units.

If you don't already have a waste heat source, it's not worth the extra space consideration of such a system, even if it's a little more efficient, so nobody uses them.

Now, fast-forward to some distant point in the future. People are burning lots of juice to mine bitcoin. In the winter, this is a great system, because waste heat can be used for everything from heating buildings, to heating water (potable, swimming pools, hot tubs, whatever). In the summer this is a problem. You got countless BTUs of heat that you need to get rid of, as well as a building to cool (including keeping your mining hardware cool). Enter absorption refrigeration. You can now use those BTUs of waste heat to cool your building, or even do industrial refrigeration. Combine a mining farm with a food packaging plant? You were gonna burn that energy to refrigerate the food anyway, right? Why not harness that wasted energy and also mine some money. It's win/win/win. How about combining a mega-mining operation with a mega casino? Can you imagine what amount of electricity is spent on just cooling a casino floor? Let alone cooling 3000-5000 hotel rooms?

There's no reason that "waste heat" has to actually be wasted.

Thermodynamics are not my strongest field, but I just did some reading for the fun of it (the german article has more detail: http://de.wikipedia.org/wiki/Absorptionsk%C3%A4ltemaschine)

Probably one should use a lithium bromide / water system which needs >70°C input and produces 5°C output at a COP of 0.7

The 70°C might be just about acceptable for silicon devices. And the 5°C are perfect for air conditioning.

But this doesn't change anything in my criticism of bitcoins PoW concept. I just don't see why such energy consumption should be necessary just to make a currency stable. PPCoin has yet to mature and proof its stability - but it already shows a possible way out.

Actually, Proof-of-burn could be a way out of bitcoin's waste of real-world resources (https://en.bitcoin.it/wiki/Proof_of_burn)
At first sight, it seems to me that this approach would even be compatible with bitcoin. Maybe even without a hard-fork. However, I'm no expert of bitcoin protocol.

Basically this would replace spending fiat currency on hardware and energy by spending the cryptocurrency itself in such a way that the same randomness of possible reward results.

You got a good point there. Food for thought, thanks. PPC probably only works if people do not let banks or ewallet providers take care of their coins. But: Isn't this risk comparable to pooled mining in bitcoin? What about a BTC guild 51% attack?

what about the Absorption Refrigerator someone mentioned? Why are the electric ones mainly used in hotel rooms only so far? Can't be that they're less efficient then.


I'm afraid such a currency will soon find itself to be under total control of MtGox and (later) Goldman Sachs.

But heating with electricity is done right now in many places, because it's more convenient than other methods in some cases. In the scenario where bitcoin becomes the world's main currency, then all of those heaters could be dual-purpose devices that also produce bitcoin.

Also there is a lot of waste energy all over the world that could be used, since bitcoin can be produced any where in the world with electricity, and can be sent a lot more easily than electricity can be transmitted.

Finally, the block subsidy will eventually end, and the block generation reward will become only what block generators get from transaction fees, which could and likely will result in the block reward being a much lower percentage of the total market capitalization of bitcoin than it is now.

I'm very tempted by the Proof-of-Stake concept: https://en.bitcoin.it/wiki/Proof_of_Stake
It certainly addresses the issue of energy consumption. I'll need to do more reading to understand the possible drawbacks. The checkpointing is a violation of decentralization, but might be acceptable. What makes me sceptical is the fact that mining can only take place with an unlocked wallet. I can already hear people whining for lost coins....

Good reads:
http://ppcoin.org/static/ppcoin-paper.pdf
http://www.links.org/files/decentralised-currencies.pdf

Just got myself a couple of PPcoins ;-)

It's funny that the developers didn't even find the time to consequently search/replace the word bitcoin by ppcoin in their software....so I guess they're at a very early stage....

What will cause even more juice to be wasted. Transaction fees shift the economic equilibrium towards higher energy consumption.

You're talking about M0 money supply. Bills and coins. They make up for about 10% of the M2 money suppy. (http://money.howstuffworks.com/how-much-money-is-in-the-world.htm) including all money located on bank accounts and funds.
Money on bank accounts and funds is most probably less "costly" if you only count storage and transaction energy and infrastructure.

If you want to convince me, you need to do a little more research.

did fuel become an issue when cars broke through ? yes ... what happened well people started fighting for it and an exclusionary mechanism was introduced ...the raise in price ! )

hardly the same but I think you get my point ...

imo energy costs will translate into transaction fees which will get bigger to include a small 0.1-1% fee extra for all the juice we waste...



this is all imo however ...'

Repeat after me:  it's only more expensive in fiat.

I'm going to have to respectfully disagree. Basing future predictions on current status quo is just bad policy by just about ANY standard. I'm not saying that there won't be any bumps in the road, but what I am saying is that if it gets to the point that it begins to be problematic, the economic incentives will be enough to drive change (economic interests are the only TRUE driver of change anyway).

The idea that just because something is produced using another type of energy makes it not 'clean energy' is disingenuous at best. In order for that statement to be true, it would require that every cell produced consumes more 'dirty' energy in the the production process than it will produce in its working lifetime, which, if that were the case, solar energy would already make ZERO economic sense. People would never be able to come out ahead, as the solar installation would cost more in energy costs alone to produce, than would ever be recovered...

In the far future you might even be right. There are clever people around that think energy will be 100% renewable and almost for free in the future. But I think we should consider the status quo and not rely on utopia when judging weaknesses of today's bitcoin technology. Bitcoin value might rise a lot faster than renewable energy will progress.

In the case of Photovoltaics one should bear in mind that today PV cells are produced in china using non-renewable energy. So their embodied energy is not renewable at all.

As long as renewable energy is more expensive than fossil and nuclear energy, I consider excessive energy use as being bad because real costs are externalized.

It's an interesting point. But it also assumes that miners are helpless victims to the current power companies. If you look at just the half of the equation, which is the equation you posted, it seems that way, but if you look at it from the perspective that energy costs can also drive people to move toward their own energy generation (Solar, Wind, Tidal, etc), in order to be competitive in that market, the total energy consumption isn't as relavent as the source of that energy being consumed.

If you think of the trilions of gigawatts of energy that are wasted by not harnessing the Sun's power... There's not a big issue with consuming a lot of energy.

The issue is with consuming a lot of energy generated with 20th century technology.

The day it becomes less cost-effective to generate energy with 20th century technology is the day it gets abandoned by bitcoin miners (and other consumers alike).

So if the issue is just what percentage of current power generation will look like when compared to future energy consumption, I don't really see it as a problem, provided that energy is coming from clean sources, who cares how much of it gets used?

My opinion is that technological developements arnt being considered here.  As technology improves energy consumption will deminish and hashing will increase.

Brenzi, the cost of the hardware will factor heavily into it, especially as electrical efficiency increases. Like I mentioned earlier with amortization. If the machine that gets 1000x mhash/joule costs 1000x, the overall resources consumed is effectively the same making the assumptions about cost being equal to resources consumed. Since electricity can be renewable but beat-to-death hardware is not very recyclable, the drain on resources probably gets worse as electrical efficiency increases.