Bitcoin Forum

Using reason and logic only. Leave the emotion and speculation to the other threads please!!!

1. Assuming silicon ASICs have now approached their limit, how much energy (in Watts) would be required if there were 10bn people on the planet and there was an average of 10 transactions per person, per day?

2. Is there any way to get the transaction time down to below 10 seconds? (i.e. suitable for buying a coffee or paying for my groceries)

3. Is centralization of the network inevitable?

1. Transaction volume does not have any direct co-relation with energy consumption of ASICs. ASICs find blocks where transactions are added.

2. 10 seconds !!! I think you meant to say 10 minutes. This is in the bitcoin protocol that at every 2016 blocks found, difficulty adjusts itself to average out the transaction confirmation time to 10 minutes. Only a hard fork can change it. But, that is not required in reality. For a coffee or groceries confirmations are not required as double spend will be more costly than the gain attacker will have. But, if you are selling a car or a house, wait for 6 confirmations, i.e. around an hour.

3. Only time can say.

Let me match your request with another request: If you want to get useful answers please do some research first. Your questions betray misunderstandings of the most basic bitcoin concepts. When you post you consume other people's time.

The energy requirement for processing a transaction are very small and are completely unrelated to mining ASICs. Mining ASICS do not process transactions. They provably expend energy to make reversal of the history of transactions infeasible and for a given security level consume the same amount of energy regardless of the transaction level.

You are confusing transaction time and transaction confirmation time. Bitcoin transactions are basically instantaneous (just the time it takes to communicate the data) but it takes time for the transaction to become irreversible. The same is true for other payment mechanisms, e.g. credit card transactions can often be reversed for months, checks take weeks to clear... Making a 1:1 comparison is hard because Bitcoin is used in different ways from traditional payment systems (including anonymously). But it is not the case that the 10 minutes mean time between blocks itself has any fundamental implication for buying coffee.

Thats up to the users of Bitcoin.

That's all very fine. The "transaction" is useless unless it's confirmed by miners. The energy question (10bn people * 10 transactions per day) still stands.


The consumer and the retailer don't care about the technical nuance you've described (that a transaction is separate to a confirmation).


I agree. I guess that would be a bad thing (i.e. potential for abuse), right?

That is a common misunderstanding.
There is no need for confirmation for micro transaction.
To buy a cup of coffee or a meal does not need confirmation.
Buying a house or a car, then would a couple of confirmations properly be a good idea before the buyer drive away with the car.

Pay attentions to gmaxwell analogy with credit cards, then will it be more clear to you how bitcoin works.

The energy spent is approximately half the mining profit (since people would spend more energy to mine if you can make money from it).  So it roughly depends on how much the people are willing to pay for the transactions (at least at the time where most mining profits come from transaction fees).

This is not a technical question but an economic one.

If the coffee shop worker knows the guy in the coffee shop, everything is fine.

You cannot run a railway station coffee stand when you don't know 95% of your customers.

Consumer to retailer transaction via credit card is not an anonymous transaction so you're not comparing like with like.

I think the answer to question 2 is "No". In the context of transaction time, both cash and plastic are superior to Bitcoin. Though I'm happy to go through the logic of how it could be otherwise.

I guess transaction fees will have to go up to incentivise the miners? It's not a problem right now, but in a 10bn people doing x10 transactions a day, how much of a problem would it be? Can a farmer pay a supplier $5 with his mobile without having to incurr 10%/20%/30%? transaction fees?

I would love to see an econometric analysis of this based on where we are now and where we're going if we keep going at the same rate (factoring in price, market cap, price of electricity, mining difficulty, transaction fees, etc.) But I don't have the knowledge or expertise.

The cash and credit card is not superior (imo).
You have the chance of false bills and credit card fraud.

There is no such things as a fake bitcoin.

The stolen CC I pay my coffee with is certainly anonymous. Just because the plastic card has a name on it does not mean I am giving away my personal information.


Payment processors could handle transactions like these, but honestly most of the time I pay for coffee they dont even check if the money I give them is real. Its a risk they are willing to take because the majority of the customers are honest. Frankly most of the customers are honest because they want to come back and as someone that once took a fake 200 € bill: my boss didnt care. Its like the robery that happens statistically once every X. You calculate with it and adjust your prices accordingly or get insurance that covers these things.


The more payments are made, the smaller the fee per payment. Assuming the blocksize is not a limit in your scenario.


Go ahead, I did know nothing about bitcoin a year ago.

I don't want to get into a trust discussion. The plastic facilitates a trust relationship.

The reality is that Bitcoin simply cannot compete with cash or plastic when it comes to transaction time. Maybe you can bolt on a trust/insurance service (i.e. hack the Bitcoin protocol), but even then I don't see how that would work without both parties identifying each other. Cash wins.

I find your rationale unconvincing. I remain very skeptical that:

- Bitcoin is a usable real-world currency (a world where the majority of human-human transactions are around the $10 mark)
- The energy required to keep the network going is sustainable
- The network will remain decentralised

,particularly when cryptocurrency technology has moved on (a lot) since 2008.

I would argue that Bitcoin is outdated and impractical.

The answer to your question (2) is "yes", you have unfortunate misunderstood how bitcoin works.

Cash has obvious problems, to mention a few:
1. You have to trust the government which issue them.
2. False money in circulations.
3. Personal safety risk when carry cash.
4. Biological hazzard, bacterial and virus.
5. Difficult to use cross border.
6. Not feasible for international trades.
7. Not convenient for large trades.
8.
9.
10.
11.

Okay, so not only do you not search first, you apparently choose not to read: I just explained that the energy used by mining at a given security level is unrelated to the volume of transactions. It is the same regardless if there is one transaction or one hundred million transactions. The amount of energy related to the count of transactions is zero. So, there you go: If you insist on getting a binary answer to your ill poised question instead of learning, the answer is zero.

They may well not, and so what of it?

Transaction time with credit cards is not better than with Bitcoin. Both are (almost) instant when it comes to processing the transaction, but the actual confirmation can take as long as 90 days as CC payments can be reverted during this period for a variety of reasons.

Cash is obviously instant and irreversible, but comes with a bunch of its own drawbacks.

Like I said, I'm not an econometric expert (though you seem to be an expert on Bitcoin).

At one extreme: free mining is available anywhere on earth (phase 5)

At the other: all miners are switched off (phase 0).

In the middle somewhere, we range from:
phase 1: just a few miners are functioning (the big value/important transactions get mined, the small transactions aren't worth it)
phase 2: a good few miners are functioning (the system is only good for big and medium transactions)
phase 3: lots of miners functioning (everyone can send BTC about, but the small guys are beginning to feel it)
phase 4: a plentiful supply of miners (pretty much everyone gets to transact with hardly any restriction)

- The days of 4 are over
- We're in 3 and will be moving into 2. Silicon has stalled and electricity isn't getting any cheaper.

To calculate the timing on the transition from 5-> 4 -> 3 -> 2 -> 1 -> 0 would require much more detailed analysis.

I think this side-discussion is irrelevant to the original question. It's obvious the answer is "No" - you cannot get below 10 seconds. This has very serious practical implications for the deployment of Bitcoin at a global scale.

You're still getting it wrong. The number of miners has no effect on how many transactions can be processed. Whether it's 1 miner or 1 million miners, it's all the same.

The only real limit to the number of transactions that's currently in place is the maximum size of each block, but there are plans to increase this and so far there's not enough activity yet to hit this maximum other than in exceptional spikes.

edit:
You can get below 10 seconds when you accept 0-confirmation transactions. For everyday purchases, this is perfectly fine as it is impractical and uneconomic for an attacker to try and profit from double spending those. Payment processors such as BitPay and Coinbase already accept 0-conf transactions and most merchants that accept Bitcoin-payments use one of those two processors.

For very large payments, you will have to wait for one or more confirmations, but these payments are almost always for things that are not delivered instantly (cars, houses, etc...).

Are you trolling?

The transaction time is almost instant, as gmaxwell pointed out.

I didn't mention cash in question 2. Yet now I "have unfortunate [sic.] misunderstood how bitcoin works".

Why are you too proud to admit that the Bitcoin micropayment system is inferior to cash (in terms of time to confirmation of the transaction)?

Look buddy... bitcoin do have some shortcomings. But, those you are pointing out are not the ones. Those are simply result of your lack of understanding.

Hi Rannasha,

Thank you for your insight. I will need to read more on the bold bit.

The average consumer and the average retailer don't care about the techie geeky stuff that you find so important. They want an instant exchange that works.

No absolutely not, I use cash from time to time, and it can be be faster than a bitcoin transaction.

I respect that you want this topic to be on a technical level, but since you ask so directly then will I have to state that I do not find cash superior, it has a fundamental flaw, it is a fiat based system, which I am strongly against. I would have no problem making micro payment with somethings else, as long as it has an intrinsic value.

Hi TookDk. I agree. I think fiat is flawed.

I think cryptocurrency should be designed to match everything that cash does, and then go further.

Currently there are serious shortcomings in the way Bitcoin was implemented. I don't think it's possible to simply ignore/explain-away these fundamental flaws.

I like to think of Bitcoin a steam engine  - a great symbol of the industrial revolution. The steam engine still has a role in the modern world, but more for nostalgia's sake (of course there are also the trainspotters and the die-hard train hobbyists). Let's cherish Bitcoin and inherit its achievements. But we can/must do better. I worry that at the moment, all we've got is a big old steam engine and we're channeling all our efforts into keeping the old beast running. You cannot transport the world's population on steam.

And they do have it. Instant transactions that are safe enough for everyday's expenses. Only you here seem to fail to understand how it works. Greg Maxwell, one of the most active and senior Bitcoin Core dev, has taken quite a bit of his precious time to explain it, repeatedly.

IMHO you seem to be here more to push an unsubstantiated claim that Bitcoin is impractical and cannot compete with traditional means of payments (i.e. credit/debit cards and/or cash) than to educate yourself. Your condescendance regarding 'the techie geeky stuff that [we] find so important', considering that you started yourself a technical discussion in the 'Development & Technical Discussion' section, can only reinforce that suspicion.

What exactly is your point here?

I understand about the unconfirmed transactions. While not ideal, it's an acceptable compromise for some.

Energy usage of the network is a far, far bigger issue (i.e. question 1., which feeds into question 3.)

Fair enough. Still, I'm not sure that you have understood Greg Maxwell's answer when he explained that the number (volume) of transactions does not affect the overall power usage of miners. "Mining" a block and confirming 0, 100, or 100 million transactions within that block does not affect power usage (or just marginally in the case of really huge volumes; for the sake of clarity, I will leave the blockchain size / storage problems aside here -- not denying these!).

Right now, the amount of energy used by mining might arguably be uneconomical or disproportionate because of the relatively high subsidy of 25 BTC per block until next halving, and the block reward is huge compared to the total fees of the transactions included in that block. However, the rapid decrease of block reward on the long run (half-life is 4 years) means that the main retribution for miners will eventually come from transaction fees instead, creating an economical equilibrium (i.e. supply vs demand) between the power costs of mining and the fees from transactions, i.e. the total cost of consumed energy will have to be paid by transaction fees.

I will concede that it's not easy to understand these concepts, and I'm not pretending Bitcoin is flawless. A lot of issues are still to be solved, but I don't feel like they are exactly the ones that you mentioned in your OP.

Regarding your question 1, you could run the bitcoin network today on a regular computer - in fact that is essentially the way it happens with full nodes, they all keep copies of the blockchain which along with the software (which defines the protocol to a large extent) defines the network. Then energy usage would be negligible.  It has nothing to do with the number of transactions per block, the limit could be 10 times higher, probably 100 times higher without a large change in energy usage.  Maybe 1000 or 10000 in a few years as Moore's law continues along.  I doubt it would double, but even if it increase by a factor of 10 or 100, it would still be negligible compared to an air conditioner or a small building.

The tradeoff there would be the security of the network.  The amount of hashing power is what secures the network, so you need a lot of power to protect the integrity of the bitcoin network and blockchain, this is the "work".  

There are two components that we're discussing here:  running the network, and securing the network.  In short, the number of transactions is not what uses the energy, it is the number computers (be they CPUs, GPUs, or now ASICs) securing the network.  

Regarding number 3,  I do not believe that centralization is inevitable, no.  As you stop orders of magnitude improvements as you had between generations, e.g. CPU->GPU->(FPGA?->)ASIC, it is much easier to predict profitability of a miner.  I believe as that time approaches, it will be easier for individuals to return to mining because they will not have to hope that they get a machine in time to have a ROI and just depend on luck to do so.  With CPU and GPU mining, we were not dominated by large concerns running farms.  Could hashing become concentrated?  Sure.  Pools could be an issue, but miners can easily switch pools - or use p2pool.  Could it occur because the value of the bitcoins in the network becomes so large that it is smart to do so?  Sure.  I don't believe it is inevitable though, it is an open question.  Ask in 10 or 20 or 50 years and the answer will be clearer.

Regarding number 2, you could have a sidechain with a faster confirmation time or any other number of solutions if you wanted some confirmation for a small transaction quickly, although as has been stated this is really unnecessary for many day-to-day transactions just as using a counterfeit detection pen on a $1, $5, $10, or $20 bill is unnecessary (and I've used $50 and $100 bills and no one has checked them too) and just as you do not need to copy the ID of every person who purchases from you using a debit or credit card.  At this point double-spends for small purchases are unlikely to be profitable to try, and unlikely to be successful.  And if they were occasionally, it is part of the cost of doing business.

BTW, you should not dismiss the difference in definition between transaction and confirmation and all the other technical details if you truly want to understand the network and the implications of suggestions.  If you truly are not trolling, think about what people are explaining here and the fact that there are a lot of technical details that deal with many issues.

Is bitcoin perfect?  I doubt it.  With innovations in the technology - few of which are trivial to implement in a running network (it is like upgrading an airplane at 35,000 feet) - it can be improved and become closer to an ideal system.

 :-)


* Somewhat simplified in a few spots.

Zero-confirmation transactions could be verified against
double-spends by connecting to other nodes.  BIP 35
provides a way to check a node's memory pool via
a special message.
 

???


The answer to that question seems rather obvious to me.


Sounds like just another alt-coin preaching troll to me.

I do preach alt-coin (XRP, specifically). And I make no apologies for that. I tell friends and colleagues (when it comes up in conversation) to be very careful of BTC and to keep a watchful eye on XRP.

The only crypto-currencies I use are XRP and BTC. I'm about 50%/50% in each.

I have an interest in BTC, but remain skeptical about its future and am prepared to leave permanently (but not just yet).

If you wish to continue branding people who want answers to BTC's fundamental flaws as "trolls" (which is neither helpful nor conducive to civilised chat), then I wish you the very best of luck with your business.

Good spotted.


Eamorr, will you please close this thread and return to xrptalk.org - and stop wasting everyone time here, you are obviously only here to troll and provoke.

Like I said - I have 50% of my crypto wealth in BTC. I have skin in the game.

I asked for answers and I got them. I even learnt new things. Even then, the die-hard BTCers are too proud to admit that their currency might have fundamental flaws.

I don't wish to continue on this thread any more. The atmosphere is hostile and unwelcoming. If BTC want adoption on a global scale, the BTC community will need to look in on themselves and sort out this attitude.

Best of luck with everything.

*beats a normally docile dog with a stick repeatedly and laughs at it*

*cries about how they should put down this rabid dog when it retaliates with what is no more than a warning nibble*

0-confirmation transactions can also be made safer. For example, I have a 2of2 multisig account with GreenAddress. If I were to pay with that account, the coffee shop (or even the car dealer) should have no problem accepting the 0-confirmation transaction as if I had just handed them cash. Their computer or point of sale terminal could do something like this (in the background, without the cashier needing to know anything about the technical details):

1) Get notification of incoming transaction: almost immediately after I send it (within a few seconds at most)
2) Check if the transaction passes bitcoind's isstandard() tests and contains at least the minimum miner's fee: if yes, proceed to step 3, else decide (based on the value of the transaction) whether to accept the 0-conf or wait to see if it gets mined.
3) Check if the inputs are from a 2of2 script: if yes, proceed to step 4, else decide whether to accept the 0-conf
4) Ask GreenAddress if it's one of their instant confirmation transactions: if yes, accept the transaction immediately, else decide whether to accept the 0-conf transaction

Meanwhile, my money is safe (GreenAddress can't run with it) and as easy for me to access as any mobile or web wallet. It's even safer than a web wallet since a signature from me and a two-factor auth is required to get that second signature from GreenAddress, and they can't run with my money because I have the second key and a timelocked transaction to refund the money to me. All transactions happen on the blockchain, I doubt GreenAddress is going to risk its reputation by signing a transaction that double-spends and if they ever do they'll almost immediately lose the trust of any merchants who rely on them. If a transaction passes the above four tests, you can be assured that it will eventually make it into a block and treat it as if it's already confirmed.

While it's true that this currently only works with GreenAddress, that's only because nobody else is doing it yet. Almost everything required is already part of the Bitcoin protocol and those parts that aren't are slowly being bolted on (for example, using the proposed payment protocol, it may be possible to eliminate step 4 entirely.. or to at least make it easier to have multiple instant-confirmation services). Personally I'd like to see more safe instant-confirmation wallets, not because I don't like GreenAddress, but because I do like healthy competition and decentralization (or at least reduction of centralization). If we had more wallets with that feature, and the payment protocol to make it easier to manage (on the merchant's end and in my mobile wallet), confirmations would be a non-issue even for larger transactions.

Someone shows up claiming to ask for answers to "technical questions":

Then ignores, dismisses the "technical nuances", speculates about the future, starts name-calling, states certain topics are off-limits, and gets into opinion:

And finally after he has wasted a lot of people's time, he resorts to saying he only wants "answers" after dismissing any he didn't like, even when correct technically:
Clearly someone came here not wanting to know the answers, but to argue argue or dismiss the technical points he claimed to want answers about. 

I'm not sure what that behavior is called.   ::)

Why do you feel the need to come out with this little rant 48 hours later?

My questions were largely answered by the constructive members.

Most people I know, when presented with new facts, re-orientate their views based on facts. You, on the other hand, seem to use anti-facts to reinforce irrational beliefs (yes, beliefs). Bitcoin is a religion to you and, like a Muslim terrorist in Paris, there's no telling you otherwise.

I wish you the best of luck with your business interests.

I actually wrote a long response above too (https://bitcointalk.org/index.php?topic=920007.msg10109111#msg10109111). 

48 hours? Again, doesn't appear that way from here.

Your reply is rich with irony regarding "new facts", seems like you are projecting your attitude.

Glad there is an ignore on here still!

The processing power required to mine a block is completely unrelated to the number of transactions included in that block. Mining a block with 5 transactions will cost just as much (or as little) as mining one with 10,000 transactions.

If that's true, no retailer would ever accept credit cards, as they effectively have a confirmation time of 6+ months.

If you still think that's a technical nuance, let me spell it out for you:

		Credit card		Bitcoin
Transaction time1		5-10 sec		1-2 sec
Confirmation time2		3-12 months		10 minutes

1 = the time it takes to make a payment
2 = the time it takes for a payment to become irreversible


No. P2P mining.

We've ascertained that. The block chain size still needs a band-aid.

Comparing apples with oranges.

Were you sick on the day the teacher was explaining algebra?

Lol. In this case, will the electricity usage be even more obscene?

Uhm, why exactly?

You seem to think are are all kinds of problems, shortcomings or limitations with Bitcoin, that aren't really there.

No, comparing credit card payments (claiming a transactions takes only seconds) to the 10 minute confirmation time of Bitcoin, is comparing apples with oranges.

Bitcoin is faster in terms of sending money around, and WAY faster in terms of transactions becoming irreversible.

That's funny, with me actually being an algebra teacher.

Obscene, you say? ??? (and why the hell it would be "even more" obscene with P2P mining is completely beyond me)

Anyway:

https://i.imgur.com/qrrDTZp.png
https://i.imgur.com/qcqKt7O.png
https://i.imgur.com/HImaRcg.png

Source: http://www.coindesk.com/microscope-conclusions-costs-bitcoin/

Your point again?

What a load of bollox.

Bitcoin is a tiny, tiny, tiny percentage of the global economy.

If 10bn people were doing 10 transactions a day, at the rate Bitcoin is going, each village on the planet would need it's own power station.

You are a retard. No amount of large letter assertions will change this.

I wish you all the best with your business.

10bn people doing 10 transactions a day can be processed with EXACTLY the same amount of miners as today.

You still don't seem to understand this, so I'll say it again: the processing power required to mine a block is completely unrelated to the number of transactions included in that block. Mining a block with 5 transactions will cost just as much (or as little) as mining one with 10,000 transactions.

Ah, the infamous ad hominem, if every other (lack of) argument failed.

Bitcoin is not a business for me. More like a fascination and appreciation of how a smart idea and ingenious invention has the potential to make a difference, and take away some power of the establishment. Just like the upcoming of the internet did.

Factually incorrect. What about the block size? How to change that? Bitcoin2.0? Why bother? There are other cryptos that have addressed all these issues.

An announcement that everyone needs to now install Bitcoin 2.0 would crash the market to 0 overnight.

I already provided you with several central refutations. You are a Bitcoin zealot and there's no talking to you.


Best of luck with the trainspotting. There are serious folks out there who have invested huge personal energy, time and money in better alternatives. People who envisage a brighter future for humanity - there's serious work to do.

Folks like you want to strap a jet engine onto a steam engine.

http://i.kinja-img.com/gawker-media/image/upload/s--JNk6QLSj--/c_fit,fl_progressive,q_80,w_636/fk67wsxn38hwkorbfkme.jpg

Plenty of options currently being discussed. Increasing block frequency, increasing max block size over time, pruning block chain at certain intervals, side chains, etc. Look at Gavin Andresen's recent posts on this. Scalability is not an issue.

Bitcoin isn't even 1.0 yet. Changes like this can be discussed by developers and the community, and once the majority agrees, these can simply get integrated in the current Bitcoin clients to become effective at a conveniently distant point in the future. No need for announcements that "everyone needs to upgarde now" or whatever.

No, you made some statements or assumptions that were simply incorrect, and now you try and come up with one excuse after the other for not properly understanding of how Bitcoin actually works.

OK, then don't.

Great, I'm all pro diversity! I suggest you move to their forums and start buggering them, if you obviously have no faith in Bitcoin whatsoever. Healthy competition is only good, may the best alternative win.

Sigh. That's your problem right there. Bitcoin is a religion to you.

Shit, you got me. Yes, I'm not into Bitcoin for technical reasons or its ingenious design, or the new possibilities this technology offers, or because it's all mathematically sound, or for the fact that it replaces trust in people or established powers with trust in open source, transparent cryptography, nahh... I'm all into it because I just have faith.

You win, sir. This also makes all your previous statements true. Enjoy your victory!

You'll find you'd actually agree with me (if you weren't so blinded by your religion).

The idea and theory of cyptocurrency is sound.

It's implementation (via Bitcoin) is not.

Care to elaborate? I'm always interested in good arguments for or against Bitcoin or alternatives. But the things you mentioned in your OP are simply not actual problems.

As for the transaction time vs confirmation time: let me illustrate this by the fact that I pay for coffee, dinner, groceries, and many other products and services with Bitcoin on a daily basis. I'm not talking about webshops or virtual online business, but typical point-of-sale situations. It works. 100% of the time. I pay in literally one second, and obviously I never EVER have to wait for the confirmation time.

I'm not saying Bitcoin is perfect, or the End To All Problems, or even near finished (strictly speaking it's still in development phase). But as a means to pay fast, efficiently, secure, both online AND offline, it already does a damn fine job.

10bn people doing 10 transactions a day? Yeah right...

Ignoratio elenchi.

Kazimir already explained the blocksize is
irrelevant to the energy usage.

The fact that the blocksize is limited
is a wholly separate issue; one that
is currently being debated now as
to the best approach to address it.

If you want to use other cryptos,
go for it.  No one is stopping you.

You still didn't make an argument why this wouldn't be perfectly possible.

The mining performance or energy usage is completely unrelated to number of transactions.
And as for block size: there are tons of options currently being discussed, just a matter of getting consensus of what's the best way to go. There is NO fundamental issue or problem here, Bitcoin is extremely scalable.

You sound like someone in the late 90s saying: this 'internet' thing may seem nice for electronic communication, like some emails and text websites, but downloading complete movies, or even streaming them at multiple megabytes per second in ordinary households? "Yeah right..."

The internet of today isn't built with 28k modems.

The only vision of the future that Bitcoin has is a 36k modem.

You seem to be unaware of many new developments, improvements, and scalability features already being discussed and planned.

Nobody who's serious about Bitcoin thinks we'll be using (the equivalent of) 36k modems.

Could you share with us the system that achieves decentralized consensus while having the capability of processing 100 billion transactions per day?

So you admit that Bitcoin can't do it?

100 billion transactions per day is roughly 1 million transactions per second. Not an unachievable number.

That's what we need to be aiming for.

Strapping a jet to a steam engine is not the solution.

yawn.  many avenues for scalability but you are only interested in arguing.  have fun.

You're avoiding the question.

Developments like Factom (http://factom.org/) however...

1 million transactions per second, using the Bitcoin blockchain, there you go.

Hey that's great. And it's good to see people are 1) acknowledging the problem and 2) doing something about it.

I hope they don't get left behind. Because a lot of work has been done and a lot more needs to be done.

Sure, Bitcoin is still work in progress. Nobody ever claimed it was finished.

But I'd say in the state we are now, Bitcoin is already way, WAY ahead of any alternative. Especially considering the path towards global recognition and mainstream acceptance.

Fair enough.

Bitcoin is currently ahead, sure. But it's brand has been very badly damaged - Mt Gox, drugs, criminality. I don't believe its adoption is any way appealing to government and/or banks. All I can say is I hope this current crash unwinds itself in an orderly manner.

http://www.coingecko.com/ gives a very good overview* of what's currently happening in Crypto-space

*the XRP community is far more active when you replace forum.ripple.com with xrptalk.org

I missed this post on a previous page:
Eh, did you ever try and actually pay with Bitcoin in real life? Say, dinner, or even just a cup of coffee? You never, EVER have to wait for a confirmation. Which is perfectly safe, because even just an attempt to double spend (with slim success rate at best) is a very, VERY expensive endeavor.

There is no logic in spending $10000 (in terms of mining power) on a feeble attempt to maybe cancel a $100 payment.

It's 100% safe to accept unconfirmed Bitcoin transactions for every day payments, like groceries, restaurant bill, you name it. Sure, if you're selling cars or houses for Bitcoin, you may want to wait for a confirmation or two. Otherwise, NO issue whatsoever.

And transaction time with Bitcoin is 1-2 seconds. Plastic is typically somewhat slower (it has to connect with the bank, and credit card company, etc). And so is cash: having to deal with change, cash registers, counting pennies and dimes and $1 notes, etc. Bitcoin is the absolute superior alternative here.

1)  Zero watts.  There is no energy requirement for bitcoin mining.  Competition drives the energy use. 

2)  Zero confirmations are nearly instant.  This is what are used by all in person POS shops which I have used (including restaurants, coffee, groceries).  There is a risk of finney attack double-spend but that risk is less than the risk these shops have previously taken to accept credit cards (chargebacks are much easier).   

3)  No.

Because this is wrong if you consider non local exchanges, which is the case of all internet services and the norm for many country.
Let's say you want to buy a movie on demand, you can always try to send your dollar bill to netflix... maybe next month you'll be able to see it.

For local exchange, yes, a bitcoin transaction is reversible. So is credit card and check, but are merchant refusing them ?

But it is a lack of perspective thinking that a BTC can't be represented in paper money. In that it would be very interesting to see banks act on such service.
Nothing prevent the creation of a paper denominated in BTC to fix to the problem. You'll can redeem x Bits against a bill to a trusted bank, as much as you could redeem gold for dollars in the past.

Such paper money would not fall in the same pitfall than the dollar, because paper BTC would be marginal for commerce.
People would use them only for local exchange, but the majority of money flows are digital nowadays, so the great majority of BTC exchanged will stay "real". There is not a lots of room for such paper to take over btc and become a fiat (ie, backed by nothing).

I believe that since such paper money would be less portable and less divisible than real BTC, there is no risk that history repeat itself, but still be very useful for hand to hand exchanges. (specially in poorly connected area)
The origin of the apparition of fiat is because it was easier to transact with fiat and credit money than shipping gold.
Which can't be the case with BTC.

Also, people originally  accepted piece of paper because they thought it was the equivalent to the amount of gold. (well, thanks to proof of reserve, there is no belief to have anymore, only proof)

Free. Stop the sophistry and just admit that Bitcoin's energy usage is obscene.

Why tolerate these risks when we have the know-how to resolve these issues?


Miners are switching off. Miners are now more likely to be in a centralized data centre than under your desk. That brings risks.

I think Bitcoin is a wonderful advancement. We would not be here where we are today without Bitcoin. She's like a big old steam engine that everyone has fond memories of.

But this bitcointalk.org blind loyalty to a crypto with underlying cracks (yes, cracks that have been filled in - some filled in better than others) is moving into Luddite territory.

What is obscene and why should we care ? no rhetoric just want to know why it bother you.

I argue that without proof of work, we can't have a currency according to the regression principle of Mises. https://bitcointalk.org/index.php?topic=928324.0 (https://bitcointalk.org/index.php?topic=928324.0)

What risk ? The biggest miners are loosing more than the small one (suffering tons of unused hardware), so I'd say it is good for decentralization.
Why should a user care at all that miners are closing ?

At the rate we're going, we'd need a power plant in every village (10bn people doing 10 transactions a day) to power the financial system. We could pay all the power station workers in Bitcoin, eh?

That's right. You know better than the senior cryptographers at Stellar, NXT, Ripple, etc.

How would you feel if mining was further centralized and all mining was done by Amazon in their data centres?

What is your calculus to reach the conclusion ?

I reach my own conclusion by educating myself on a wide range of domain. I don't really care about who said it. If you want to change my mind, just give me a better explanation about the origin of money.

Why would it happen ? Miners suffers directly from Bitcoin crashes, big one takes the biggest hit.
And if it happen, why would it last when it is not profitable to do so ?
As in POS, miners have something to loose to screw up with Bitcoin.

We almost reached the critical 51%. Are you saying it could never happen again?

Look, I love Bitcoin. It's great. I wouldn't be here without it. But it's time to move on. Open your mind and investigate for yourself what's happening.

I personally does not care. Since it can't last. Moreover, whatever power they get, they can't rewrite the past. I am scared that a fork happen for technical reason, but miners are the least of my worries.

Sure educate me, but you did not gave me convincing arguments yet.


What is your calculus ?

Best of luck with your business.

You can bring a horse to water...

My job here is done.

Stellar and ripple aren't even cryptocurrencies, so what are you talking about?

It would not be ideal if mining was all done by Amazon,
but they are not the only big data center company.
Centralizing of mining may be an issue, and something
to keep an eye on, but I don't think it is any kind
of fatal flaw.

Any more questions?

I came because I thought you would show me the water. :(

Oh come on, not this sh!t again... As has been explained to you numerous times by now, the amount of mining performance (and thus, power consumption) is NOT related to the number of transactions.

100 transactions a day require exactly as much mining power as 100 billion transactions a day.

Constructive criticism and feedback is more than welcome, but this is becoming a bit of a sad trolling attempt. Do you personally hate Bitcoin for some particular reason, or what's up with this?

Frankly speaking, this topic doesn't belong to this board.
However, we can estimate Bitcoin mining power consumption in 12 years, provided that Bitcoin replaced all M1 money supply in the world, what is very unrealistic, but we want to make a conservative estimate. As of 2009 M1 was about 20*10¹²USD http://dont-tread-on.me/wp-content/uploads/2011/02/SmallGlobalMoneySupply.png
In 12 years block reward will be 8 times less than now, let's assume that fees still constitute a minor part of block reward, i.e. block reward is 3.125 BTC, that means 18.75 BTC created per hour. Value of 1 BTC equals roughly 1 million USD of 2009 purchasing power (in 2027 USD doesn't exist already, since BTC completely replaced it). Let's now assume, that miners spend half of their revenue to pay for electricity (that's rather unrealistic, but we want to make a conservative estimation), so we have 9.4*10⁶ USD per hour to pay for electricity. If we take a reasonable price of 0.07 USD/kWt*h, that results in 1.3*10¹¹ Watts of total power consumption by all miners in the world. In 2007 the world produced about 20*10¹⁵ Wt*h of electric power, or 2.3*10¹² Watts on the average. We can see that under such unrealistic conditions bitcoin mining consumes less than 6% of total world electric power.
Hope I didn't make a mistake in calculations.

BTW, we can conclude that even in 12 years Bitcoin is unlikely to replace all money in the world, since 6% of total electric power produced in the world is a bit much, I'd say that bitcoin will need not less than 16 years to achieve that  :)

The discussion does not belong but who thought it would deviate here. Finally interesting response though.
Cross checked, your calculus is right.

I'll sleep on it tonight, thinking about if this model is economically relevant though.
A nice test is to plug the current mining data and compare the bitcoin price, to check if the model coherent. I'll plug that on excel, it bugs me. :p
Having the current market cap, and assuming miners pay 50% of their bill in electricity, we should be able to retrace the steps and see if it is coherent.
I'll do that tomorrow I think ;)

I plugged the number corresponding to current BTC price / Mining consumption etc... your model predict 0,008% worldwide of consumption currently.
With mining power of 300,000,000 GHZ and 0,65 W/GHz (I think current mining tech), this give us consumption of 195,000 kw and in kwh/year 1,708,200,000.
Given the power prodruction stat at http://en.wikipedia.org/wiki/List_of_countries_by_electricity_production (http://en.wikipedia.org/wiki/List_of_countries_by_electricity_production)

I reach to 0,0074% of world consumption for mining when your model predict 0,008% so it seems to match very well.
I won't bet I did not make any mistake, but I was impressed to find that it match your model.

stdset, I reposted your response at https://bitcointalk.org/index.php?topic=520977.160 (https://bitcointalk.org/index.php?topic=520977.160), it is a better place for people to discuss it.

You asked how much energy bitcoin requires, not how much energy miners are choosing to use.  The answer is still zero watts. 


These risks have always been tolerated in commerce.  Remember it was bitcoin that first introduced the idea of a public confirmations for a transaction, as it is the first public currency.  People survived with no public confirmations at all, so I wouldn't complain too much that getting one takes 10 minutes.  That being said, if you have an idea for improving the consensus algorithm we would all love to hear it.   

[/quote]



I think you are confusing your centralization issues.  A centralized data centre is different from a centralized currency.  Pools are centralized.  Miners are not.  And by the way, fiat is centralized.  I hope you have stopped using it completely to avoid those risks you mention. 

Well, not exactly, you'd at least have to solve difficulty1 blocks... Still the person who watches this single miner uses more energy per day than the computer to calculate the hashes.

Looks like mining market is more or less saturated now.

lol, nice :) 

I have another comment for those that complain about energy use.

Why do you think the cost of production of money should be 0 watts?  Don't you see the incredible waste this has lead to? 

In the end it matters little how we justify it here or whether we agree or not.  The cost of production of money now is determined on an open market of public currency mining.