I've read in recent days about bitcoin mining and the assertion that the process uses a lot of electricity.
Until I read that, I was under the impression that bitcoins were digital - just 0's and 1's - and that there was no real coin involved.
If anybody on this forum understands bitcoins, could you please explain. Thanks.
the quick and dirty:
Bitcoins are digital 'tokens' that have to be generated via 'math'. It is that math that uses the electricity typically either via a GPU (graphics card) bigger the better or if you have real coin an ASIC.
The math is designed so that it can't be circumvented or someone find a shortcut to. There is a math based upper limit to how many coins can be minted/mined, the process of generating said tokens. these tokens then show up in the 'blockchain' a distributed public record of all the coins, where they come from and where they've gone.
This is a very basic summary in lay terms. If you want the particulars you're gonna have to do the research yourself. and no you can't just copy/paste a coin to make more of them because it won't be a valid and unique token each time you do it.
The math is CLAIMED TO BE designed so that it can't be circumvented or someone find a shortcut to. There is CLAIMED TO BE a math based upper limit to how many coins can be minted/mined, the process of generating said tokens.
You might want to check out these resources:
The mathematics of Bitcoin (Grunspan, et al)
The mathematics of Bitcoin (Crossen)
I'm aware of the idea behind the security of crypto, but still have strong doubts that there cannot and is not any backdoor way of the developer not scamming the system.
It's usually impossible to prove a negative, so I'll grant that it may be possible, though unlikely.
However:
1) The system is open source, as is the mathematics.
2) A very high number of highly qualified cryptographers have looked at the system.
3) There is a very, very high financial value associated with 'scamming the system'. A successful 'scam' of Bitcoin could easily be worth billions.
Given that, the odds of there being a latent scam seem extremely low. Not zero, perhaps, but extremely low.
By contrast, there are far more (and far more obvious) ways of scamming the existing non-crypo commerce system.
None of that says that crypto is 'a good thing' or you should bet your house on it, but as a mathematical transaction system I would be inclined to bet on its security.
The major exception is that quantum computing may, when commercially viable, destroy a number of assumptions on which most deployed mathematical security systems are based. Indeed, governments are starting to both 1) store massive volumes of encrypted communications and 2) seriously look at moving to new algorithms on the assumption that at some point in the not-that-distant future nearly everything encrypted by currently secure algorithms will be trivially decodable at very low cost.
Yeah, there are plenty of scams in cryptocurrencies (including bitcoin) but they generally take the shape of scams that have been regulated out of the normal financial system - pump and dumps of shitcoins, ponzi schemes, etc.
If somebody finds a way to invert a sha256 hash, though, they're gonna be rich and bitcoin will have nothing to do with it.
Reversibility is complicated by several factors:
- Iterative cryptographic structure
- Avalanche effect amplification
- Built-in mathematical complexity
- Loss of entropy between input and output
Changing a single input character in a long string results, on average, of about half the bits of the resulting hash being flipped randomly. Not saying it's not possible, but it's certainly not possible with current state-of-the art computing and mathematics.
If somebody finds a way to invert a sha256 hash, though, they're gonna be rich and bitcoin will have nothing to do with it.
Abolutely. SHA256 is believed to be quantum-safe, though. Algorithmically, it's not all that impossible to find a text which will generate the same hash (EDIT: see below - this is largely wrong), but making that text also function in the way it needs to in order to subvert whatever function the hashed text is providing is far, far more difficult.
And, yes, there are a myriad of well-understood non-cryptographic scams in cryptocurrencies. There have also been a few poorly designed cryptocurrencies which are vulnerable to technological hacks. Bitcoin seems highly unlikely to be, however.
I don't know what you mean when you say that algorithmically, it's likely to find a collision - they certainly exist, says the pigeon hole principle and I don't think anybody argues that one is wrong, but if you can trivially (or even just reliably) generate two inputs that produce the same sha256 hash i think that would be news to me?
You're right. I misread the source document I was looking at. Even with quantum computing, the fastest search is O(2^128). That's a huge improvement over O(2^256), but it doesn't really get you anywhere at all.
Sorry for the confusion.
There is a math based upper limit to how many coins can be minted/mined, the process of generating said tokens.
I thought I read somewhere that the company 'managing' bitcoin in the absence of its creator is actively looking to increase that limit.
AJ
I thought I read somewhere that the company 'managing' bitcoin in the absence of its creator is actively looking to increase that limit.
From my understanding of how bitcoin works, I don't think they can't do that without invalidating all existing bitcoins.
Basically, they would have to release a bitcoin 2.0 with a higher limit, but it would be a new crypto currency starting from zero, not an increase in the limit on the existing bitcoin.
Perhaps I got the wrong end of the stick.
My newspaper reported last week that the rate at which bitcoins are generated is to be halved, as has happened before. I thought mining bitcoin got progressively less productive as the number of bitcoins in circulation approached the limit, but apparently it needs some assistance.
AJ
I thought mining bitcoin got progressively less productive as the number of bitcoins in circulation approached the limit
It takes progressively more computational cycles to "find" each new bitcoin as the limit is approached.
The people who mine bitcoin for a living compensate for this by throwing more and more hardware at it, as well as faster hardware.
Despite their attempts to compensate, the rate of "production" of new bitcoins continues to drop.
I thought mining bitcoin got progressively less productive as the number of bitcoins in circulation approached the limit, but apparently it needs some assistance.
Halving theoretically doubles the work necessary to generate the same number of coins (that is, the work that used to generate 50 coins, now generates 25). That allows the period of coin generation to continue, than if all blocks cost the same to mine.
Consider it similar to a gold mine, where the yield per ton mined decreases as the mine is worked. Eventually, the mine is exhausted, or the work necessary to extract and refine the ore costs more than the ultimate value of the gold bullion created.
I've read in recent days about bitcoin mining and the assertion that the process uses a lot of electricity.
Until I read that, I was under the impression that bitcoins were digital - just 0's and 1's - and that there was no real coin involved.
Not an expert, but my understanding is that Bitcoin is based on a cryptographic cypher using a key of a fixed size. This means that there is a finite number of possible bitcoins.
Bitcoins are "mined" by using a computer to run the appropriate cryptographic algorithms to generate new keys, and thus new bitcoins.
As the pool of "undiscovered" bitcoins diminishes, it takes more and more computer hardware to "mine" bitcoins at a useful rate.
That hardware takes electricity to run, at this point, a lot of electricity.
How much electricity? There was a story a year or two ago about a company mining bitcoins that set up a deal with a local electric utility to co-locate a server farm of dedicated bitcoin mining machines at an electric substation.
Note that it's not just the process of mining a bitcoin which takes electricity; confirming a bitcoin transaction requires the transaction to be recorded in a block (and as a practical matter, it's not really considered confirmed until it's in more than that) and block production requires computation, which requires electricity and time.
As a non-transactional store of value, that's not a huge problem (aside from the energy consumption, since the energy used could be spent doing literally anything else more productive). As a currency, it's more of a problem - it's not super practical to buy groceries with bitcoin if the transaction isn't complete until an hour after you start the checkout process, for example.
The wait for validation is a problem. For what it's worth, the Lightning network and other solutions are designed to overcome that limitation.
As a currency, it's more of a problem - it's not super practical to buy groceries with bitcoin if the transaction isn't complete until an hour after you start the checkout process, for example.
Volatility in exchange rates between bitcoin and fiat currencies (or gold, for that matter) is also a concern.
The value of a payment in dollars or euros is generally stable from transaction to transaction, so that if you take in currency and pay out currency as change, there is little risk.
The value of bitcoin can fluctuate wildly from one transaction to the next. That creates problems in setting prices properly.
Note — this is also true when two currencies circulate side-by-side, though with fiat currency, the exchange rate will fluctuate far less (i.e. has lower volatility) than it will for digital currencies. There are also limits to volatility for gold and fiat currencies imposed by exchanges and regulations, which is not true for bitcoin or other digital currencies.
Just something to be aware of.
Bitcoin is not "currency" - at best it is an investment vehicle.
Four definitions:
Currency (n) — something (such as coins, treasury notes, and banknotes) that is in circulation as a medium of exchange
(n) — a system of money in common use within a specific environment over time, especially for people in a nation state
(n) — something that is used as a medium of exchange; money.
(n) — a generally accepted form of payment
All of these fit bitcoin.
By the anonymous* original inventor intent it is was meant to be a currency.
That it isn't particularly well designed to be one, well, a multitude of technical articles have been written discussing that.
* It was created under a pseudonym, and to my knowledge, despite persistent rumors and a compelling theory or two, there's never been unambiguous public confirmation about who worked under said pseudonym.
"Overwhelming evidence" shows that Australian computer scientist Craig Wright is not bitcoin creator Satoshi Nakamoto, a UK judge declared Thursday.
Comments on the article are worth reading too, perhaps more than the article (with isn't uncommon on ARS)
It's not an investment vehicle either, but given that bitcoin is described as "the first decentralized cryptocurrency" and supporters of it tend to compare it to fiat, which is a currency, it's useful to describe the ways in which it falls short of something that people might actually want to perform transactions with.
Bitcoin is a virtual token with inherent scarcity in the design (there's limited amount that can exist) and is the first original "product" showcase for the blockchain concept.
Digital tokens are awarded by the algorithm in exchange for... something.
Bitcoin is "proof of work" token, to claim a bitcoin you have to prove that you have done a certain amount of complex computing. Exponentially more for each next, and, you have to be first to compute the block with the new bitcoin. (For curious example of alternatives, there was/is a derivative "proof of storage" coin where you would fill multiple hard disks with garbage, insanely huge nonsense files that effectively act as lottery codes for periodically issued "coins" ... not sure how successful that has been.)
By now however, the maintaining of the blockchain transactions is taking a lot of computing (and energy) on its own. Well, it's (mostly) handled by the network of the dedicated "miners" themselves, and... I'm not knowledgeable about the the technical aspects, but have got an impression the two workloads are at least connected nowadays, e.g. I believe (perhaps wrongly) that "merely" servicing the bitcoin blockchain may earn you a bitcoin eventually. It might just be that in public perception it's merged...
The original core "mining" activity is more akin "brute force" chipper breaking. That's where the deeply specialized hardware comes in.
I believe (perhaps wrongly) that "merely" servicing the bitcoin blockchain may earn you a bitcoin eventually. It might just be that in public perception it's merged...
Transaction verification (adding the transaction to the blockchain) is competitive, and miners earn fractions of bitcoins per verificaiton.
At the current time, verification can also be rewarded by additional blocks (bitcoins), but that will end when all 21 million tokens have been issued. At that point, the only compensation will be the aforementioned transaction fees.
Pro tip: Whether it's learning about bitcoin or or to fix your washing machine, there is almost certainly a YouTube tutorial on it.
The new place just built near where I live is a "Makerstar" facility. (Actual company, look it up)
They have 5000 "mining computers" running 24/7 and use about $1M a month in electricity.
They have to locate near an electrical substation to guarantee constant power, and the facility has Seven large transformers on site.
Hope that helps.
Side note:
This is owned by a CCP corporation and lots of people have wondered if bitcoin is the only "data" they might be mining.
Look at a company named "MakerStar"
Out local one has SIX large transformers feeding 5500+ "mining computers".
I know for a FACT that they have to be built near a power substation, and use approx. $1M a month in electricity.
As a side note…
This company is Chinese owned and there is a lot of speculation about exactly what information they may actually be 'mining' for.