Bitcoin Forum

I would like to create awareness of an economic effect that will make BTC mining in general an illigal activity. In short: Mining will become too expensive for full-cost paying 'legal' miners (who pay for their power consumption). But this is not going to stop parasite-miners who will crowd them out...

While thinking about rational behavior of BTC users and miners I thought about the fair price of a BTC and how it is tied to the cost of mining a BTC. I realized that the current situation in which it is possible for all enthusiasts to run (seemingly) profitable mining rigs is a short term market abnormality which is inevitably doomed to disappear soon - completely independent of the BTC price level at the exchanges. I predict the price for mining per BTC (i.e. the costs for electricity (kW/h) that has to be afforded) will in the long-term be higher than the exchange market price that can be earned for selling it - thus for 'normal' people mining will become a money-losing endeuvor. Mining will inevitably become a parasitic (mostly illegal) activity.

The reason for this is quite simple: Whenever the market price for one BTC is higher than the mining cost it will motivate people to participate in mining thereby making the mining task more difficult for all miners. At first I thought this would allow profitable mining only up to the point where the running costs for mining stay below the market price. But then I realized that today there are two completely different groups involved in mining: the full-cost paying 'legal' miners and on the other hand the 'illegal' parasitic miners who use secretly computers in public organizations (universities, youth centers etc.) or privat companies. The big difference between those two groups is that the latter does not have to pay for their power consumption what makes mining attractive for them even when the actual costs for the host organisation owning the machines would be prohibitive.

In the BTC mining business the costs for electric power will dominate everything. At first rational miners in regions with the highest energy prices will be squeezed out by miners in regions with lower costs. But finally as the parasite group keeps growing it will eventually make legal mining too expensive .. even for the lowest non-zero energy cost payers.

In the end all mining activity could be performed by illegal/parasitic miners. While the price for one BTC will be higher the price for mining a new one. It is a simple crowding-out effect that places in if the number of parasitic computing power keeps increasing - the parasites are the only ones who keep making profits even when their hosting organization has much higher than break-even expenses.

I wonder whether this has been discussed before and if it could become one of the major points of critique in the public debate (beside the money laundering thing of course). One of the good things about the parasitic BTC mining is that it may act as a support for the BTC value - because everybody then knows that a substantial amount of money had been stolen/sucked out of some business entity for mining the BTCs.

(Due to the high number of posts I am not 100% sure whether this topic has been addressed before - if it has, please let me know the thread uri.)

/blueling

a troll just escaped the newbies section.

The "illegal" mining you mention isn't terribly profitable.  Most university, lab, school, and other large computer "labs" don't have the horsepower and certainly don't have high end graphics performance necessary to mine effectively.

In order to perform "illegal" mining, you're going to need Bot Nets.  While CPU isn't good for mining, 100,000 CPUs will mine enough to make it "profitable" for someone to rent bot nets and feed them mining code.

The downside is that the users are probably going to notice their CPUs running at 100%.

This is the proposal that Symantec has already published:
http://www.symantec.com/connect/blogs/bitcoin-botnet-mining

:-)

set the newbie barrier to 100 posts and 500 hours  :-*

Does the cost of gold reflect the amount of money it takes to produce one troy ounce?

Or could it be that it's scarce and has properties that have increased it's value over the centuries?

I have a great business plan for you: Go mine gold at $400 per Kg and sell it at $13,000. Why buy a gold watch or a gold bar when you can go straight to the source.

I think OP has a good point actually.

The numbers are so enormous that using 100,000+ cpu's would have massive impact on difficulty, forcing mining costs up. Then there is the small percentage of zombie machines that will have a descent graphics card or two that can be put to use. Mining clients are well suited to the botnet paradigm and can be easily integrated.

I guess the question will be whether botnet owners want to do something quite so brazen.

 

Some people have a low threshold of "troll." I thought the OP's post was quite true. The cost of mining bitcoins will surely go below the cost of the electricity required to mine them, because many people can find ways around paying for electricity. 

Long term, however, the equipment required to mine will get more and more advanced. It is unlikely that normal company or university computer systems would have the hardware to make mining worthwhile even if the electricity is free - even controlling 100 university computers right now is pretty worthless unless they've got gaming GPUs.

+1

yeah, i guess the best way would be to have access to solar- or wind-energy...

I think that the OP has a point, but not in the near term. I'm mostly a miner, and my gear should be payed off before the next difficulty increase. I'll keep mining with all my systems (each family member has a dual GPU setup now) until the cost of electricity is greater than the value of the BTC mined. I'm expecting this to be several months off, but will be quite happy if the value goes up to cover difficulty increases.

One thing the OP hasn't taken into account is the ASIC and FPGA mining that can occur at very low electrical use rates. I think before the "legal" miners get squeezed out by the "illegal" miners, that the FPGA and ASIC miners will take over, so all that is left are them and the "illegal" guys.

Lets not forget about the many kids that are using their parents electricity, or college students in a dorm with "free" electricity. These two do't fall under "illegal" in my book, but they also don't have to directly pay for electricity.

solar / wind have H-U-G-E setup costs to them. In fact solar energy is considered more expensive than traditional forms of energy production for this reason. With government incentives home panels are priced to pay for themselves over the course of ~20 years. If bitcoin mining profitability went to parity with traditional energy costs, that would represent a loss for a solar panel user.

As for illegal activity, I don't really see that being the real takeover step. A cpu miner can get something like 1/100th the hash rate of a similarly priced gpu. So a 100,000 person parasite botnet is the equivalent to about 1000 gpus. It is estimated that there are about 30,000 GPUs mining currently, that would require 3,000,000 compromised computers to reach parity while cpu mining. I'm guessing that 24/7 100% CPU load would attract even the noobiest trojan infected noob to wonder what is up, thus risking the botnet owner lotso f ocmpromised computers. These computers themselves have value (for illegal activities, such as ddos or what have you) and losing them because of mining might be a net loss. Hence I don't really see much incentive for people to take over bitcoin with zombie comps. More profitable to infect people and steal their wallets or ddos rival pools for pay or something.

I think you make a good point. Not being a miner though, I couldn't care less. As long as the network is kept online what does it matter if it's run by botnets?

I don't think this will hold in a post ASIC era. No one, unknowingly, will allow installation of ASICs on their computers and everyone will notice very slow PCs.

Don't you suppose the IT departments of countless companies are about to find out what's happening on port #8332 of their machines?
I would think that large companies, followed by mid-size and even small companies, will start blocking access to "api.bitcoin.cz", "deepbit.net" just like they block access to "facebook.com" and "myspace.com" today.

Universities -- same thing. They will start dealing with the leeching of electricity, etc. and of course the students are paying a ton anyhow -- I'm sure electricity is covered. But using more than 1 PC's worth will draw attention, and might be clamped down on.

As for the kids in mom's basement -- the moms and dads of the world are about to get a wake-up call when the 1st "Post-bitcoin-mining" electricity bill comes due. I know, bitcoin has been around longer than that, but a LOT of people discovered it in late May.

Matthew

Yet they exist. There are more and more simulation labs running models on GPU-s; there are  weather, aerodynamics, rock movement models. I know a lab with 64 5750s. If one gets his miner disguised as one of the models that are being run there, he can get easy access to 15%-20% of the power without real risk of being caught. That is around 3GH for free.

I manage about £100,000 worth of IT equipment at work and did consider mining from there, unfortunately its pretty terrible for mining as it will all be CPU mining. I tried it a little as I was curious and I got about 10mhash/s on one of the systems.
Considering I have 4 of these in total I would get 40mhash for 100k worth of kit, it is sooo not worth risking it.
Now if I knew about this last year then I may have added the price of several ATI cards, but hindsight is a wonderful thing.
I am sure there would potentially be malware out there that would mine secretly on peoples machines if they got infected. It would be hugely noticeable though unless the malware limited to CPU mining and only to 1 core, then I guess you could keep it hidden.

My plan is to set up mining computers as servers for companies for a discounted rate (yes they will pay me to sit the server there) I'll maintain them for this discounted rate included, and since the processor is sitting there doing nothing, the "server" will happily be serving files and what not.  Eventually, they will have bought the server for me and I'll just make the bitcoins by throwing either two or four powerful gpu's in there.  This will be legal as they will be signing a contract for my services and I'll be getting paid to mine rather then paying electricity to mine.  :) Get fucked everyone else?

OP is not a troll, just wrong. Trolls seem to abound in 500+ posts category.

GPU miners have 100:1 gain over CPU, and ASICs have 100:1 over GPU. You need 10.000 CPUs to catch up to a single $50 ASIC, so it follows energy prices are not relevant, it's mostly best tech wins. Building a 100.000 or 1000.000 computer botnet is not free, and has a black market value higher than the equivalent ASIC mining. I predict ASICs will arrive in ~ 3 months if the price stays stable.

That is a "risky" prediction given that some company has already announced they will have them in September-October. Although in this kind of projects, delays are the norm not the exception.

Please, stop writing random numbers.

I love that I'm simultaneously accused of making predictions that are both random and obvious :)
Yes the LargeCoin thing was announced, no they don't have anything to show for other than a Google spreadsheet. Not even a website. There's some controversy if the company that produces the first ASIC would sell it, it seems so damn profitable to seize the opportunity and become the top miner. Regardless if LCs become available for sale, I expect a huge difficulty rise in about 3 months - but only if the price stays over $10.

The part about 100:1 over a GPU is based on my technical insights and my day job, I've detailed the numbers in another thread (http://forum.bitcoin.org/index.php?topic=17804.40):


I'm obviously estimating the orders of magnitude, not exact numbers.

int returnRandomNumber() {
return 4 //created by rolling a dice, totally random
}

How wrote that doesn't have a clue. There are too many "experts" in this forum tossing out numbers and previsions. There are several problems with the implementation just described and only a newbie will even think about it.

What about if Bitcoins for some reason needs to change encryption algorithm?

What is the relevance of changing the algorithm (effectively, an alternate blockchain) in the context of ASICs for mining the current blockchain ? Are you implying you can anticipate such a change and future-proof your device ? You do know what "ASIC" means do you ?

I read about the LargeCoin ASICs rumors yesterday for the first time. Before that I only heard about experiments with FGPA chips. Indeed if it worked - ASICS would become the primary (most cost efficient and thereby most effective) way of mining - mostly because they are more energy efficient compared to GPUs in Watts per GigaHash. Therefore I would not agree that 'energy prices are irrelevant'. Rule of thumb should be: More efficient machines will crowd out less efficient ones. As soon as a miner figures out that running costs are higher than the market price for him he has to stop mining immediately - even when he believes that the price for bitcoins might rise dramatically - buying at an exchange from somebody who is able to mine at lower costs would be more effective. If holding/investing in BTCs is not an option for such a sub-threshold-miner he has to invest in more efficient hardware or simply will go away...

Very efficient specialized hardware (compared off the shefs products available to the masses) in the hands of a single entity could have a significant impact on the bitcoin ecosystem. If nobody can catch up this entity could reinvest all generated coins into hardware until it would possess >50% of the network's computational power. This would then allow to dictate transaction fees by not accepting blochs of other miners because eventually the longest blockchain would be generated by machines of that entity itself.

If there will still be competition between a high enough number (>oligopoly) energyefficiency will be their primary area of optimization.

Regarding my original post I agree that it might not be possible to run a ASICs super-computer with significant power consumptions hidden in the corner of a youth center. Parasitic mining would make senes only with the latest generation of minig hardware (a factor of up to 1/10 might be ok - but if others have 100 times more effective machines it will not be worth making a profit of 1% of the energy costs paid by your boss).

As far as I know the price for mining a single troy ounce currently is >500 USD. If it would be possible for everybody to dig a hole in their garden to mine an ounce with half an hour work it would be worth near nothing. It is pretty sure that you as an individual cannot effectively mine gold for less money then it is sold.

Could you please provide a reference for the 400 USD per Kg statement?

Oh, and the "average person" can't mine Bitcoins either.  They do not have the technical knowledge to get a mining rig up and running.

Wht does it matter if the network is run by Botnets, well, it means all the remaining bit coins will be held by very few people, then we will be back in central bank territory.
The original poster is correct, when mining begins costing serious energy, parasitic mining will be viewed as stealing energy, then how will people mine.
Either by parasitic illegal mining, botnets etc, or by collaboration, I don't think many individuals will be able or even willing to mine.
The parasitic miners will end up owning significant shares of BTCs, yet the original intent of Bitcoin was that BTCs should be widely distributed, not into the hands of a few.
The webpage mining facilities are a way that people can collaboratively agree to mine legally, and also ensure BTCs don't end up in the hands of a few.

 

Hey, it's the economics subforum.

People pull stuff out of their ass.

There seems to be the misconception that faster hardware could generate more coins. But this is only true in so far as you might be able to get a bigger share of the fixed daily production of 50 * 6 * 24 coins that are generated on average. Lets say I would have a super-computer which could generate 20% of the current bitcoin network procossing power in Hashes per second.. but the cost for operating it would be much higher than what the expected amount of solved blocks/bitcoins would cost at an exchange. Then I would not power up my super node but start buying BTCs from a cheaper producer.

Currently the electricity costs for mining a BTC are lower than the price payed at exchanges like Mt. Gox. But already today the profit margin is different for systems with different efficiency rates and also depending on the price for electricify which a 'legal' miner has to pay. We will very likely see in the next couple of month if not faster, that the profit margin for current GPU hardware will go towards zero. And the important point here is that it is not possible to buy a faster GPU that is less energy-efficient to increase the profit margin - if you cannot improve the efficiency of your system or find a 'provider' electric power who is cheaper .. you would loose money continuing to mine with your inefficient (sub-threshold) system.

Therefore: Speed alone (hashrate) does not help you. A 'legal' miner can only make profit (and e.g. generate the money for his/her rig) when he is using hardware that has a kW/h per GHash efficiency above a certain threshold.

Viewed differently: Imagine the daily 'production' of ~7200 coins would be worth 100.000 USD at Mt. Gox and that price would stay constant. At the same time the number of miners would increase until a saturation plateau is reached. If all miners would daily convert their coins (not acting as traders/speculator) an ever increasing fraction of the 100.000 USD would go to the suppliers of electric energy...

And as a student if you got caught misappropriating those resources for something like mining, you'd be expelled (most likely).  Happened to several people using computer resources for their own purposes, Bitcoin mining would be even worse, because you're not folding proteins, you're making money for yourself.

USC (and others) have specific policies against it:
http://web-app.usc.edu/scampus/1100-behavior-violating-university-standards-and-appropriate-sanctions/

"11.35
Theft (or attempted theft) of property or of services within the university community; possession of stolen property regardless of origin; or misappropriation of university resources."

I'm not mining, I'm just testing the vulnerability of the SHA algorithm, it's for a paper i'm writing.

And if you are a scientist who run the models in that lab as your daily routine? Or if you are the lab supervisor?   :P