Bitcoin Forum

I never followed mining too closely - but the difficulty grows like crazy http://bitcoin.sipa.be/speed-lin.png

It is quite possible that at this rate it suddenly outgrows the profitability level for most miners and if they decide to switch off the mining rigs - then bitcoin transfers become slow. This can undermine the bitcoin system causing the price to drop and lowering the mining profitability even more forcing people to switch off further rigs.

In 2011 the system proved to be quite robust against price drops - many miners did not switch off immediately after the price made mining unprofitable - and the system was maybe slow at some point but not catastrophically so - but this time it might be different with the difficulty explosion.

Mining with already produced hardware only becomes unprofitable if the electricty costs outweigh the income from mining and for ASICs (which make up the vast majority of the network), this is not even remotely the case yet.

What may happen is that the difficulty growth will slow down considerably as buying new hardware become less and less appealling, but as long as the network hashrate doesn't collapse all of the sudden there's no problem (a gradual decrease isn't an issue either).

Difficulty-adjustments will ensure that in the event of miners shutting down the difficulty goes down as well and mining profitability will increase again.

Exactly. Eventually the hardware will catch up with Moore's law and mining will be run by big corporations with access to cheap electricity and large data centres.

Bitcoin's future isn't related to the profitability of miners any more than gold's future is linked the to profitability Goldcorp etc. It is linked to wider adoption and the possibility that something better will come along.

Yes - but the adjustments don't happen immediately - and especially down adjustments will happen slowly.

That's a reasonably good scenario for the Death of Bitcoin. It happened to Namecoin, prices fell faster than the difficulty could adjust making it decades till the difficulty would catch up. They barely limped to the next adjustment to use merged mining.

This is only an issue if a large amount of mining power is switched off at the same time. If the mining power is halved at the start of a new difficulty-cycle, it'll take 4 weeks for a downward adjustment, which is long, but still reasonable. And an instant 50% reduction of mining power is absurd outside of catastrophic incidents.

Like I said before, existing miners won't be turned off anytime soon. And when they are finally turned off, old, less efficient miners, which will have a lower hashrate are turned off first, making any potential difficulty drop rather smooth.

Scrypt alt-coins are far more susceptible, since they're at the mercy of profit-switchers moving large amounts of hashing power from one coin to the other. SHA-256 alts can be merge-mined with BTC if they have been setup for it, so they're more protected.

There is nothing different in the current growth rate from the historic growth rate. Since its inception Bitcoin difficulty has grown exponentially and I expect this continue for some time.

https://bitcointalk.org/index.php?board=81.0

It appears that "sudden" means November-December.

Lots of tears in some of those threads.

My GPUs where still minig at lost when BTC where exchanged at 2$ .. I sold those BTC over 200$..
So, I dont plan to stop mining as I will sell those over 2000 $   ; P

Hint: Avoid selling.

Those miners with electricity included in the rent won't bother, and those miners with high electricity cost will shutdown first and buy coins instead, that will drive up the exchange rate and in turn improve the rest of the miner's profitability

Hmm - but instead of paying for your electricity you could buy BTC and you'd have them more than with mining. Then you could also sell them over 200$ and be better overall. There is no rational case for mining when you pay more for electricity than you'd pay for the BTC bought directly.

Buy maybe your argument is that some people never get this idea that they could buy more BTC then they mine - so they will be mining BTC regardless of the economy of that. That's possible.

People are lazy and it is easier to keep mining than figure out how to buy bitcoins.

Mining is also about protecting the network from attack. If people stop mining because it is no longer profitable, and mining becomes too centralized and suffers a successful attack, everyone's bitcoin will quickly become worthless.

There is no good reason to discourage people from bitcoin mining at this point. We're all still early adopters. There's a lot more to bitcoin than making the maximum return on your investment.

I heard this rumor too.

The inefficient will be replaced by the efficient.  Hashrate will still increase.

It's not like an on/off switch, when mining starts to become unprofitable the difficulty will adjust accordingly until mining becomes marginally profitable again for the average miner. The speed of the network is still increasing exponentially, but this will level off at some point and perhaps even decrease a bit for a while but it's not going to go to 0 or become so slow that it harms the verification process.

Fud theory:

Mining becomes almost unprofitable. Nobody is investing in mining equipment but there is still a small gain to be made.
Somebody with deep enough pockets continues to buy equipment (although at a loss) just to drive and then keep the entire mining unprofitable for most miners at a certain loss even for themselves.
Eventually more miners quit and the mysterious organisation comes close to 51%.

And then bitcoin becomes worthless and they wasted all their money to steal worthless stuff.

The reason it could be far more extreme this time is because of the massive list of preorders of ASICs. This creates a lag in the adding of appropriate hash power and makes it very easy for malinvestments to occur since noone knows exactly when he'll receive his ASIC, and thus, whether it will amortize at all.

With GPUs it was a different situation because your local hardware shop will have a different competency level compared to BFL or Avalon.

What makes this even more interesting is that ASICs have 0 value (apart from materials perhaps) besides generating Bitcoins, whereas GPUs used to be modern GPUs that could be reused and resold for gaming machines.

It's a FUD theory , of course it ends with worthless bitcoins.

That's not entirely correct.  I don't need a gold miner to trade gold but I need a bitcoin miner to trade bitcoins.

Why? You don't need to have bitcoin miner to trade bitcoins.

Dear 'high'commander,

you are high and stupid and have missed the point completely. stfu and go die.

Mining already IS unprofitable for most miners.......

In order for me to send bitcoins I need to broadcast the transaction into the network and the miners need to record it in a block.  No miners = no transactions.

This is not entirely correct. An ASIC can continue to have value as a space heater in winter partially subsidized by the BTC generated even after the marginal cost of electricity is well above the value of the BTC generated. This is because they are far more compact and portable than the typical GPU based mining rig.

"What if mining suddenly becomes unprofitable".

...too late.

People will continue to mine, however. If not to protect the value of their BTC but also for tx fees.

Really? You got the point - be real and prove it, or don't act like a single-celled organism.  ::)

User705,
Alright. Appreciate your response. Because I was aware of that and read about it on wiki a while ago. As I understood the following:
Yes, everything that is broadcasted is being recorded by everyone, just like in any p2p network, but it doesn't say that peers MUST be mining. Is it understood/assumed so? If anyone can confirm that, would be good to know.

as long as bitcoins are valuable the decrease in mining - people going offline should be slow since if there is an opportunity to make money people will turn on their devices

If speculators are smart, they will anticipate the difficulty correction, and try to set the price to reach an exponential apex to coincide with the beginning of the ROI-pocalypse. February/March 2014 if you asked me to guess.

why didn't you read the next phrase

This complete record of transactions is kept in the block chain, which is a sequence of records called blocks.
No mining no blocks no transactions

We either need blocks or all the transactions can be stored in memory at every node.... But then the coins can't be spent again until they are in a block.  So, yeah, we need mining to keep this shit rolling.

However so far the race of ASIC and bitcoin rush has made bunch of wasteful blocks... ouch.

A curious scenario would be where the difficulty has wiped out any ROI in BTC terms, but a wild exchange rate rise makes mining profitable against running costs. I guess that's two definitions of mining profit: ROI above capital outlay for equipment, and then there's profiting on price/energy input against price/minings output. You'd be unwise not to make the most out of the second possibility while it existed.

I did read that phrase. Yes, everything is recorded in the block chain, but again, as I understand how protocol works, miners DO NOT need to know how and where previously mined coins are being sent.  

Guys, think simple. Let's pretend almost all blocks are found and everyone decides to stop mining, nobody will be able to send and receive coins? Nonsense, right? Think again OR find a fact that clearly states that mining has to be in place in order to trade coins.

You're mistaken rewards with blocks
In the end there will be no more bitcoins offered as a reward for mining a block but that doesn't mean blocks won't be mined anymore.

I hope this passage is as clear as it can be:

"Mining, or generating, is the process of adding transaction records to Bitcoin's public ledger of past transactions. This ledger of past transactions is called the block chain as it is a chain of blocks. The block chain serves to confirm transactions to the rest of the network as having taken place. Bitcoin nodes use the block chain to distinguish legitimate Bitcoin transactions from attempts to respend coins that have already been spent elsewhere."

That's not possible.  I think you are thinking about high but stable difficulty but the only reason current roi projections all show negative is because of difficulty rising.  That will not be forever but no one can tell you when it stops.  A good endpoint to calculate is when pure electricity costs are higher then btc generated without even taking capital outlay into account.  At that point it's not unreasonable to assume not many new miners will come online and therefore difficulty will stabilize and then grow only with exchange rate growth.

eh?

That's not impossible at all. If the price goes high enough, it takes the miner that's been pushed out at $140 (due to total monthly returns being worth less than the electricity costs) and puts them back in the profits above running costs territory. They'll still never hit ROI on the ASIC, but they'll at least nibble a little bit more back. The difficulty trend doesn't matter as long as the exchange rate increases enough.

Mining is extremely profitable against running costs right now.  You don't need any higher exchange rate for that.

+1

Bolded for emphasis.

I think someone in the Mining sub calculated that 1st gen 130-110nm ASICs would maintain profitability against running costs up to a difficulty of nearly 2 billion. While we're only hitting 150 million right now, I think we are projected to multiply current hashrate (1 petahash) by many, many times if all these other ASIC manufacturers deliver product. And that's just what's been promised between now and February 2014. Big "if" in some cases, but I suspect they will mostly pull through within a 6 months timeframe, even if they don't deliver to their own (promotion influenced) targets. So things may finally settle around March next year, and it could easily reach above 2 billion difficulty, even if one manufacturer scams or folds.

I think this is true if you only consider the cost of the mining process and not the initial investment the mining equipment.

if difficulty goes down it will be more profitable, im missing the problem?

Since difficulty follows price, and price has corrected from the high, I expect difficulty to peak out too and correct due to unprofitable miners switching off equipment, just like in 2011. I would also think this will happen somewhere the coming months.

Since these are all feedback loops I wonder whether this event will have an influence on the btc price. Anyone knows?

difficulty will continue to go up for a long while still

once we reach the limits for producing the most efficient bitcoin mining machine, then hash rate will fall, but I think we are a long way off from that. idk but i feel like there will be more than 2 generations of ACIS before we reach this technological limitation.

thanks but not my question

I don't think large scale mining farms can operate effectively, since there are lots of miners with free electricity included in rent. That is easy to achieve by millions of individuals but not possible for an organization

That's why we have farmers who take their water by hand from the wells and use digging sticks and no farms with irrigation and harvesting machines.

Also i would love to see how many days you'll mange to stay in a rented house after the owner gets hit with a 1000$ bill.

With first gen ASICs, you would need 200 GH/s to use $1000 in electricity in a month.  That much hardware is not exactly within most hobbyist budgets.

With BFL Monarchs, you would need about 25.7 TH/s or roughly 42.8 cards to use $1000 in electricity in a month.

And a 4800 piece of "hashing" is within most rent a home hobbyist budget?

My theory was as simple as it gets , it will cost less and less money to make an actual attack for the 51%
.
Don't forget that my FUD theory doesn't mean hey we buy 10 millions of equipment and get the necessary hashrate from the first second they plug in.

They just have to buy equipment as other miners do (even receive profit in the first stages ) and invest invest until this is going to make a loss for every miner (including them) and at that point when people quit  >1 hour > 10 blocks and bitcoin is doomed. (which I hope not)

Also...
Ebay buys online payment firm Braintree for about $800m to strengthen its PayPal division's presence on mobile devices

How much will they pay to "weaken" the other payment methods??

And if they try to change the rules, we fork.

What you think the result of a hard fork will be on adoption and exchange rates?

Each of those miners maybe only draw 300w for 60Gh, but the number of those miners can be millions, thus the hash power will be very distributed around the world. Any single mining farm can not afford too much electricity and heat burden, they will be quickly overwhelmed by millions of miners with free electricity

Maybe some large mining farm can design water cooling devices and sell the hot water as a heating service, but due to quick technology advance in ASIC devices, that plan is also risky

Also keep in mind small miners don't price out all the extras it takes to run the equipment.  Large scale facilities have to pay salaries, insurance, etc... etc...

I think then, the miners that are remaining will profit from lower difficulty.

everything that produces heat from electricity will mine, it will be a standard chip / resistor

toaster with mining resistors

hot water heating element

bar heater

hairdryer


think about it.

It could make sense at some point in the development of mining manufacturing, likely when the transaction fees regularly become a significant part of the block reward. That could take another 2 reward halvings though, so until then, it makes more sense to take the route of massive farm + selling complete mining units. And making sure the farm's online before they start shipping, of course

after smoking a spliff (or two) i imagined a home radiator company that builds radiators that are actually miners. once bitcoin price is high enough one could market this as "heating for free"  ;D


edit: lol, misread "what if whining suddenly becomes unprofitable..."

And now?

1)  0.6 W / GHash = 0.6 kW / THash =>  1200 kW / 2000 THash
2)  if you will mine 1 hour then it will cost  1200 kW * $0.15/kWh = $180   and you will mine 6 * 25 BTC =  150 BTC
============
result: running costs are 100 times lower than bitcoin price if you are using 0,28 nm asics  (but hardware is too expensive)

Can you please tell me where does this come from?

1. Power cost (USD/kWh)  $0.15
2. http://cointerra.com/product/terraminer-ii-1ths-networked-miner-january-batch/  28 nm asic  1 THash/600 Watts
EDIT: or http://www.butterflylabs.com/monarch/
EDIT1: https://hashfast.com/shop/baby-jet-third-batch/
3. Network total   2164.711 Thash/s  (now)  http://bitcoin.sipa.be/

The thing is that you can't mine with that now :) you'll be able to mine in January
And unfortunately most of the network hash isn't provided by KNC , it is composed mostly of the last generation mining equipment.

I made the mistake of overlooking he said the running cost and thinking about mining cost (which does  involve the cost of equipment) but still don't think getting into mining was my brightest idea.

We had 19 THash at the begin of the year.  Now we have 2200 THash (so 19 THash is old HW  and 2180 is new one .. asics)

Edit: and at the end of year we will have 10,000 - 20,000 THash ... new 28 nm asics

yeah , but those are running at 6-10W/GH , which is 10 times more at best.
Well it still leaves room against the current electric bill but I doubt that will see any mining equipment bought before August still running in January.

My point was that miners who haven't changed their equipment are having a hard time right now.