Topic: miners shutting down? (Read 2819 times)

If the difficulty increase like this, soon some giant miner would hit break-even for their electricity...

I doubt that difficulty is going to drop, at least not for a very long time. We're currently seeing the beginning of the end of the mining-manufacturer bubble; the companies sit on thousands of chips that they can't sell anymore. I expect a big wave of dying ASIC companies, but till then, the only choice of these companies is to turn to self-mining with their own unsold equipment. ASICminer is an example of that, but I think that the recent rise of DiscusFish's percentage in global mining is all the cheap ASIC manufacturers in China starting to self mine or cooperate with big industrial mines in order to get rid of their pre-produced equipment.

difficulty won't drop as long as I continue to double.. nay, treble, my moneys in mere months via Extremely Lucrative Cloud Mining Opportunities

I posted this chart over a year ago:



I even got the exchange rate right.

Keep in mind that for $/TH you have to take in to account variable production cost for the OEM's, not the consumer list prices. And you have to calculate with their energy costs, not ours.

No way!  What'chu talkin' about Willis?  KNC Miners is offering 400ghash for **$250** for a 6 month contract!   You get the privilege of renting their 'super cooled arctic' miners in return for paying them $250 up front and getting $200 of that back in 6 months.  Sounds like a steal.

Once difficulty hit 500ph we should see some stability.

Current hash rate is 245 PH

Peak illogical hash rate was reached a month or so ago.

Peak stupidy will be reached between 400 and 600 PH or when Bitcoin goes to $0.01c

Other than that, best not to employ logic when it comes to understanding the thinking behind adding 150 to 200 more PH to the network while the demand for bitcoin is so that the price is in freefall.

They will probably quit all together. These mining companies are making faster ASICs that we can't afford & selling them to huge farms. So we literally funded our own death! The death of home mining! Now we will have to pool our money together just to get some coins!!!

if the price keeps on going down some Big farms and home miners would quit for awhile

Those that have high electricity is shutting down their miners and buy Bitcoin with it.. So we should see some rise in price

I suspect these are just two different aspects of the same issue.

One way it to look at what is the lowest economic bitcoin value for mining with the CURRENT difficult.  This shows that a price one third of the current bitcoin value would make some large scale mining farms uneconomic ($150).

The other way is to look at difficult level changes to find the point where large scale mining will stop growing, based on the current prices, as Nagle has done.  This shows that at current prices a three times increase in difficulty to about 100,000,000,000 to 200,000,000,000 (say 10% for 8 to 10 difficulty changes) would make mining uneconomic.

In summary large scale mining is currently quite profitable.  But a three times increase in difficulty or a corresponding fall in bitcoin value price, would make large scale bitcoin mining uneconomic with current hardware and bitcoin values.


This thread asks "How long till the difficulty starts to drop since the price is not keeping up with difficulty?"  

Using the above estimates of mining profitability it looks as if large scale mining will not stop rapidly growing, and difficulty will not stop rising till early/mid next year.  At this point the cost of mining on a large farm and the value of the mined bitcoin will basically be in balance. 

Difficulty changes are then likely to follow the value of bitcoin even if with a lag.  It is the same as the amount of gold mining around the world changing as the value of gold changes.  If it is profitable to mine lower grade gold you can be sure miners will appear.  If it becomes unprofitable for certain gold mines they will eventually close.

As the regular 2016 block difficulty adjustment has a 2% or so statistical variance we should see in 2015 the occasional negative difficulty change, even if the total mining hash rate is still slowly rising.

But big drops in difficulty are not likely until mid 2016 when the bitcoin block reward changes from 25 to 12.5 bitcoins.  The hashing difficulty should then half very quickly. (This assumes the halving does not also affect the bitcoin price.)

Those numbers need further examination. A large-scale miner has costs for construction, floor space and staffing. And the difficulty reduction has to be calculated in.

Play with this calculator. That's set for the assumptions the poster gave.  Bitcoin price is assumed constant at 400, and difficulty is assumed to increase at 10%/adjustment. Break-even is in November 2014. At end of life, profitability is 70%.

Now try a price of $300. Breakeven is at the end of January 2015. But at end of life, profitability is only 6%.

At $290, end of life profitability is only 0.07%. Below that, it's never.

Not seeing any way to get profitability at $150 given realistic assumptions.

Even if it got to the point that they had to shut down, they could just as easily power back up once it is profitable again.  Plus I have to imagine that there are farms running on renewable energy like solar or wind that at this point have very low overhead.  Those facilities would probably remain profitable for quite some time.

Large scale mining farms would appear to be profitable with current mining hardware and current difficulty down to about $150 per bitcoin.

This assumes the miners operating in northern areas with air cooling, can get electricity at about $0.05KWh and can make their miners for a marginal cost of about $1,000 to 2,000 per 4 terahash miner.

So don't expect any large scale miners to turn their miners off at this stage.  They will still be adding capacity.

What is more likely in the near future is a reduction in the number of companies making the ASICs.  Only the most power efficient chip manufactures will survive.  Mergers or takeovers are very likely, especially as the move to below 20nm chips starts.  

By the middle of next year I would expect mining to be dominated by 10 to 40 large scale mining farms, each with several megawatts of power.  The number of large scale farms will depend on the bitcoin value.  Large scale farming total earnings rise and fall by about $100,000 per month for each $1 change in bitcoin value.  So at $500 per bitcoin, large scale farms earn about $50 million per month in total (roughly).  

In summary,  in the long term bitcoin value will drive difficulty and also the number of large bitcoin mining farms.  But with Bitcoin still in the $300 to $500 region we are a considerable distance from seeing the least efficient large scale farms pulling out of the game.   Expect the difficulty to keep rising.
 

The key take away for me is investment in the next generation of ASIC chips is a becoming a very iffy proposition. The next halving will occur in their "productive" lifetime.

I still have usb miners running wont shut those down anytime soon.  Though I have a s3 on the way I cant wait for.  Shutting down my gpu rigs and starting to part them out, alts are getting boring.

Lol at large mining operations leaving before home miners!

Large mining operations aren't playing on the same field as most home miners. Their equipment costs pennies on the dollar compared to retail mining equipment. And they pay a fraction of what home miners pay for electricity.

They'll be running years after the last home miner shuts down.

The marginal cost of mining is very low. They'll keep on mining until mined BTCs are worth less than electricity consumed. Now it is only Europeans with expensive electricity and old ASICs who are at this point.

The problem is over-investment in new capacity. I don't think anyone, not even producers, will make back their investments in hardware and data-centers now. The reason they still put up capacity may be backlogs or overpriced cloud hashing contracts.

The most inefficient hardware is leaving the network all the time when the difficulty goes up. cpu->fpga->1st gen. asics->etc. You can't see that from the hashrate graph, because new mining power offsets this.

Big miners may close if the price was to rise too high I suppose which is probably to be expected however smaller miners would continue no matter what. Seeing the bigger miners close would only encourage smaller miners.