The Best Mining Rigs Are Now Barely Profitable -- Now What?

[Puppet]

Even for them it will start being unprofitable past 15 billion difficulty.

ROFL. You really think those asics are made from pure gold dont you.

[samsonn25]

Now what?....I will tell you what.

The smart residential miners are closing up shop this summer because before the end of the year no one in a house will be able to make money.

The smart commercial miners are already planning their exit strategy for the end of 2015.

The smart industrial miners are making connections with governments and banks for when mining becomes 100% centralized.

Mining is pretty well over folks...most people just don't realize it yet.

I have to disagree with everything you said.

PUE of home miners = 1.03

PUE of industrial miners = 1.3-1.5

How can you speak for "smart miners" without being one?

I agree there are retail miners. Commercial. And manufacturers.  I know of 1 big maker selling the farms now to exit by xmas.

[a.miner]

There is one variable missing from your equation that is very important.

The cost of the mining equipment. With this variable added it seems we are already at the point of the no sum game, at least for the retail and commercial miners. Manufacturers might have another 30-45 days because they can mine at the manufactured cost of the machines.  Even for them it will start being unprofitable past 15 billion difficulty.

If I understand Puppet's graph correctly, each curve assumes a different cost per terrahash of mining rigs.

I do agree that even the players who make their own rigs based on their own ASICs are going to feel the crunch soon.

And that's why I ask... what now?

Something has to break.  But what?  And in what direction?

[samsonn25]

Sit back and watch the fireworks 4th of july.

[pjviitas]

There is one variable missing from your equation that is very important.

The cost of the mining equipment. With this variable added it seems we are already at the point of the no sum game, at least for the retail and commercial miners. Manufacturers might have another 30-45 days because they can mine at the manufactured cost of the machines.  Even for them it will start being unprofitable past 15 billion difficulty.

If I understand Puppet's graph correctly, each curve assumes a different cost per terrahash of mining rigs.

I do agree that even the players who make their own rigs based on their own ASICs are going to feel the crunch soon.

And that's why I ask... what now?

Something has to break.  But what?  And in what direction?

I am this thinking that it will be like any other example of centralization.

Bad for most...good for a few and generally devastating for the economy involved.

ASICs will not be usable by anyone without a commercial or industrial service which will drive the price down.

This will in turn decimate consumer demand which is actually a massive segment of the bit coin economy.

[samsonn25]

There is one variable missing from your equation that is very important.

The cost of the mining equipment. With this variable added it seems we are already at the point of the no sum game, at least for the retail and commercial miners. Manufacturers might have another 30-45 days because they can mine at the manufactured cost of the machines.  Even for them it will start being unprofitable past 15 billion difficulty.

If I understand Puppet's graph correctly, each curve assumes a different cost per terrahash of mining rigs.

I do agree that even the players who make their own rigs based on their own ASICs are going to feel the crunch soon.

And that's why I ask... what now?

Something has to break.  But what?  And in what direction?

I am this thinking that it will be like any other example of centralization.

Bad for most...good for a few and generally devastating for the economy involved.

ASICs will not be usable by anyone without a commercial or industrial service which will drive the price down.

This will in turn decimate consumer demand which is actually a massive segment of the bit coin economy.

And alot of asic companies will stop making machines because of low demand, and negative return on investment as electricity will cost more than the btc it mines.

[a.miner]

I am this thinking that it will be like any other example of centralization.

Bad for most...good for a few and generally devastating for the economy involved.

ASICs will not be usable by anyone without a commercial or industrial service which will drive the price down.

This will in turn decimate consumer demand which is actually a massive segment of the bit coin economy.

I couldn’t agree more about the evils of centralization.

But we have already determined that a large number of small miners can operate more cost-effectively than a small number of large miners.  So hitting the wall of difficulty should hurt the large miners harder than the small ones.

This should actually lead to decreased centralization.

Unless I am missing something?

And a lot of asic companies will stop making machines because of low demand, and negative return on investment as electricity will cost more than the btc it mines.

Yes.  And this, combined with inefficient miners going out of business, should drive the difficulty back down to the point where it makes sense for the most efficient miners to continue or resume operating...

And then there is the wild card introduced by the price of BTC, which the large players have more control over than the small ones.

So can we expect some kind of "BTC business cycle"?

How is this likely to play out?

[pjviitas]

I am this thinking that it will be like any other example of centralization.

Bad for most...good for a few and generally devastating for the economy involved.

ASICs will not be usable by anyone without a commercial or industrial service which will drive the price down.

This will in turn decimate consumer demand which is actually a massive segment of the bit coin economy.

I couldn’t agree more about the evils of centralization.

But we have already determined that a large number of small miners can operate more cost-effectively than a small number of large miners.  So hitting the wall of difficulty should hurt the large miners harder than the small ones.

This should actually lead to decreased centralization.

Unless I am missing something?

And a lot of asic companies will stop making machines because of low demand, and negative return on investment as electricity will cost more than the btc it mines.

Yes.  And this, combined with inefficient miners going out of business, should drive the difficulty back down to the point where it makes sense for the most efficient miners to continue or resume operating...

And then there is the wild card introduced by the price of BTC, which the large players have more control over than the small ones.

So can we expect some kind of "BTC business cycle"?

How is this likely to play out?

I work for a power utility and have a pretty good idea about what types of electrical services are available and what it takes to upgrade an electrical service.

With this in mind, I don't understand how miners are going to be able to keep plugging in increasingly more miners without paying for costly upgrades to their electrical service.

Just by running some simple difficulty growth projections I predict that miners will cease to be able to make a profit off a typical 200A residential electrical service.

Basically, the infrastructure is not in place for distributed mining.

On another note, Bitcoin mining is not the only thing challenging power commissions...grow ops and electric cars are 2 others that come to mind

[rocks]

Create an excel spread sheet that calculates the break even hash rate for a given BTC price, ASIC efficiency and cost of electricity.

Adjust each assumption up and down, you'll see the break even hash rate adjusts as described above.

I did the excel sheet ages ago and the results are obviously not quite that simple, unless you assume free hardware (or an eternal break even time expectation).

Nice chart

You are assuming that miners as a whole are rational people by pricing in a 1-year break even requirement.

For my calculations I took the assumption that miners are irrational people and will continue to buy equipment up until the time the utility company's monthly bill is larger than the BTC generated. This happens when electricity cost equals revenue generated. At this point miners start to shutdown, not because they want to but because they have to since they are going broke.

If you look at the GPU era, this is exactly what happened in 2011. The hash rate overshot break even, and then started a downward slide before leveling out. At the rate people are still buying equipment today we are well on our way to doing this yet again.

Only this time there won't be a very large existing second hand GPU market for gaming to take the supply of equipment that will pop up on ebay. That's when we hit fire sale time this time around.

[majakn]

we need new good SHA-2 coin
 

[davejh]

Thanks for double checking, your calculation is right. I was going off of memory and switched the break even hash rate with break even difficulty in my head.

Just checked and my assumptions last year were $200 per BTC, $0.10 per KwH and 0.8W per GH/s. This worked out to a break even hash rate of 375 PH/s at a difficulty of 52 billion. (I remembered this as 50 PH/s). If you take your assumptions and double the price to $400/BTC and use 0.7W efficiency I got a break even hash rate of 857 PH/s, so yes that's right.

What's interesting about this calculation is you can see how the difficulty should rise or fall once the ASIC market settles down.
1) Global hash rate should go up and down with BTC price. For example if you double the price of BTC the hash rate should double.
2) Global hash rate should go up and down opposite of the price of electricity. For example if the cost of electricity doubles, global hash rate should drop by half.
3) Global hash rate should go up and down with ASIC efficiency. For example if ASIC designs improve by 10%, global hash rate should go up by 10%.

Regardless of the exact number and break even date though, the basic point still holds. Once the hash rate rise to where the most efficient ASICs are break even with electricity, "datacenter" style operations are not profitable and lose cash flow every single month. At that point the only thing for them to do is sell off their hardware and exit. Who knows how it will happen, they might get decent market prices from home users or it might be a fire sale, but my guess is mining will become more decentralized than today.

Even 875 PH/s is not that far away, probably 6 to 12 months. Here is the log chart: http://bitcoin.sipa.be/speed-ever.png. Each of those bars on the vertical axis represents a 100x increase in hash rate. We are at 50 today which is only 14x from break even using your numbers....

I'd been thinking about this a few weeks ago and finally got time to build a simulation of the hashing last night (it's a couple of hundred lines of C). Here's the extrapolation:



The hashing rate here is in PH/s. The 3 predictions are for the price of BTC in 2 years time.

The full write up is at: http://hashingit.com/13-megawatts-of-mining

[tuneman1980]

Please share how you are able to see in to the future.

Magic crystal ball?

Well, I'm just looking at historical trends in difficulty and trending them out.

Difficulty is now at almost 7 billion and increasing at 11 percent every ten days.  That puts it at 45 billion a year from now.

Unless the difficulty stops increasing, Terraminers and Neptunes won't be able to pay for their own electricity in another 12 months' time, let alone recoup the purchase price.

Even these new SP-30s are only about 3 times the perfomance per dollar, so they won't make it past 16 months from now.

And those estimates are very generous.

So... are people expecting difficulty to level off, or much more efficient technologies to emerge...

or what?

Difficulty will have to plateau at some point unless everyone is just mining based on future price.  Here's my historical analysis http://www.cryptocoinstats.com/difficultytracker.php

[a.miner]

I'd been thinking about this a few weeks ago and finally got time to build a simulation of the hashing last night (it's a couple of hundred lines of C). Here's the extrapolation:

http://hashingit.com/images/articles/20140428/hash-predict.png

The hashing rate here is in PH/s. The 3 predictions are for the price of BTC in 2 years time.

The full write up is at: http://hashingit.com/13-megawatts-of-mining

Well done, davejh!

This is a pretty close match to what I'm expecting, except that I agree with rocks, in that we should expect an overshoot and then a correction in hash rate.  Don’t know how to model that realistically, though...

[davejh]

Well done, davejh!

This is a pretty close match to what I'm expecting, except that I agree with rocks, in that we should expect an overshoot and then a correction in hash rate.  Don’t know how to model that realistically, though...

I'm not sure if we'll see an overshoot as such but I think we will see people turning off older mining hardware sooner than they'd expected as the operating costs become larger than what they generate. Newer hardware is much more capable though so we'll probably not see any actual drop. A new ASIC might be 10x higher performance (or more) for the same input power than the ones being dropped off though so we may not really notice.

As an example, in the last 2 months the new capacity is almost the same as all of the previously-existing capacity. It makes no sense to switch any of the mining hardware if it's generating more than it costs to run. The hardware may never ROI but the only sane thing to do is ignore the sunk cost, unless a buyer can be found. It makes no sense for that buyer to switch it off though as long as it generates more than it costs to run.

The single-purpose nature of the mining ASICs really does suggest that things will have to be different to when people decided to switch off GPU mining hardware that could be "repurposed" to rendering graphics :-)

[Puppet]

I'd been thinking about this a few weeks ago and finally got time to build a simulation of the hashing last night (it's a couple of hundred lines of C). Here's the extrapolation:

 ...

The hashing rate here is in PH/s. The 3 predictions are for the price of BTC in 2 years time.

The full write up is at: http://hashingit.com/13-megawatts-of-mining

I like the chart, but I cant agree with your assumptions. IN particular electricity costs. Average electricity costs are irrelevant, what matters is lowest electricity cost (where mining is feasible). Because that is where mining will relocate, and places where electricity price is substantially higher, miners will have to shut down. Here is a read for you:

http://www.spokesman.com/stories/2014/apr/26/northwests-cheap-power-drawing-bitcoin-miners/

A map comparing energy rates led them to Central Washington, where hydroelectric dams churn out electricity that costs industrial customers less than 2 cents per kilowatt.

You are assuming 10x that price.

Secondly you are incorrectly factoring in asic "improvments". THere are two factors that matter, price per GH and watt per GH. The first one doesnt need Moore's law to improve dramatically from where we are today. ALl thats needed is difficulty going up further, and prices will come down proportionally. We are no where near asic manufacturing costs yet, in fact today marginal production cost of these chips is so low, that its irrelevant, they might as well cost nothing. Asic alone, I estimated HF golden nonce elsewhere at $0.04/GH in wafer production and packaging cost (assuming they can hit 800GH with their rev).

As for watt/GH; also there we do not need Moore's law to get a doubling in 2 years. THe very same asics being sold today almost certainly have the possibility to be downclocked and downvolted to achieve that doubling (or more), be it at the expense of performance per chip. ALmost all miners sold today are being run and sold at the top end of the voltage/clock shmoo plot, which is evidenced by the fact that they typically barely overclock at all. To see what changing of voltage and clock can do, look at Bitmain S2, which is the exact same chip as in the S1, but is twice as power efficient. Im willing to bet KnC, HF, CT, etc can easily double their power efficiency too on their existing designs, once that trade off becomes worthwhile. But since electricity cost is only a tiny fraction of most miners cost, and pricing of hardware is far more dependant on hashrate than power efficiency, it doesnt make sense for 28nm designs today. Doesnt mean they cant.

[davejh]

I'd been thinking about this a few weeks ago and finally got time to build a simulation of the hashing last night (it's a couple of hundred lines of C). Here's the extrapolation:

 ...

The hashing rate here is in PH/s. The 3 predictions are for the price of BTC in 2 years time.

The full write up is at: http://hashingit.com/13-megawatts-of-mining

I like the chart, but I cant agree with your assumptions. IN particular electricity costs. Average electricity costs are irrelevant, what matters is lowest electricity cost (where mining is feasible). Because that is where mining will relocate, and places where electricity price is substantially higher, miners will have to shut down. Here is a read for you:

http://www.spokesman.com/stories/2014/apr/26/northwests-cheap-power-drawing-bitcoin-miners/

A map comparing energy rates led them to Central Washington, where hydroelectric dams churn out electricity that costs industrial customers less than 2 cents per kilowatt.

You are assuming 10x that price.

Secondly you are incorrectly factoring in asic "improvments". THere are two factors that matter, price per GH and watt per GH. The first one doesnt need Moore's law to improve dramatically from where we are today. ALl thats needed is difficulty going up further, and prices will come down proportionally. We are no where near asic manufacturing costs yet, in fact today marginal production cost of these chips is so low, that its irrelevant, they might as well cost nothing. Asic alone, I estimated HF golden nonce elsewhere at $0.04/GH in wafer production and packaging cost (assuming they can hit 800GH with their rev).

As for watt/GH; also there we do not need Moore's law to get a doubling in 2 years. THe very same asics being sold today almost certainly have the possibility to be downclocked and downvolted to achieve that doubling (or more), be it at the expense of performance per chip. ALmost all miners sold today are being run and sold at the top end of the voltage/clock shmoo plot, which is evidenced by the fact that they typically barely overclock at all. To see what changing of voltage and clock can do, look at Bitmain S2, which is the exact same chip as in the S1, but is twice as power efficient. Im willing to bet KnC, HF, CT, etc can easily double their power efficiency too on their existing designs, once that trade off becomes worthwhile. But since electricity cost is only a tiny fraction of most miners cost, and pricing of hardware is far more dependant on hashrate than power efficiency, it doesnt make sense for 28nm designs today. Doesnt mean they cant.

Well part of the reason I chose to write this simulation was so that it's easy to adjust the assumptions and see what happens :-)

For the electricity charges I didn't factor any other overheads such as the PUE of the facility or attempt to amortize any other costs. Certainly though a 2c per kWh cost will allow for a larger peak hashing rate - the slowdown still happens, just on a slightly different curve. I just ran the numbers again and looked at the projected hashrate with a final BTC price of $900 (for consistency). With 20c/kWh we hit a rate of 640 PH/s, 10c/kWh we hit 766 PH/s and 2c/kWh we hit 893 PH/s. That goes to my comment that lower electricity costs certainly support a movement towards more centralized infrastructure where electricity and cooling are cheaper.

For W/GH/s then I agree that voltage and frequency scaling can probably take things down to 0.325W/GH/s at a loss of probably something like 30% of the performance (assume a square law - it's over-optimistic but it's somewhere to start). That would give us an improvement of about 1.4x but means we need more ASICs to hit the same hashing rates. That's part of why I'm pretty happy that we can get a 4x overall improvement in the next 2 years. Moore's Law would normally give us 2x as a result of die shrinks and so I'm assuming another 1.4x to be found by improvements to the ASIC designs.

The simulation allows that initial 1.4x to be factored in very easily. Let's assume that an ASIC vendor reduces the cost of their devices so that the cost/GH/s remains the same (they're effectively reducing prices 30%) so now 2 devices use the power that 1 previously did but we're now getting 1.4x the hashing capacity. Now the numbers work out as: 20c/kWh we hit 766 PH/s, 10c/kWh we hit 842 PH/s and 2c/kWh we hit 910 PH/s

I totally agree that the equipment cost is the biggest single factor. I'm not sure that anyone currently producing mining ASICs can get to $32 per device though - they have lots of business overheads that have to be amortized with the current batch sizes that they're all building (my guess is that those overheads dominate in fact). I doubt any of the mining ASIC companies are getting particularly great fab costs right now for much the same reason - they're not building a couple of million devices a month. Could someone dramatically change this? Absolutely yes.

Again, let's try halving the cost of each hash immediately and we're still folding the 1.4x we just took so now our manufacturer's are shipping devices at 2.8x lower cost: 20c/kWh we hit 1270 PH/s, 10c/kWh we hit 1507 PH/s and 2c/kWh we hit 1699 PH/s.

Certainly those final numbers are higher. The original projection with 20c/kWh and current operating behaviour works out to be an 11.2x increase over the 57 PH/s we started with. The most optimistic projection with 2c/kWh would now be a 29.8x increase - they're still dramatically slowed down from the past 12 months.

[Puppet]

With 20c/kWh we hit a rate of 640 PH/s, 10c/kWh we hit 766 PH/s and 2c/kWh we hit 893 PH/s.

Which only shows your assumptions cant be right. Ultimately mining is converting electricity in to bitcoins, you would expect a near linear correlation, not a 40% increase for a 1000% decrease in elecricity rates. The only thing preventing that from being truly linear is hardware amortization. YOu must have quite unrealistic assumptions for those.

For W/GH/s then I agree that voltage and frequency scaling can probably take things down to 0.325W/GH/s at a loss of probably something like 30% of the performance (assume a square law - it's over-optimistic but it's somewhere to start). That would give us an improvement of about 1.4x but means we need more ASICs to hit the same hashing rates.

? The fact you would need more chips to achieve better efficiency doesnt again reduce the efficiency. IT does increase cost, but not compared to todays market prices, only compared to todays marginal production costs.

The simulation allows that initial 1.4x to be factored in very easily. Let's assume that an ASIC vendor reduces the cost of their devices so that the cost/GH/s remains the same (they're effectively reducing prices 30%) so now 2 devices use the power that 1 previously did but we're now getting 1.4x the hashing capacity. Now the numbers work out as: 20c/kWh we hit 766 PH/s, 10c/kWh we hit 842 PH/s and 2c/kWh we hit 910 PH/s

Im not sure you understand how mining hardware pricing works. Only two factors really matter; miners (as in end customers) breakeven point or perceived breakeven and manufacturer breakeven (on marginal costs). Currently these two are several orders of magnitude apart. As a result, miners are being sold for prices that miners perceive as potentially profitable. BUt that price point is linearly correlated with difficulty, so those prices will keep going down, way, way faster than Moore's law, at a speed limited only by manufacturing capability, until prices approach marginal production costs of the most efficient vendors. Now you can debate what those are, but asicminer claims $0.2/GH wafercost and my guess is that larger MCM 28nm devices are at least 2-4x cheaper per GH.

I totally agree that the equipment cost is the biggest single factor.

Actually, I dont agree Electricity cost will become the biggest single factor IMO

For a ballpark guess, lets take $0.2/GH as marginal hardware production cost and 0.5W/GH efficiency. So each TH would cost $200 to produce and draw 500W. 500W @ 3 cents per KWH (including cooling, overhead) = ~$131/year.

$200>$131, however we are looking at the "endgame" here, and therefore hardware prices will stabilize and their value loss can be written off over several years. If you just amortize over 2 years or assume a resale value of 50% of the purchase price, electricity has become the biggest factor.

I'm not sure that anyone currently producing mining ASICs can get to $32 per device though

If by device you mean the asic itself, then obviously that can be done easily. If you mean a complete rackmountable miner, then obviously not, but in 6 or 9 months, no one is going to be selling miners in fancy enclosures with touch screen LCDs. Look at KnC's datacenter, or bigmegapower; naked PCB's on "ikea" shelves. Thats how it will look.

they're still dramatically slowed down from the past 12 months.

Now thats something I and I presume everyone else will agree with. No one ever expected a monthly hashrate doubling to continue perpetually.

[jimmothy]

As for watt/GH; also there we do not need Moore's law to get a doubling in 2 years. THe very same asics being sold today almost certainly have the possibility to be downclocked and downvolted to achieve that doubling (or more), be it at the expense of performance per chip. ALmost all miners sold today are being run and sold at the top end of the voltage/clock shmoo plot, which is evidenced by the fact that they typically barely overclock at all. To see what changing of voltage and clock can do, look at Bitmain S2, which is the exact same chip as in the S1, but is twice as power efficient. Im willing to bet KnC, HF, CT, etc can easily double their power efficiency too on their existing designs, once that trade off becomes worthwhile. But since electricity cost is only a tiny fraction of most miners cost, and pricing of hardware is far more dependant on hashrate than power efficiency, it doesnt make sense for 28nm designs today. Doesnt mean they cant.

You do realize bitmain s1 could achieve 1.2w/gh at the wall right? That's a tad bit more than half as efficient than an s2(1.1w/gh)..

Got any better evidence of your claims?

[TheMinex]

And a lot of asic companies will stop making machines because of low demand, and negative return on investment as electricity will cost more than the btc it mines.

Yes.  And this, combined with inefficient miners going out of business, should drive the difficulty back down to the point where it makes sense for the most efficient miners to continue or resume operating...

And then there is the wild card introduced by the price of BTC, which the large players have more control over than the small ones.

So can we expect some kind of "BTC business cycle"?

It's not a large or small players affair. It's all about efficienty. Large player will never be able to get free electricity, and like someone said, they are in disadvantage in terms of efficienty due to cooling (I add + setup e maintenance costs )  Small player, watching the other hand, can mine even at loss for very very long period hopeing the BTC price will rise, can easily have free electricity and can work 'for free' in maintenance & setup.

Those are the whys miners are now trying to kill each other, but I completley agree with you that, in this rude scenario, the first thing to fall down will be investments in mining hardware, and consenquentely in mining makers margins, probably leading a situation that you describe; a strange BTC business cycle driven only by huge price rise ( small rises will be compensate by worst efficienty miners that have switch off (maybe after a redistributio/decentralization due to a try to recover initial hardware investment by both large & small players ?? ))

....what a wonderfull strange world  

[Puppet]

You do realize bitmain s1 could achieve 1.2w/gh at the wall right? That's a tad bit more than half as efficient than an s2(1.1w/gh)..

You moron, S1 and S2 use the same chip. So of course S1 is capable of achieving similar efficiencies if you underclock and undervolt them just like in the S2. Thats the whole point. The difference is that S1 by default uses a clock and voltage setting that results in ~2W/GH at the wall while for S2 they picked a point that roughly cuts that by half.  

Got any better evidence of your claims?