Bitcoin Forum

Looking at the difficulty estimate from https://bitcoinwisdom.com/bitcoin/difficulty at the time of this post, there's a possibility that there will be a negative difficulty change. Small one, and there are still ~6.4 days to go, but what type of impact would this have on BTC prices?

Estimated Next Difficulty:   39,581,348,249 (-1.78%)
Adjust time:   After 905 Blocks, About 6.4 days

Price at Bitfinex = 368.5 USD

Any guesses?

No impact per se, as just a single adjustment would have, but I believe there is a close relation to the hunter-prey equations. (http://en.wikipedia.org/wiki/Lotka%E2%80%93Volterra_equation)
http://upload.wikimedia.org/wikipedia/commons/a/aa/Volterra_lotka_dynamics.PNG

Nice graph EM!

Great reference!

However I'm not sure if this relationship does really represent the complete scenario, because there are more than two variables who interact:

1. the mining difficulty
2. the miners
3. the BTC price
4. the market participants (buyers or sellers)

ya.ya.yo!

Unlike 2011, we might never see a period of continuous difficulty drop, now this trend is too big to stop and there are too much reserve hash power from retired 1st and 2nd generation ASIC miners, a slightest drop in difficulty will re-activate those miners thus immediately push back the difficulty to where it was

it should be more difficult, more expensive. why the price is not rasing  :(

Difficulty has little observed effect on price. Difficulty has gone to the moon while the price went down slightly.

Price affects hash rate. Miners who are losing money drop out.

I think you guys have it backwards.  The cause is the price, the effect is the difficulty.
Difficulty has almost no influence on price, because the supply of coins is independent of difficulty.  
The price is the incentive of mining, and hence, of difficulty.  As long as the *cost* of difficulty is below the price,
there is an incentive to mine more, which will result in higher difficulty.
When the cost of difficulty is higher than the price, mining is not profitable any more.
A miner can choose to continue and go broke, or stop the losses.  Difficulty will hence decrease until
mining becomes profitable again.
The cost of difficulty as a function of difficulty is given by technological advances.
But I don't see at all how difficulty can have any influence on price.
Difficulty should follow price (with some lag), corrected for technological advances.

Those who have active miners will continue to mine regardless of difficulty, and drops in difficulty for 1 or 2 periods are not going to encourage more people to mine. Prices are more dependent on demand and news.

I don't agree. The only circumstance where difficulty would have no influence on price is one where someone could pay for the expense of mining with bitcoin. This is most likely not the case, and miners must sell the btc they mine to pay for things like electricity, which would reflect the downward pressure on BTC price. However, with a decrease in difficulty, miners are effectively producing more at the same expense, which may alleviate some of that selling pressure. How much, I'm not sure.

Miners don't produce more.  The production of coins is independent of difficulty.  It will always be 3600 coins a day (until 2017 or so).  Satoshi produced 7200 coins a day with his PC (before the last halving).

What you are saying is that miners, at high difficulty, will have a different saving/spending attitude for bitcoins (that is, they will only convert what they need to cover their costs) than when the difficulty is lower.  In as much as miners are hodling, that may play somewhat, yes.  

In as much as miners systematically convert into fiat, and take their profit in fiat, there is no effect either.

If the total network hashrate is 277 Petahashes and stays constant, won't they produce 3600 coins faster at a difficulty of 39 bln than they would at 40 bln?

The difficulty re-targets after 2016 blocks.
Simply put if the hashrate is higher than anticipated more blocks are found in a shorter time so difficulty increases to compensate for that.
If the hashrate drops it takes longer for the 2016 blocks to be found and difficulty decreases after the next re-target to compensate.

Difficulty adjustment is slow so it can lag behind hashrate changes.

[edit] To clarify. Think in terms of the time it takes to find blocks and not the amount of bitcoins found.
If your hashrate increases you will find blocks faster than anticipated so the 2016 blocks until the re-target are found ahead of schedule.
The protocol will then increase the difficulty so if the hashrate does not increase again the timing should be roughly on target

[edit2] to consistently create blocks faster than anticipated the hashrate has to keep increasing.

finally happening - difficulty going down!  I see minimal impact.  I seems support at low 300's is incredibly strong - still remember the famous 30K ask wall at stamp got gobbled up.  Without the auction, I say prices is stabilizing.  But the auction may set a new ceiling to the price - which maybe lower than where it's trading.

People can't enjoy a lower difficulty for more than two weeks as more miners will just put their machines back to work.

But it's a welcome drop, for once people don't see their rewards drop  :)

People are not willing to pay a premium for bitcoin & as the demand falls the price will only continue to fall... Miners minting bitcoin is just jet fuel on the fire  :'(

He said miners will mine more at the same expense, not just "more."  That is true if difficulty were to drop.  Individual miners will use the same electricity to produce more coins for the same cost in power. 

Miners will not "mine more at the same expense".  They will mine the same thing at less expense.
Individual miners cannot "produce more coins for the same cost of power" if not other miners "produce less coins for much less power".

Because the total number of coins mined must be the same.
If some use "the same amount of power" and hence "mine more", then that means that other miners mine less, and consume vastly less power (maybe stop mining all together).

Globally speaking, the miners mine the same amount of coins, and use less electricity.  The do not mine more coins with the same consumption.  Some do, and others mine less. 

That means that miners make more profits (or lose less money, depending where they come from).

The point is that it could be that those increased profits go into hodling coins.  That's the ONLY way in which difficulty can have an influence on the price: by increasing the demand for holding coins (by the miners themselves, because they want to allocate their profit in coins).

this makes no sense.  how are miners jet fuel on the fire.  they have always been there producing the same amount of coins regardless of price.

Ummm because of how a marketplace works??? I understand they have always been producing the same amount the point is LESS PEOPLE ARE WILLING TO BUY BITCOIN & facts are more money is being changed into USD than ever before from BTC. People were willing to pay over a thousand dollars for one coin before! Now you can't find people willing to pay 400 because the miners are racing each other to sell you their coins first and not their competitors.

Bitcoin price is being decreased so it can become more mainstream! Don't fight the cheap coins coming soon!

 If difficulty goes down then a miner will mine more coins with the same machine.  The same machine will obviously continue to use the same amount of power.  Not sure where you are getting your information but it is inaccurate.

I want some of whatever he is smoking. Almost had me convinced.

all apple farmers are evil too because they compete for the sell-price as well!
but the most evil is wheat! just see how the price is fluctuating... fucking bastards

 I know! I had to read that a couple of times.  I'm pretty sure he could get a marketing job with BFL or KNC.

he's probably already working with GAW. seems to have the correct intellect.

Don't get off topic guys...

The point is the miners always been producing the same amount the point is LESS PEOPLE ARE WILLING TO BUY BITCOIN & facts are more money is being changed into USD than ever before from BTC. People were willing to pay over a thousand dollars for one coin before! Now you can't find people willing to pay 400 because the miners are racing each other to sell you their coins first and not their competitors.

Bitcoin price is being decreased so it can HOPEFULLY become more mainstream! Don't fight the cheap coins coming soon!

The total number of coins mined remains essentially unchanged if difficulty goes down.  
If, as miner A, you keep your machine running at the same power, you will of course mine more coins.
But that has to be compensated by another miner B who is mining LESS coins for the sum to remain the same, right ?
Now, with difficulty going down, that means that in order for miner B to mine less coins, he has to reduce power consumption, and more than proportional to the reduction of the amount of coins mined (as difficulty went down), right.

So yes, some miners, like miner A, will mine more coins at same power, but other miners, like miner B, will mine less coins at much less power.

So on average, your miners are going to mine the same amount of coins, and use less power.  That is exactly what difficulty expresses !

If all miners were to use the same power and mine more coins, difficulty would not go down.
The very fact that difficulty goes down means that power consumption (at equal technology) goes down.  As the total number of coins mined is a given, it means that a miner, on average, spends less power and mines the same number of coins.  Of course, some will mine more and other will mine less.

He still doesn't get it  ::)

Individually a miner will mine more at the same expense ( a drop in hashrate that leads to a drop in difficulty means your individual share of the overall hashrate increases, hence you are more likely to find a block. In terms of pooled mining it effectively means you see a higher income in btc paid out by the pool for your provided hashrate).

Overall however (all of them combined) miners will use less resources to produce the same amount of blocks (given that the hashrate then goes into a stable state where difficulty does not change).

Your statement is only true if you regard the entire network of miners.
It does not make sense if you think about it. If 10 people share the production of something equally everyone gets  1/10th.
If 2 leave (drop in difficulty due to drop in hashrate) the product is shared between 8 and not 10.
If the amount produced stays the same regardless of how many are contributing (this is where bitcoin uses difficulty to adjust that production) the remaining 8 will get a larger share.

Obviously if leaving also means stopping your input of resources so overall the production cost sinks by the amount of resources you stopped supplying.

Yes. 

I saw this as such:

10 miners mine 1/10 at an expense of 20 (each spends 2 and produces 1/10).

8 miners mine 1/8 and 2 miners mine 0 at an expense of 16 (lower difficulty).

Those 8 miners spend each still 2 of course, and they mine now 1/8 instead of 1/10, so they "spend the same and produce more", but 2 miners used to spend 2 and spend 0 now, and produce 0.  Those miners "spend less and produce less" (here radically 0).

The *average miner* now still mines 1/10 (8 at 1/8 and 2 at 0), but spends only 1.6 (8 at 2, and 2 at 0).

So the average miner mines the same, and spends less.

Here is your conceptual mistake. Using an arithmetic mean makes no sense here because you want to look at the individual miner and you do not want to include miners that are not mining)
Bob and Alice are still mining by spending 2 resources so they spend exactly the same as before but receive more.
Satoshi is no longer mining so he is spending 0 resources and receives no more coins (not that he needs them ;-) )

You are basically taking satoshi and adding him to your mining calculation which logically makes no sense.

[edit] do you see the problem in your assumption? I can add an infinite set of miners that do not mine (does not affect mining at all) and it would make your average miner spend close to 0...

This becomes nitpicking, but OK.

Consider the fixed group of people who will mine "before" and "after" the difficulty drop.  Let X be a randomly chosen miner in that group.

For some choice X (say, Alice), the production will maybe increase, but for another choice of X (say, Bob), the production has to decrease.  That was my point.  The extreme case is when Bob stops mining all together, and for that special case, you eliminate him from the group.  But it could just as well be that Bob switches off 90% of his mining equipment.  Then he's still mining, but much less.  The case "he switches off 100%" is simply an extreme case of "he switches off part of his mining equipment".

Somebody has to switch of mining equipment to have the difficulty go down, otherwise it wouldn't go down. 

No, it's not nitpicking, it simply correcting you, on account that you were wrong in your OP statement.

A miner is an entity that is hashing in an attempt to find a block and he is spending resources to hash.
If you treat Bob as two entities (Bob and Bob' where Bob keeps mining while Bob' quits) your problem reduces down to miners that keep mining and miners that quit.

A miner (or part of a miner) that quits gets 0 while spending 0 resources while the others spend the same amount of resources for a larger share (basically the amount of bitcoins earned by the now stopped miner is distributed proportionately to the hashrate over all remaining miners).

The set of miners that stopped mining should not be a part of the group you want to sample because they are doing nothing.
Like I said before I can create an infinitely large group of non-miners and they have 0 effect on the bitcoin network.
Including them in any calculation on miners is incorrect because you are altering the total number of miners by including non-miners.

What you are trying to do is include non-miners (well miners that stopped) in the set of miners.

[edit] by the way I am totally ignoring the fact that difficulty lags behind hashrate changes and that hashrate is not fixed so the actual model is more complicated but overall it more or less follows the described mechanism.

Yes, that's obvious.
You could say that the mining units still running, will mine more coins, while they are consuming the same amount of power.  Only, there will be less units running.

If all miners in the world stop, and I keep my PC running, I mine all the coins in the world and my PC will even consume somewhat more (because I don't even run a mining software right now of course).

The important point is that in order for the difficulty to decrease, miners have to stop mining, or to decrease mining, and will hence consume also less or nothing any more.

My point was that if you considered the CHANGE for a given miner "before" and "after", you shouldn't subselect only the sample of those that kept mining.  The change for those that stopped mining was also a change: namely from mining (a lot) to mining less or nothing any more.

It would be like saying that soldiers become stronger and more victorious in war time.

No, soldiers that survive are maybe stronger and more victorious, but a lot of others died.  If you eliminate those that died, the balance of those surviving is not the "typical evolution" of a soldier in war time.  There are those that are victorious, but there are also those that die.  If you don't count them "because when they are dead, they are not an active soldier anymore", then of course war looks like a fun party for soldiers :-)

But I think we get each other's points.

So when I'm running my miners at home at 1 TH on Monday and difficulty drops on Tuesday, but there is no change in what I pay for electricity, can you explain how I am mining the same thing at a lower cost?

Also, isn't it true that a difficulty drop could be caused by something other than miners going offline? If the prior difficulty increase was too great, wouldn't the difficulty drop to compensate?

Of course, *until* the next difficulty adjustment, you can mine somewhat more.  But only for 2 weeks.


yes, but that's on a small-time scale, of the order of one or two adjustments.  I admit that I was thinking on longer time scales where these small correction fluctuations are smoothed out.

I would guess that in general, as long as difficulty is for a long time rising, there is actually a little bit more mining (that is, somewhat more than 3600 coins per day), because when difficulty rises systematically, there is each time more than 3600 coins/day mined, and that's why difficulty is adjusted with a lag of 2 weeks.  On the other hand, if difficulty lowers systematically for a long period, there are somewhat less than 3600 coins mined per day, because that is what makes difficulty fall.

But I would expect these to be rather small effects.

 You really need to read some factual material.  If all miners stop mining and you fire up your PC, the odds are that you will get sweet f$ck all Bitcoins.  You will probably never mine a single block let alone the 2016 needed to get to a difficulty retarget.  For god sakes man you cant learn a subject by preaching about it.  READ:

https://en.bitcoin.it/wiki/Difficulty (https://en.bitcoin.it/wiki/Difficulty)

You know, I was following what you said after your first post because price does seem to lead difficulty but that soldier analogy is just bizarre. Totally bizarre.
 

Yes you're right if they switch off so fast (that is, within just a few periods) that difficulty is not adapted to their switching off.  Then indeed, to get to the difficulty-lowering with a PC, I would still be stuck for a while with a HUGE difficulty as you point out.  As I pointed out already, I was talking about evolutions much slower than the 2-week period of adaption.

We get here an extreme example of what I said, that during difficulty-decrease, somewhat less than 3600 coins are mined per day.  Here, the step in difficulty, if taken at once, is so huge, that this "somewhat less" changes into essentially 0.

I'm assuming all changes slowly as compared to the 2-week adjustments.


I thought it illustrated quite well what I wanted to say.  If difficulty goes down (for a long period), and you look what that means for someone who has been mining, it means that maybe some of them will mine more coins at same power, but at the expense of others who will mine less coins with less power, or some who will even stop mining.  Not all miners who were mining before, can mine more coins at same power !  If some mine more, others have to mine less.  If all of them do the same thing, then it simply means they all consume somewhat less (switching off part of their park) and will mine the same amount.

I was looking at the viewpoint of a random miner BEFORE difficulty went down, and how things happen to HIM. 

That is like looking at a random soldier before he goes to war, and look at what happens to him.  Some will die, other will become victorious.  Just picking out those that will become victorious, and saying that going to war renders you victorious, was in my eyes similar as saying that when you're a random miner, when difficulty goes down, you'll mine more coins.  No.  Some will (like some soldiers will survive and be victorious).  Others won't and will stop mining (like the soldiers who die).