Bitcoin Forum

I've read at multiple places that some of you think that a higher hashrate means a higher price, while in fact I think it's the opposite.

But before, here's a little economic theory:

People will do hard stuff to obtain something because it is valuable, not the other way around.

With concrete example:

A man will go underwater to get a pearl because it's valuable; the pearl is not valuable because a man has to go get it.

People go deep to mine gold because it's valuable; gold is not valuable because people have to go deep to mine it.

In relation to Bitcoin, we could say that miners mine Bitcoin because it's valuable.
(Otherwise they would just fork the client and mine their own coin. Clearly the "mining" in itself isn't the goal.)

Further than that, here's a little graph:

https://i.imgur.com/rWxurTM.png

What I want to show you is that a surge in price occurs before an increase in hashrate.
The price is driving the hashrate, not the other way around.

Each time there's a little bump in the price, you can see a response in the hashrate, lagging a few months behind.
The lag is understandable, as it takes time to purchase and set up a mining rig.
In case of falling prices, it can take a while for the miners to accept to stop their operations.

Some of you argue that the current rising hashrate indicates a future higher price.
On the contrary, I say that the rising hashrate is only a response of the recent price increase.

I agree with your overall point, but the data doesn't quite tell the story you are trying to portray.  The piece you are missing is that the 2011 run up and the recent run up in difficulty were both triggered by a new technology that dramatically improved the efficiency of mining.  In 2011 it was GPUs and in 2013 it is ASICs.  ASICs are still being deployed and I expect a minimum of triple the current hash rate before we see difficulty top.  As much as 10X the current difficulty may happen, depending on how the price goes.  <-- Again, I agree price ultimately drives difficulty, but technology improvements shift gears.  That being said, as soon as 28nm ASICs make up the majority of the network, we will see technology improvements capped by Moore's Law.  We will then be in "top gear" and we can drive this thing like we stole it.

I'd be interested not in the hashrate but guesses on an average electricity used vs price. ASICs are a pretty big change in technology and the hashrate may be distorted for this transition period.

New and improved technologies do push the hashrate up, but one can still see the clear relation between price and hashrate.

In your explanation, where do the FPGAs fit?

Yes, you are right.

Most people have it backwards.

Mostly it's just wishful thinking on their part that more difficult mining will somehow cause their coins to be worth more. It's not true at all.

So, if you are right, then you expect hashrate to imminently begin to go sideways for a couple of months?

I expect it to follow the same pattern as the price: to peak and slowly diminish after that.
Last price peak in 2012 induced a hashrate increase that lasted for months.

Anyone has any data pointing to a reverse relationship, where hashrate would be driving price?

Imagine two identical gold mines, mined by two different companies, both offer stock. Mine Co A has a bunch of labourers with pick-axes and donkeys to pull mine carts, Mine Co B has a drill, dynamite, rail engines, and rock crushers. Obviously Mine Co B's stock will be worth more because they will be more efficient.  

Hash rates are simply a lag indicator of price...

Except Bitcoin isn't a stock in your example, it's the gold.

Well then you're gonna have some explaining to do in regards to timing because we are not going to peak and slowly diminish any time soon as Avalon batch 2 will be delivered in 1-3 days followed by batch 3 and more asicminers.

I also disagree with your visual interpretation of price always leading. If you really zoom in and look carefully I could make the opposite argument that hashrate leads. It's just that the price spikes are so violent in comparison that it looks like they're leading.

Your gold mining analogy is also off as gold miners do not have a philosophical motivation to help or secure a network as do bitcoin miners.

Is the collective wisdom NOT buying Bitcoins purely because they deem the difficulty too low? If not, it has no effect on price.

However, I believe there is another effect, and it's that an increase in profit margin is bullish because miners can afford to speculate with a larger portion, ie hold them. Currently, the profit margins are of course rapidly decreasing due to sideways/downwards price action and still skyrocketing difficulty. My assumption is that the percentage of the daily mining subsidy (3600 BTC) that is released to the market is increasing because the costs that have to be covered are rising.

I don't expect hashrate to top out very soon, and that's thanks to the incompetence of the suppliers who are lagging extremely. There's already hardware for multiple hundreds of TH/s it seems. I do expect it to top out again sometime, as always, people will order far more ASICs than profitable for some of them in this arms race and some of them will have to drop out. Funny thing though that the prices for ASICs can plummet far further on the open market than GPUs because they have no other purpose than generating profits in Bitcoin's POW. Quite risky particularly when you have no clue when they will ship.

So, what you are saying is that the Avalon batch 2 and the batch 3 of asicminers are not going to increase the hashrate?
Because such a jump in the hashrate sure would look like a peak following the price in my book.

Reddit was full of people investing in mining a month ago. Now it's rather... silent.
My guess is many bought mining equipment and they will receive it in the coming weeks/months.

Do it! I'm open to discussion. If you can zoom in and circle the interesting parts, I'll be happy to look.

Ok... and the consequences of the philosophical motivations are....?



While some argue some of my points, I've yet to see an opposing argument showing that hashrate influences the price.

Indeed.
I've yet to see a single person making that argument.

"Eh, Bitcoin is interesting, but the hashrate is too low for now... network isn't secure enough."

"Wow, Bitcoins look like an interesting hedge. Time to allocate a few % of my holdings and pour a few millions into it – ah, the difficulty is only 15 million and hashrate 150 TH/s. How overvalued. Will wait a couple months, then." ;D

Seriously, back when we routinely had people buying multimillion USD worth of BTC, I don't think the difficulty was of any consideration. I don't believe it is a metric people base investment decisions on, other than mining hardware.

One case that can be made though is that it's a cyclical alternation between investment in BTC and investment in mining hardware as the one gets unprofitable and the other more attractive. But I'm not sure, most of the time, both are attractive at the same time and unattractive at the same time.

Since being in bitcoin since January of 2011, the only real dip in hashrate that I witnessed was for a brief few months from about August to November 2011. Otherwise it's been steadily if not parabolically increasing.

My explanation to fit my perceived notion and experience as a miner of this is that there is literally a rush to secure as many bitcoins in hand as fast as possible in anticipation of a price rise and before the asymptotic decrease in rewards kicks in. This is all irrespective of the price.

Also there is a community aspect to bitcoin mining that does not exist for gold.  It's not purely profit driven.

Interesting. If they want bitcoins, why wouldn't they just buy, instead of mine?


Ok.
Does that give the hashrate any way to influence the price?

Here (https://bitcointalk.org/index.php?topic=7427.0) is a great past analysis of it that begun in early May 2011.


There was another dip beginning December 2012 following a period of sideways. Two consecutive difficulty decreases, one increase and another decrease that ended up the bottom.

Sure, some are more inclined to mine at a loss etc., but the upwards motion is first technology and second price driven.

That actually allows me to highlight another reason why miners mine.

For enjoyment. They're geeks after all and this is what they do.

Hashrate influences the price by making speculators more confident that their money won't disappear from a 51%attack.

The cyclical alternation is probably the best explanation as there are no absolutes when talking about anything related to prices.

Why not mine their own little Bitcoin fork? Same hassle, same fun?  ;)
But yeah, I see your point.

Like I said, I've never seen anyone saying he is concerned by the "low" hashrate.
I also think this is way above the head of many speculators.

We must not forget free electricity for the average college geek miner.

And then who doesn't like the idea of printing up money?

Especially when you don't have to dig.

You're right. This is very much over the heads of most speculators.

I talked to a couple of guys at the conference looking to get involved by buying up alt coins. I warned them that speculative attacks could easily be carried out by bitcoin miners looking to destroy competitors. The thought clearly had never occurred to them.

Sure enough in the very next session crazyrabbit explained to everyone how freicoin had just gotten taken down by an attack. And now more recently we get another attack on feathercoin?

The mind of a miner is a unique thing.

This may be true in the real world but it's not true in Bitcoin world.It would only be true if the mining companies were allowed to mine as much gold as they wanted, but that is not the case in Bitcoin world, you can only mine a set amount per day.

So what would happen is this:

The mine owner would go to Mine Co A, see they only have labourers with pick-axes and donkeys and so tell them "you guys can go mine on the edge of the mine where the rock is soft and the gold is easy to get to, that way you can still get your 100 nuggets per day

The mine owner would go to Mine Co B, see they have all the best stuff and say " because you guys have all the best stuff i'm sending you deep into the mine to work on the hardest rocks that are tough to get out, so you'll also end up getting 100 nuggets per day.

In Bitcoin world company B would NOT be much more valuable than company A, they would be worth the same amount because they both produce exactly the same amount of gold per day/month/year

I am not involved in mining at all so could be completely wrong BUT from the anecdote above it i would probably have to disagree with Frozenlock. This is because:
A hash rate that goes higher and higher means it's harder and more expensive to mine the nuggets. If companies were having to buy newer, tougher more expensive machinery to mine and the nugget stays at $1 you would hit the point where all companies would go bust because the rising cost of the evolving machinery would mean that no one could make a profit, so the price of the nugget would HAVE to keep increasing so that the companies mining were always able to turn a profit.

or to take Frozenlocks example:
"People go deep to mine gold because it's valuable; gold is not valuable because people have to go deep to mine it."

Actually, if it only cost you $1 to go deep and mine that gold you'd be happy selling it for $5, but if it cost you $10k to go mine that gold would you still sell it for $5? No of course not, the price would be over $10k because you'd still have to turn a profit. In a sense he is right that gold is not valuable because people have to go deep to get it, BUT it's value IS and WILL BE determined on how much it costs you to get deep enough to mine it. If you could go out with a snorkel and grab a pearl you'd be happy to sell it cheap, if you had to buy a super yacht and build one of those funky submarines to grab the pearl would you still sell it as cheap? of course not, the bottom line price of the pearl WOULD be determined by the cost of obtaining it.

If the hashrate goes up and up then it means it's becoming harder and more expensive to mine the bitcoins, which means the price will have to increase.

Exactly, most people couldn't give a fuck, and that's why I find the idea questionable of there being lots of money on the sidelines but just waiting for difficulty to increase.

You are right on the money about the scamcoins, and that includes Blightcoin as well. There can be no second proof of work chain in the long term, doesn't even matter if the algorithm is different. Eventually, it will be attacked by people with vested interest in Bitcoin. It all amounts to sucking money from unsuspecting people.

Ok.

Let's look at the period between November 2011 and December 2012.

I contend that the hashrate was going straight up while the price kind of bumped along sideways and slightly up.

I could make the argument that that hashrate rise is what led to the parabolic price spike beginning in February.

Not exactly.

If it costs you $10k to dig an ounce of gold, it doesn't mean that's the price it will have on the market.
It means you won't mine it unless gold is valued at more than $10k.

Same thing with mining bitcoins.
If it becomes more expensive to mine, less people will mine.
(While the supply should stay more or less the same, thanks to the network automatic adjustment.)

Really, labor theory of value? :D

I'm preparing for the broken window fallacy.  ;)

You could.  ;)

The relationship would be far less obvious... and it seems the price bottomed before the hashrate.
Unfortunately I don't have the tools to make a more detailed graph of this period (at least not for the hashrate).

You could say it's less obvious but we both are stuck having to explain the complete divergence occurring this very moment between the two.

But if price breaks up following the hashrate then...  :D

Lag between price and hashrate, like every other time?

Some of the rise in hashrate is probably because of the price-peak. But the hardware coming online these days have been ordered months ago.

I think the rise was due to both increased interest and miners who wanted to increase their holdings without buying mining-equipment since they knew difficulty would soon skyrocket anyways.

More btc generated daily, miners selling as they mine, competing with each other, lowering their prices as they go. Disappointed low hashrate miners running like a rat from a sinking ship sell everything they have, resulting in even lower price.

Also: who BUYS btc?! Those who want to get on with mining, to get themselves some asics, etc? Traders who want to get into while they can? in General - people who want to get profit from BTC. And profit for everyone is not possible.

As soon as it fails to look profitable enough, people will lose interest, and will not buy anymore. Guess what kind of result it will lead to.

this thing is like a ponzi, kept alive by greed to a certain point.

BUYERS:

Not a Ponzi, a pyramid.

People joining the pyramid can very much continue to profit provided the size of the pyramid can grow to accommodate them. Everyone can  profit, up to a point. The real question is where that point lies.

By your own admission, Bitcoin is as yet under-developed. So if one assumes that Bitcoin will become more useful, then the pyramid can and will grow. If BTC is to be successful as a real currency, that point may lie very far in the future.

(On the flip-side: examine your own beliefs... If you believe the pyramid cannot grow further and there can be no future use, then why are you here?)

SELLERS:

Supply is constant, so it can effectively be factored out of the above argument. I therefore agree with the OP -- I don't see how hashrate is relevant to price.

Yeah, a pyramid was what I meant.

lackluster language skills on my part

This is an interesting view, if more people join the mining operation, their profit margin will shrink and create a downward pressure on the price. But this is under the assumption that they have no fiat money to cover their cost, actually many people are converting their fiat into bitcoin in this process

IMO, price is always a leading indicator for fundamentals, the reason price increased so fast at the beginning of this year is because of reward halving and difficulty rise caused by ASIC devices. Just after Avalon batch 1 delivered, the price erupted. Many investors start to quit mining and buy coin directly (Because there is no way to get a ASIC device in a reasonable time frame)

Currently most of the investment money is going into ASIC device order, there are less support for the exchange price, but soon after the number of ASIC devices reaches 36000, average per device return will drop to 0.1 coin per day, at that stage investors will have no other choice but buy coins

More BTC generated daily?

The difficulty is readjusting to keep the supply at a fixed rate. Even though this means that for a short period of time slightly more BTC's than targeted are mined per day when hashrate increase drastically, It sounds like you have misunderstood how BTC-mining work.

Slightly = at least 20% witnessed by my own eyes.
Do not know about precise numbers.

A pyramid scheme will collapse when the available money runs out. That's the reason housing collapsed when FED start to tighten the money supply in 2007

But for money, it is very different. Have you ever heard about a pyramid scheme of money? No, because money itself is the unit of counting and its value is perceived as a standard. Even in a very deflative environment like great depression in 1930s, the value of money never rose by more than 40%

pyramid scheme collapses not when the money runs out, but when the interest does

Pyramid schemes collapse when you run out of idiots.

All bubbles are pyramid schemes, and bitcoin has just been in a bubble. However bitcoin itself (long term) is not a pyramid scheme nor a bubble.

Current asics were sold long before the recent price spike happened.
First asic delivered before price went parabolic.
I'm sure the interactions are more complex than 'price leads hashrate'.

You can also view it like that, then that's the reason bitcoin is not a pyramid scheme, since the interest for money never dies

Hashrate has an effect on price, no doubt about it.  

During the time between October 2010 and March 2011 Hashrate drove up price. (Contrary to Frozenlock's observation) The first publically available GPU mining software came online during this time Difficulty increasing changed the CPU miner's landscape, demand to buy XBT drove price to $1USD. ( a growth stage)  

Price fell as mining supply and demand equilibrium was reached. (a development stage)

Once GPU mining was established,  Price Drove up Hashrate.  May and June of 2011.(consistent with Frozenlock's observation.)  (a growth stage, followed by a falling price and a development stage)

November, December 2012 ASICs were in-house being tested, Hashrate increasing - no notable effect on price. (Consistent with Frozenlock's observation.)  (still part of the development stage)

Mid  January, Avalon ships first ASIC, Prise starts accelerating. Anticipation of increased Hashrate, drove up demand and Price (contrary to Frozenlock's observation) but consistent with March 2011 Hasrate increasing driving price. (a growth stage)

We are in a consolidation stage mining supply and demand equilibrium is being reached, (development stage) once an equilibrium is reached,  (the manipulators "conspiracy"  will pump up the next run when the price of ASIC's has fallen and a market balance between mining and XBT profitability is reached. )

Also relevant is XBT 7200 per day were being created relative to trade volume.  And now we have XBT 3600 being created and relatively higher trade volume, but another order of magnitude on Hashrate. (Result is new dynamic in distribution - arguably "stronger hands", or risk takers who believers in Bitcoin as opposed to profiteers utilising existing hardware)

Once the full magnitude of halving and supply and mining production is in balance sometime after all ASIC's Avalon, Butterfly Labs have shipped and ironed out bugs, we'll see another growth stage.  

I fail to see that, could you provide some data?
A temporary surge in the hashrate, reflected later on the price would be interesting to see.


If I understand your argument correctly, you are saying that an anticipated hashrate increase would drive up the price.
Thus an increase in hashrate wouldn't be caused by a preceding surge in price; the increase in hashrate would have been anticipated by the market and the price adjusted accordingly.
Is that it?  ???


Except that the supply is mostly constant. The network adjusts itself to a hashrate increase.

FPGAs were too complicated for the average miner and even though I had the expertise to operate them, I knew ASICs weren't far behind, so I did not invest in FPGAs.  I'm buying ASICs now.  I would attribute the widening gap between difficulty and price before the parabolic move to the small amount of FPGAs that were brought online.

http://www.jdidesign.com/files-in/Hash-Price.jpg
When GPU Mining started, in July of 2010 you see a huge increase in hash, (green) as the market reacts you see an increase in price, taking Bitcoin form $0.06 to $0.38 in October.  Then as the technology is adopted, you see another order of magnitude of growth and the price goes to $1.00.    

In the image the yellow bands are what I refer to as Bitcoin growth stages, the result is likely Price driving Hashrate. But it seem to be triggered by an increase in Hashrate:
1) When we moved from CPU to GPU we had a Hasrate jump up an order of magnitude higher, followed by an order of magnitude jump in price.
2) the Hash rate also seemed to increase slightly prior to the June 2011 boom.  
3) GPU + FPGA the December 2011 growth stage was somewhat driven by news of FPGA's  (FPGA's didn't have much of a following as anticipation of ASIC's was high and they offered little benefit over GPU's)
4) in Jan of 2013,the market reacted with demand, just after the inevitable increase in Hasrate and the start of the cretin demise of GPU mining.

Let me put it this way, every miner knew the Hashrate would go up and GPU mining wouldn't be viable, it wasn't going to happen until it happened.  Only after Avalon shipped was the increase in Hashrate inevitable. The resulting Hashrate increase we have now is a result of the previous development stage (the shipping of ASICs ordered well before the 2013 increase in price.) This Hash increase is mainly driven by the previous development stage.  The price increase was catalysed by the fact that cheep coins were no longer going to be available.

This whole thing is dynamic you are partially correct there is feedback. Although the same thing also happened before the June2011 bubble, Difficulty increased faster than Price, but for simplicity I put my attention to the overwhelming trend that Hasrate increase were driven by an increase in price.

What adjusts is how the supply is distributed and who gets the XBT has an impact on price.  

Frozenlock:

When there is an increasing hashrate there is an increased rate of bitcoin production (since there is a lag between hashrate and difficulty adjustment). Could this result in a small "wealth effect" resulting in miners investing more of their fiat in bitcoin driving the price higher?

I honestly think this is a more complicated issue than many people think it is.

For example, Bitcoin is subject to the "Tinkerbell Effect", where the more people believe bitcoins have value... the more bitcoins have value. Thus, if a miner works really hard to get a bitcoin and so believes that it must be expensive... to some extent the miner's belief is making it true! (of course, no individual can make a substantial effect, but in aggregate this can be dramatic)

Exactly, the value of money to its root, is wide acceptance and trust. For fiat money, these are ensured by the government and central bank, for bitcoin it is ensured by mathematics and network