:

[chodpaba]

:

[gusti]

Interesting if both are correlated in any way.
Can you pls explain what the grey and yellow dots mean ?

[Raulo]

You can also look at the problem as mining income per, say, 100Mhash/s of hashing power:
http://bitcoin.atspace.com/income.html

Of course the mining income drives difficulty. The fast rise in difficulty in late February and March was due to rise in profitability and many people jumping (or expanding) the mining bandwagon.

Since the currently mining income per hash is lowest ever, we should see significant slowing in difficulty rises (if the BTC price stays the same), except maybe for the current round which might result is slight lowering of the difficulty. In theory, the mining income should converge towards the cost of electricity plus depreciation of the capital and with for the best GPU cards and we are still above this level.

[Hal]

Very nice analysis! Since the price has been at or near 1 for a month now, this would predict that difficulty should now level off for at least a month. Should be interesting to see!

[slush]

This makes sense to me - when price is going up, many people is interested in buying new cards. I noticed this on IRC weeks ago; once price is going up, much more people is solving issues with their new cards .

[TiagoTiago]

Perhaps you could try inputting the data into Eureqa and see if it finds any usefull patterns?

[marcus_of_augustus]

... and of course once a majority of market participants become aware of this correlation, then the feedback of price expectations based on knowledge of the difficulty trend becomes inseparable from the open-loop correlation signal itself ...

[S3052]

As with any of these oversimplistic and lopsided correlations, this correlation will fail to make sense at some point. Those who believe too long in a certain correlation will be surprised if it doesn't exist any more.



See USD / stock market correlation, see Oil / stock market correlation, etc. Over some months, or years there is a positive correlation, over another period the correlation is non-existent, and over another period it is negative.

Be careful with this.

[S3052]

Disagree in terms of who is leading what:
BTC vs. other fiat currencies (largely BTC/USD, BTC/EUR) are LEADING indicators, the others are following.

[S3052]

All righty. It's clear.

Let's continue to watch the exciting developments.

[Raulo]

That is what I am saying. "Price", that is $/BTC, leads difficulty. It is tightly coupled to operation cost.

I agree.

A little further analysis shows that we are already there. When you total costs for power, hosting, depreciation, etc. you will find that we have substantially, already reached equilibrium. Any changes to this will be more or less incremental. The ratios will be very solid until we see a quantum leap in efficiencies.

I disagree. You analyzed only a short period of time, a time far from equilibrium. Long-term price/difficulty ratios are trending down:
http://bitcoin.atspace.com/income.html

and there is still a lot of going on and the equilibrium is not yet achieved.

The trend was down due to increases of efficiency in CPU mining and introducing of GPU mining. But until recently, mining was oligopolistic. Only tech-savvy people were able to mine with GPUs but it has become more "democratic". Pools which are a recent invention also changed the dynamic quite a bit. These processes are not over. Moreover, mining is still too profitable for anybody to drop out. Mining income for ATI cards is still a few times the electricity prices. Including equipment depreciation does not change it. And there is still a possible a dynamic of concentration of mining in countries with cheap electricity. $/BTC/difficulty ratios that are unprofitable for countries with expensive electricity will still be fairly profitable for those with cheap electricity. 

And there is still possibility of botnets or others with "free" electricity and the recent "mystery miner" that changed current difficulty substantially might have been such an example.

I think the difficulty/BTC price ratio will be still trending up (except for the current round where the "mystery miner" artificially pumped the difficulty). Much more slowly than during recent mining gold rush but still respectable 10%+ per 2016 blocks until probably around 150,000-200,000, where it will slow further.

[Raulo]

We are not really in disagreement except, perhaps, where it comes to quantifying the level of equilibration.

Well, yes.

I doubt that many miners have actually worked out a long-term business plan, there are a lot of costs that the system as a whole has already accounted for that I don't think many individuals have really figured out. The costs for 'hosting' and 'depreciation' are not insignificant.

I'm not claiming it is insignificant but the profitability is still much larger. And you are neglecting quite a lot of market with only electricity costs (and there are also miners with no electricity costs). Take compute4cash participants. They bought equipment for gaming and are just using it for mining when idle. There is no capital cost involved.

Part of the reason for this is Moore's law. In order to stay on the edge of profitability mining operations will have to continually invest in new equipment, but

You don't have to invest in new equipment. As long as BTCs that trickle are worth more than electricity and the expected payoff is larger than the resale value of the equipment, the miners will stay online. And this is sufficient for keeping the difficulty high.