Bitcoin Forum

I don't know if im alone in this, but how do you read those?

I don't know either. Someone on IRC asked for such graphs :)

The higher the value of bitcoins the more lucrative generation becomes. It makes sense that the value of bitcoin drives the difficulty.

This will be the case as long as there is a 50BTC reward for generating a block. When it moves to transaction fees, it's a different story.

Ah I wanted to see this chart.
But...
What are those data points? There should be a ton after the price spike to .5 that are much lower. That curve doesn't seem to fit the actual data?
Was the price just sampled right when the difficulty changed? Could you sample more frequently. I think the graph will look pretty different.

More datapoints (intervals are 86400 seconds)
http://bitcoincharts.com/media/stuff/price_vs_diff.png

I'm sure that I don't really need to remind the members of this thread, but to anyone who doesn't already know...

Corrolation is not causation.

In all likelyhood, both these variables are responding to the same unknown.  That unknown is likely the number of people who are interested in bitcoins at any particular point alont the timeframe noted.  But that is just an educated guess, since there is no really proxy for the number of users in the network.

tcatm: could you please post a csv or tab-delimited data file with the data? :)

Yeah, it does feel to me like price drives difficulty. But I think "demand" drives both really.

It seems to me more interesting price-time and difficulty-time on one chart

One reason price might follow difficulty is that mining should not be too profitable (because nothing should be too profitable, the world doesn't leave free money lying around). Therefore the price of Bitcoins can't rise too much above the cost of mining (counting equipment depreciation among the costs of course). The cost of mining is proportional to the difficulty (approximately). Therefore we might expect to see price proportional to difficulty.

We do see a nearly proportional relationship in the 1st graph, but that data set was incomplete. I'd like to see that last graph redone with a linear difficulty scale so we could see how the proportionality holds up with more data.

Your latest chart is price vs log(difficulty).  Can you show price vs difficulty?

Exactly right about no free money lying around. But that doesn't mean one follows the other, only that they will maintain a close relationship.

Those are interesting charts. I think the two dominant effects are the influx from the Slashdot article (when the difficulty crossed 100), and the rise of GPU mining (roughly, difficulty of 1000 and later).

Yeah, that (log) difficulty one wasn't particulary useful.
Price vs difficulty is another story.

+1