Hey gbianchi - I haven't read through the entire thread but earlier in the thread it was pointed out that the model would eventually predict an infinite value for bitcoins. What about adding a secondary function into the original that represents price influences other than addresses so that the final equation is a ratio of the two added together. Basically I'm saying that for now the addresses thing you've figured out might be the largest controlling factor but as we approach using up the addresses that the influence of the secondary equation becomes the stronger of the two and eventually becomes the equation.

Also I was curious to see if the dates where the price spikes and then levels out over the long run such as April 13' and then again at December 13' are related to asics. I know the second spike I mentioned was around the time that the silk road shut down and I took that to be increased popularity at the time because of the negative association with the silk road ending but still wonder if the spikes relate to 1st gen vs 2nd gen asics. Anyway, sorry if this has been discussed already!