Thank you very much Doog! The python code really cleared it up for me.

I am wondering if someone can help me make sense of the algorithm used to constuct a given roll.
For definiteness, lets' use the following bet: https://just-dice.com/roll/307724627

I initially construct the two strings:

s1 = 1496:'server seed':1496
s2 = 1496:'client seed':1496

The instructions then become ambiguous because it is not clear how these two strings should be concatenated before hashing with the SHA-512 function. I've tried a couple combinations and none give me the desired roll (once I take the first 5 characters from the hashed value and convert them to an integer).

Someone in the chat room suggested:

1496:'server seed'-'client seed':1496

But this didn't work either. Anybody have a clue?

Why don't you transfer your shares to havelock and dump them there?

I propose an estimate for a lower bound to the price of bitcoins. The aim of this is to evaluate the longterm intrinsic value of the cryptocurrency.

Eventually, we'll get to the point where mining for bitcoins won't be a feasible task. The only way for the network to hold its ground, is for miners to keep on mining irregardless of mining difficulty. This can only mean that the transaction fees miners get for doing their job need to, at the very least, balance the cost of running their machines.

At the moment, avoiding all the various transaction fee caveats, transactions earn miners about 0.0001 BTC/kB with your average transaction coming at 0.5kB. VISA is built to handle a peak burts of ~8500 transactions per second (tps) which generally occur during the holiday season. However, a hypothetical future society using bitcoins as a global currency will also trade stocks in bitcoins.

Now, I cannot find a statistic on the total number of stock market transactions but I know that the global markets trade around ~697,000 $/second (roughly 22,000 billion/year in 2013). This, along with the $67 average share price gives an estimate of ~10000 tps (identical to VISA's peak transaction rate)

Taking 30000 tps as a good reference rate for transactions happening in the world every second overall (stock markets+consumer market+an extra 10k to simulate traded physical currency), this would come out to 15 MB/s of transaction size for a total fee generation of 1.5 BTC/s.

Alright, how much does this 1.5 bitcoin need to be worth to even out the cost of running mining equipment?

Assuming the sum costs per mining remain constant (devices get more powerful and energy efficient but we require more of them) in such a way as to saturate the lower bound on the estimate price we want to give bitcoins, we can calculate how much it costs to run today's network. The network is currently mining at ~4000 TH/s. Being generous and granting the whole network a power efficiency of 1 Watt/GH (currently the efficiency advertised for the top of the line KnC Asics), then the total cost of running the network (taking prices of 0.15$/kWh) is of 0.5 $/s!!!!!!!!

This would give a long term lower bound to bitcoins of 0.33 $/BTC.
This is, in my opinion, the true intrinsic value of bitcoins as opposed to the extrinsic estimate we made using the prices of gold as reference.

References:

1) http://www.statista.com/statistics/270127/largest-stock-exchanges-worldwide-by-trading-volume/
2) http://bitcoin.sipa.be/
3) https://www.kncminer.com/categories/miners
4) [Suspicious link removed]j.com/moneybeat/2013/06/04/the-average-stock-price-is-expensive-get-used-to-it/

Zipmaster