I have run some Excel spreadsheet simulations on ROI and want to warn people that buying hardware just to mine BitCoins won't pay off.
I just bought a dual 6970 system, but I bought it before I did the simulations. I based my purchase on one of the bitcoin calculators. The problem is that the network is growing at 4% per day in computing horsepower. That means that your return is losing 4% per day because you essentialy have to compete with everyone else for a fixed rate of bitcoin production per day. It isn't exactly like that on a day-to-day basis, because the difficulty doesn't adjust continuously. But the difficulty function adjusts often enough to almost exactly approximate you having to compete directly against everyone else.
When you take the 4% growth into consideration, your break-even goes from 2 months to never. In my case, I spent $1,000 to produce about 700 MH/S. Everything else remaining the same, my total cumulative bitcoin production will amount to around $200 before it reaches the cost of electricity in about 60 days. At that point I will just have to shut it off and sell the parts to my gamer friends for half off.
The only way mining makes sense is if you already have a gaming system and mine in your off times. But you better do it while it is still worth more than the electricity. As I said before, that breakeven point will be reached in about 64 days from now.
Some people might say that the exchange rate of bitcoins may go up to compensate. That may or may not be true. But if it is, it would make more economic sense to buy bitcoins now and take the gain, rather than buying hardware to generate them.
I wish that you could paste images directly into the messages, because I have a chart that shows the return falling off a cliff right after purchasing your equipment.
There is a graph of the kind of computing horsepower you have to compete with here:
http://bitcoin.sipa.be/Anyway, I want to warn people not to make the same mistake I made. The only way to beat the system now is to come out with a custom ASIC for cheap enough per chip that you could dominate the entire network and grow your hardware fast enough to knock out any competitor before they could produce their own ASIC and make a profit.
The frustrating thing about this is that I first read about BitCoin around April of last year (tax season) when I was researching anonymous banking. I was going to invest $1,000 just as a pure speculation. I never got around to it though. If I had followed through, I would have around $7 million now.