I see a lot of people are under the extremely erroneous impression that mining is meant only to produce new coins.
Mining is bitcoin's lifeblood. It's vital for bitcoin's security. It verifies transactions. It forwards blocks. It advances the entire blockchain. No mining, no bitcoin. So make no mistake, mining will still continue, even after the bitcoin creation is finished. Miners will gain more money from transaction fees once income from bitcoins starts to deplete. This, of course, presumes Bitcoin will be used more and more for transactions, which I think is the case, judging by the constant growth in popularity.
Also, the concentration of hashing power in only a few hands is quite unlikely, since bitcoin has a nice natural defense mechanism against that. The more centralized it becomes, the less people trust it. The less people trust it, the lower the value. And since controlling a significant chunk of the network requires a significant investment, you'll be shooting yourself in the foot by doing it, since as soon as people get a whiff of what's going on the bitcoin value will go down like a Singapore hooker and by extension, ruin your ROI.
Yes, mining will NOT be as profitable as it was some time ago, and many people will quit because of this, but make no mistake, the moment mining stops for good bitcoin will be gone.
Topic: calculations (Read 1437 times)
Difficulty has nothing to do with mining odds. Difficulty acts to slow down the global network, it does nothing to change the distribution of block finding.
The odds of you finding the next block first depend only on the fraction of global hashing power that you possess.
However, the latest few difficulty increases imply that the network has seen and expects to continue seeing roughly 3.5% daily increases in hashing power. Now obviously we can't sustain 3.5% growth forever. The bad news is that it really doesn't have to continue for very long at all to invalidate most other projections.
Note that Fergalish did his math at the dawn of a new era, when CPUs gave way to GPUs with hundreds of times more power. I see no evidence of such a shift happening now. People are simply brining new GPUs online at a rate of around 3.5% per day.
The odds of you finding the next block first depend only on the fraction of global hashing power that you possess.
However, the latest few difficulty increases imply that the network has seen and expects to continue seeing roughly 3.5% daily increases in hashing power. Now obviously we can't sustain 3.5% growth forever. The bad news is that it really doesn't have to continue for very long at all to invalidate most other projections.
Note that Fergalish did his math at the dawn of a new era, when CPUs gave way to GPUs with hundreds of times more power. I see no evidence of such a shift happening now. People are simply brining new GPUs online at a rate of around 3.5% per day.
one thing for sure is that the mining days are limited one way or another. Diffuculty rises are only half of the story as well.
Yes, soon no one is going to be mining, because difficulty levels are going to be so high, so why bother starting now, eh?
one thing for sure is that the mining days are limited one way or another. Diffuculty rises are only half of the story as well.
Six months ago, fergalish did the same calculation for CPU mining. Next year, we'll probably read the same news for FPGA mining.
In this case, I think you have over-estimated the rate of difficulty growth. The last couple of difficulty rises have been particularly big because of the sharp price growth during April/May. There seems to be a lag of about six weeks between price rises and difficulty rises, so I expect the difficulty will increase a bit more slowly from next month.
In this case, I think you have over-estimated the rate of difficulty growth. The last couple of difficulty rises have been particularly big because of the sharp price growth during April/May. There seems to be a lag of about six weeks between price rises and difficulty rises, so I expect the difficulty will increase a bit more slowly from next month.
I did similar calculations..
I'm also pessimistic that this will be left on the big miners hands and it will no longer be profitable..
Next jump will be around what..? 340000/350000..?
I'm also pessimistic that this will be left on the big miners hands and it will no longer be profitable..
Next jump will be around what..? 340000/350000..?
Current global hashing power: 1.97E+12 Hashes / Second (source)
That puts the odds of any given 1 Mhash/sec winning a block at 0.0000506%. Over the course of "today" (the next 144 blocks) it has a 0.00729% of finding a block.
The current growth in difficulty suggests an expected growth in global hashing power at 3.5% per day. That means that tomorrow, the chance per block will be 0.0000489%, and overall 0.00705%.
Assuming this rate of growth, the limit (sum of the series over an infinite lifetime) is 0.215% chance of finding a block.
So, a GPU, or whatever, with 300 Mhash/sec of power has about a 65% chance of finding a block between now and the end of time. The expected payout is thus 32.34 BTC, or about $226 at $7/BTC.
Thus, with my pessimistic assumptions, you should not spend more than 75 cents per Mhash/sec, which is still possible, but getting harder.
Note also that I'm assuming zero operating expenses (free power and cooling).
That puts the odds of any given 1 Mhash/sec winning a block at 0.0000506%. Over the course of "today" (the next 144 blocks) it has a 0.00729% of finding a block.
The current growth in difficulty suggests an expected growth in global hashing power at 3.5% per day. That means that tomorrow, the chance per block will be 0.0000489%, and overall 0.00705%.
Assuming this rate of growth, the limit (sum of the series over an infinite lifetime) is 0.215% chance of finding a block.
So, a GPU, or whatever, with 300 Mhash/sec of power has about a 65% chance of finding a block between now and the end of time. The expected payout is thus 32.34 BTC, or about $226 at $7/BTC.
Thus, with my pessimistic assumptions, you should not spend more than 75 cents per Mhash/sec, which is still possible, but getting harder.
Note also that I'm assuming zero operating expenses (free power and cooling).
Jump to: