Topic: What will happen when the 21M bitcoin will be minned - page 6. (Read 26062 times)

1) I thought you talked about the drop to 25BTC, not 12.5BTC
2) 16 weeks was correct. 2 times "few" months > 2 * 2 months > 2 * 2 * 4 weeks = 16 weeks

A single T/H added to the network at the current moment would decrease time it takes to reach the next difficulty adjustment by more than a full day.  A full week would require a doubling of the network hash rate (~24-26 T/Hash).  A full month would require ~175-225 T/Hash.  This is easily forseeable.  

Let's say that the average miner has an output of about 5 g/hash.  If you take an average network hashrate of 13 t/hash, and divide that by 5 g/hash, you're looking at about 2600 miners.

Now, let's assume the average cost of a 5 g/hash mining setup is $4000.  In terms of ASICs, $4000 will get you approximately 120 g/hash (you can buy 3 of BFL's $1300 40/ghash models).

Now, multiply 120 g/hash (or .12 t/hash) by 2600.  You get 312 T/Hash.  And there you have it, the time it takes to reach the 12.5 BTC is decreased by more than a month.

Then, take into account all the new miners that will enter the picture in the next 4 years.  If the number of miners has reached ~2600 in the past 4 years (actually, it's been less than 4 years, and miners didn't REALLY start to enter the picture until the first quarter of last year), then it's reasonable to assume that the number of miners will at least double, if not triple, quadruple, or even more.

Let's say we have 4 times the number of miners by the year 2016 (2600 x 4 = 10,400), then we're looking at ~1.245 P/H.  Then, also take into account the declining cost of ASIC units over time, and this number could increase even more.

I think it's quite reasonable to assume the jump to the 12.5 BTC reward will occur at least 2 months earlier than anticipated.

Edit:  BTW, I think you meant 8 weeks = 2 months, and not 16 weeks = 2 months.

To make this assumption come true, you would have to significantly increase the computing power over several retargets. My bet is ASICs will not accelerate the whole progress of getting to 25BTC block reward at all as they will not be available before that. Definitively not in an amount to compete with existing hashing power.

To have the jump "a few months sooner", you would have to have a constant increase of hashing power of 100% per retarget for two "few months".
Doubling network speed per retarget/week would take 2 few months to arrive few months early. As two months is the minimum to qualify as "few months" and "two few months" would be more than 16 weeks we would have to see 16 weeks of doubling that would end at about 655PH/s. Hmm … I wanna see that

Stupid roundoff error.

Rjk's got it.  When all the ASIC devices start reaching consumers (whether its from BFL or some other competitor) then the difficulty will rise accordingly.  At worst, the jump from 25 BTC to 12.5 BTC will come a few months sooner than originally expected.  It is likely that when ASICs start reaching consumers, we will see a few weeks where the time it takes the network to solve 2016 blocks (i.e. the # of blocks needed for a difficulty readjustment) will be shortened to a span of a few days rather than an average of 2 weeks.  But, once the majority of ASIC purchasers have received their devices, then we will be back to the 2 week average.

The difficulty will go up to match the new speed and keep block production steady. And there is no need to worry that we will run out of difficulty levels; they can go to a hugely high number to match whatever new technologies can throw at it.

What happens if the BFL asic is real and an extra 10000TB/S is added in minining

Yeah but it will converge near to 0 in 15 years.

You might want to recalculate that.  Technically we will never reach 21 million coins but it will be a lot longer than 15 years before the subsidy goes to zero.

The block reward continually halves; we will asymptotically reach the theoretical limit of 21 million. So yeah, you'll probably be dead...

Cool thanks for answering.

More miners would be good from a security perspective, but the network would still function with few miners.

Is the network will need more and more miners to be supported?

Well, that's about 100 years into the future, so we can't say for sure. However, miners would still be needed to secure the network, as well as to process transactions, so the general conclusion is that they would live on the transaction fees. At this point, the transaction fees aren't much, but when there is no block subsidy people might be more willing to pay a fee in order to have their transactions processed.

Regardless of whether a subsidy is created on each block, the blocks must still be created for the network to function.

the 21 million bitcoin will be mined in 15 years... I will be certainly not dead. Well I hope so!

The world ends. I guarantee you that you will be dead when that happens ...

I will personally buy every accessible Bitcoin, then dominate the world.

If the network need miners to be supported. What will happen when all the 21 million bitcoins will be mined and no ones will be mining and supporting the network anymore?