Guys... whether or not it is "profitable" to generate bitcoins in my ridiculous manner mentioned is beside the point.
If and when a large government entity wishes to destroy bitcoin. Is this one of the ways they could go about it?
The idea was... that they would generate 2016 blocks within a matter of nanoseconds, and then disconnect from the network entirely. Thus setting the generation difficulty bar so high that they "ruin" basically everything.
Even transactions would cease to occur would they not?
A more interesting question is what an entity that controls 1%, 5%, 10% of the processing power devoted to mining could do with such an on - off - on cycle of generating coins. Here's what I think happens:
cycle 1: The entity turns his miners on generating coins at a rate that pushes the difficulty up for the next cycle.
cycle 2: The entity turns his miners off. The difficulty is scheduled to go down next cycle.
cycle 3: The entity turns his miners on again generating coins at a high rate pushing the difficulty up for the next cycle.
cycle 4: Other miners start noticing the pattern and join the on - off - on cycle since it makes sense to only spend electricity during the easy cycles.
cycle n: It becomes obvious to all miners that it is counterproductive to generate coins during the hard cycles and they therefore wait for the easy cycles.
This is a self reinforcing pattern and is unstable as the number of miners remaining in the hard cycles goes to zero the difficulty will flip flop by a factor of 4 (or whatever is the maximum allowed ratio) between the easy and hard cycles. Also the easy cycles will draw more and more newcomers.
More alarmingly, since the difficulty will be slamming against the artificial bounds (the max by 4 factor) it no longer reflects the true difficulty (since it is no longer calculated from the equation that controls block production rate to near one per 10 minutes).
I will give a made up example to show the dynamic: Assume you want to generate 2016 coins per 20160 minute cycle and the available processing power can generate 1000 coins per minute when the difficulty is 1. We set the difficulty to 10000 to get a stable 1000/10000 = 0.1 coins per minute or 2016 coins per cycle.
cycle 1: Entity turns on and adds 10% capacity. Coins produced = 1100/10000 = .11 coins/min (time to produce 2016 = 18327 min) -> difficulty is set to 11000
cycle 2: Entity turns off. Coins produced = 1000/11000 = .091 coins/min (time to produce 2016 = 22176 min) -> difficulty is set to 10000 again
cycle 3: Entity turns on. Coins produced = 1100/10000 = .11 coins/min (time to produce 2016 = 18327 min) -> difficulty is set to 11000
cycle 4: 10% of the miners notice the pattern and stop mining the difficult cycle. Coins produced = 900/11000 = .082 coins/min (time to produce 2016 = 24640 min) -> difficulty set to 9000
cycle 5: Production = 1100/9000 = .122 coins/min (time to produce 2016 = 16495 min)-> difficulty back to 11000
cycle 6: Now 30% of miners catch on. Production = 700/11000 = .064 coins/min (time to produce 2016 = 31680 min) -> difficulty set to 7000
cycle 7: Now new miners want in on the easy cycle. Base capacity up another 10%. Production = 1200/7000 = .171 coins/min (time to produce 2016 = 11760 min) -> difficulty now to 12000
cycle 8: 90% of original miners are in on the game. Production = 100/12000 = .008 coins/min (time to produce 2016 = 241920 min) -> difficulty should be 1000 but slams into boundary at 12000/4 = 3000
cycle 9: The game has attracted 50% more miners. Production = 1600/3000 = .533 coins/min (time to produce 2016 = 3780 min) -> difficulty should go to 16000 but stops at 12000 (at the 4x3000 limit)
cycle 10: Production near zero and difficulty down to 3000 again
Difficulty will slam between 16000 and 3000 and most coins will be produced during the 3 day easy cycle with long 24 week periods with high difficulty and few miners.
This kind of instability is inherent in systems that have delay such as happens with the calculation of a new "difficulty" after a delay.
I bet someone controlling even 1% of capacity could, through the amplification of other miners joining him once they notice, cause hard/easy cycles where production in the easy cycle is more than twice desired production.
