[AgentME]

>because they stand to make $100,000,000/yearly from their (now not so secret) advantage.

If anyone is confused how the 30% advantage that ASICBOOST gives translates into this huge number, remember that the 30% boost applies to revenue, not profit.

The revenue of mining in the long run approaches the costs to mine (basically the cost of electricity), leaving extremely slim profit margins. Let's say the usual revenue from mining over some period of time is $1.02, and the cost is $1.00, giving a profit of 2 cents every time unit. A miner using ASICBOOST could have a revenue of $1.32 every time unit, giving a profit of 32 cents every time unit, which is at least an order of magnitude more profit than anyone else. This miner then has more resources to spend on buying hardware and scaling up.

(The above paragraph goes for any type of optimization and isn't necessarily nefarious. It can be a problem if one miner effectively keeps an optimization secret for too long, because they'll continue to grow in size and could get more than 50% of mining power, which is a specific point that causes huge problems with Bitcoin. But the real unique issues about ASICBOOST are about how it encourages empty blocks, and how it has been secretly incentivizing some groups to argue against any incompatible protocol change.)

[midmagico]

Your arithmetic is off. Revenue doesn't increase. Costs decrease. Effectively the increase in profit is similar, though, to your result. Just for different reasons.

Ignoring depreciation or hardware costs, the arithmetic is like this:

Pre-ASICBoost: Revenue: $1.02 Costs: $1.00 Profit: $0.02

Post-ASICBoost: Revenue: $1.02 Costs: $0.70 Profit: $0.32

The closer to costs that revenue for non-AB mining is, the closer to infinity the comparative profit margin is.

That is, AB doesn't increase hashrate. (That would be a very, very much more complicated equation since the hashrate reward pie is essentially static.) All it does is decrease the amount of power required to do the same work.