[
It requires hardware support, so you can't add it to miners that dont support it in hardware. Also there is a difference between Overt (Dragon) and Covert (bitmain) versions of Asicboost so unless bitmain add overt to their miners it won't help them.
Thing that I'd like some clarity on, ASICboost was only suppost to improve the efficiency of a miner, not its hashing speed, so it did the same but used less power. Overall of course this would lead to either being able to push the miner faster for the same power usage (if the hardware could handle it) or running more individual miners but using the same power. (e.g. if it gave a 25% reduction in power, you could run 5 miners for the power you used to run 4 miners)
Is this new version of overt ASICboost different? There are hints that the Dragonmint miners will only run at 1/4 speed on non-asicboost pools. That sort of doesn't line up with how we all thought ASICboost worked.
https://arxiv.org/pdf/1604.00575.pdf is the whitepaper for this new boost.
Sections 3.4.1-3.4.2 talks about how it is different than the Bitmain version. What Bitmain did was prune the merkle tree of transactions to find a collision. This way doesn't state what it does, only that a different way is to "is to calculate many merkle roots at random and filter them based on their last 4 bytes until sufficiently many are found that collide in their last 4 bytes." and also "Another, more efficient way to produce colliding block headers is by using bits inside Chunk 1 but outside the merkle root that are free for the miner to choose. This does not require finding any merkle root collisions. Instead, one merkle root is chosen and fixed. The free bits are then updated in a loop and the respective Mid states are computed from Chunk 1. Each Mid state obtained in this way gives a new colliding work item, so that this method is extremely efficient."