I'm talking IV, NOT, BV, how accretive to IV would a huge buyback be

That's what I addressed.

If you assume that current fair value is (for example) $722k per share (1.6 times known book at the moment), then $100bn of buybacks at 1.3 times book would increase true value per share by 15.9%.
If you think the fair value per share is currently higher than $722k then the gain would be a little bit more, and vice versa.

And probably decrease is by some unknowable amount because of the loss of optionality. There are certain things that are possible with that much ready cash that aren't possible without it.

It's rather moot as there aren't $100bn of shares available for purchase at that price. And Mr Buffett isn't going to reduce the benefit to the foundations for the sake of giving more money to existing shareholders. But it is a fine demonstration that buybacks don't increase value per share very much unless they are both (a) huge and (b) done at a material discount to fair value. They do work, but only at the margin.

Jim