AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
M Two Accelerator
Graphic was a big deal. We have to expand the memory ban with because the gpu lives on that memory ban with. There's no point putting more compute in if you don't follow it up with more memory banwith. And so that combination of the two is probably the biggest target that we set. Many of those transistors ended up there. They ended up in extra memory system, and they ended up in some very targeted performance enhancements for the cpu and the neral engine and other places.