Re: you can see a 20% performance boost for having the memory controller on-chip on very many crucial applications. So in other words, you are wrong about it being 30%. Thanks for the confirmation.