Well when paired with some of the new direct storage API’s this could offer substantial performance increases for gaming especially in scenarios where there is insufficient GPU ram to store all the needed assets on the card, or where an unstated asset is called upon.I’m sure there’s a reason, but what would this accelerate workload wise?
Depending on the changes to AMD’s new memory controller they could see a decent performance increase especially in GPU bound scenarios, specifically focusing on the 1% lows.