Here's a proper look at how timings and memory speed improves Ryzen performance, unlike AT's take.
He's using a Vega, comparing 2666C16 auto timings, The Stilt's timings for 3200C14 and 3466C14, and 3600C16 with auto subtimings. All B-die.
Unsurprisingly, auto subtimings are a disaster with 2666C16 and 3600C16 performing mostly similar in these five games, and true performance is hidden in tight subtimings.
If results like these had been posted in that article, it'd have been an eye opener for many people. Hopefully future Ryzen iterations are less reliant on such fast memory to perform as well as shown here.