Shortly after Microsoft’s tell-all reveal of the Xbox Series X’s specifications, Sony followed up with our first detailed look at the Playstation 5. While we’re still not sure whether the console will look like a trashcan or a banana, we now have a better picture about its capabilities, both in comparison to existing PC hardware and to the Xbox Series X.
The Playstation 5 is set to feature a Zen 2-based processor with 8 cores, sixteen threads, and a peak clock frequency of 3.5 GHz. Unlike the Xbox Series X (and consoles in general), the Playstation 5 features a variable clockspeed that’s tied to system power draw. This means that variations in clockspeed will be identical across different PS5 units, and will not depend on external thermal conditions.
The PS5’s GPU design is unexpectedly narrow and fast. The 36 CU RDNA2 part looks very similar to the RX 5700, with the exception of added ray-tracing hardware and the fact that its running at a remarkable 2.23 GHz at peak. This translates to 10.28 TFLOPs of raw compute, which is just 15 percent less than the Xbox Series X. The real world performance deficit might be a bit wider, however. Tests have indicated that Navi doesn’t scale that well with clockspeeds. Bumping the clocks up to 2.1 GHz on the RX 5700–an 18 percent boost over stock, yields just 5-10 percent higher performance.
In terms of memory, Sony outfitted the Playstation 5 with 16 GB of GDDR6 memory, across a 256-bit bus, delivering 448 GB/S of bandwidth. This is, again, right in line with first-gen Navi and a bit slower than the Xbox Series X.
Compensating for this is Sony’s remarkable custom SSD design. The 825 GB unit has an effective transfer rate of over 8 GB/S when factoring in compression, making it the single fastest storage device on the market. We expect this to have a major impact on world streaming, loading, and other storage and I/O intensive workloads.