In a tear-down of the Xbox Series X expansion card, it was discovered that Microsoft’s next-gen console uses Hynix’s 128-layer QLC NAND flash technology (referred to as 4D NAND). While QLC based SSDs are cheaper and denser than traditional TLC solutions, they’re also slower and less durable. You can read more about that here.
Update: As far as the NAND storage is concerned, it’s either one of these two:
- SK Hynix Ships Samples of High-Performance 96-Layer 1Tb QLC 4D NAND: 96-layer QLC NAND.
- SK hynix Starts Mass-Producing World’s First 128-Layer 4D NAND: 128-layer TLC NAND.
- Considering pricing and availability, it’s likely the latter.
This explains why storage on the Series X is markedly slower than that used on the Sony PS5 (2.4GB/s vs 5.5GB/s raw). The latter is most likely using TLC based flash storage which is faster but also a notch more expensive.
Moving onto the memory controller, we’ve got a Phison E19T controller which is a midrange part, supporting up to 2TB of storage which is likely to be the upper limit for the expanded storage on the Xbox Series X. In line with the 2.4GB/s raw data speed supported by the Series X, the Phison E16T tops out at 3.75GB/s.
For Microsoft, the primary advantages are on the software side. With the DirectX 12 API already out on the PC, and DirectStorage being utilized on the RTX 30 series GPUs, we know what it’s capable of.
NVIDIA’s RTX IO which is basically a different name for DirectStorage significantly speeds up NVMe drives on modern PCIe 4 platforms. The low-level access given to developers by DirectStorage allows data to be directly send from the SSD to the video memory, compressing and decompressing it using dedicated hardware in the GPU. Furthermore, DirectStorage also helps send batch I/O requests while earlier every I/O request big or small was individually sent, adding unnecessary load to the processor. RTX I/O or DirectStorage, however, needs to be implemented by the developers for every game engine.
As per NVIDIA’s own tests, RTX I/O cuts down the load times in an NVMe based SSD by nearly a fifth. The test was conducted using a 24-core AMD Threadripper processor using a Gen 4 NVMe drive, and it took all the cores 5 seconds to decompress and load the assets. With RTX IO, the same operation was performed on the GPU in just 1.5 seconds.
Therefore, as you can see, Microsoft can use DS to improve the data compression/decompression speeds and I/O calls to make up for the relatively lower raw drive speed.