Placing SSDs in slots 'B' and 'C' will slow PCIe x16 to x8—should this affect your setup?
Placing SSDs in slots 'B' and 'C' will slow PCIe x16 to x8—should this affect your setup?
Hey everyone, planning a new PC for the upcoming X3D chips and RTX 50xx releases. I'm checking the motherboard docs and noticed that placing M.2 drives in slots B and C (using CPU PCIe lanes) will limit the x16 slot to x8 speeds. My board is an X870, and I'm considering models like the 9950X3D, 9900X3D or 9800X3D. Since PCIe 5.0 is standard now, I want to keep the same layout. With three M.2 drives on my Z490 (currently PCIe 3.0 with an RTX 3090 at x16), I'm curious if running M.2 in B/C will slow down a future RTX 5080 or 5090 when they hit the market. Should I avoid using that x8 speed, or is it fine? Thanks, Rhys.
It's quite fascinating, though. PCIe gen 5 really pushes performance boundaries. Gen 4 is already impressive, so PCIe gen 5 takes it even further. As long as you're not experimenting with unusual setups, everything should work smoothly. If you're avoiding the topic, consider swapping to a bigger drive or using a riser card for better access.
Essentially, running a Gen 4 40 series card in an x8 Gen5 slot (at Gen4 speeds) still offers similar performance to Gen 3 in an x16 setup, which is why my current configuration with the RTX 3090 works fine. Not a huge difference, I think!
The most demanding part of my work is flight simulation. It can get quite heavy when entering specific airports with high-definition scenery, though it usually stays within a single load rather than a constant strain on VRAM.
I understand that VRAM can be tricky, but I think I heard someone mention that MSFS really needs a lot of it because it pulls textures from Bing maps, downloads them online and renders everything at once. I’m not very familiar with the details of VRAM, but I know that having more memory is usually better for performance.