Nvidia RTX 3080: PCIe 4.0 vs 3.0 scaling analysis

1. Introduction2. Testing Methodology3. 3DMark Synthetic4. Control5. Death Stranding6. The Division 27. Far Cry New Dawn8. Gears 59. Ghost Recon: Breakpoint10. Horizon Zero Dawn11. Metro Exodus12. Middle Earth: Shadow of War13. Red Dead Redemption 214. Shadow of the Tomb Raider15. Total War Saga: Troy16. RTX On - Control17. RTX On - Metro Exodus18. RTX On - Shadow of the Tomb Raider19. Performance Summary (15 Game Average)20. Closing Thoughts21. View All Pages

Here we present frame rate figures for each test system used today, averaged across all 12 (plus 3 re-tested with RTX on) games on test today. These figures can disguise significant variations in performance from game to game, but provide a useful overview of the sort of performance difference we have seen.

There's a couple of key points to make here. First of all, if you've gone through our game-by-game breakdown, you will not be surprised to see PCIe 4.0 makes very little overall difference compared to PCIe 3.0. Almost nothing, in fact – at 1440p, the average difference between the 3900XT with PCIe 4.0 and the same CPU with PCIe 3.0, is a mere 2%, or 2FPS. At 4K, the difference is reduced to exactly nothing.

The other point to bring up is CPU bottlenecking. At 1080p, the 10900K is simply a faster gaming CPU than the 3900XT, performing 10% better on average despite its PCIe 3.0 limitation. Even at 1440p and 4K, it's simply averaging higher frame rates than the 3900XT with PCIe 4.0.

Become a Patron!