NVIDIA GeForce RTX 4080 vs NVIDIA GeForce RTX 5070
Side-by-side benchmark comparison, specs, pricing, and value analysis.
Benchmark Comparison
Gaming figures are average frames per second, measured by Tom's Hardware GPU hierarchy. Geometric mean of 15 games, ultra settings, no upscaling. Ray tracing figures are DXR-enabled at 1440p. A dash means the card was not tested at that resolution. Productivity and Steel Nomad are synthetic scores on their own scales and are not comparable to the fps columns.
Specifications
| Spec | GeForce RTX 4080 | GeForce RTX 5070 |
|---|---|---|
| VRAM | 16GB GDDR6 | 12GB GDDR7 |
| Memory Bandwidth | 717 GB/s | 672 GB/s |
| Shaders | 9728 | 6144 |
| Boost Clock | 2505 MHz | 2512 MHz |
| TDP | 320W | 250W |
| Architecture | Ada Lovelace | Blackwell 2.0 |
| 3DMark Steel Nomad | 6,585 | 5,256 |
| Best Price | $1,199 | $549 |
| Value Score | 33/100 | 69/100 |
Verdict
The NVIDIA GeForce RTX 4080 is faster but costs more. The NVIDIA GeForce RTX 5070 offers better value for money. Your choice depends on budget vs. performance priority.
NVIDIA GeForce RTX 4080 vs NVIDIA GeForce RTX 5070: Which Should You Buy?
The NVIDIA GeForce RTX 4080 is 14% faster at 1440p ultra (130.4 fps against 114 fps). That is roughly one settings notch — enough to matter if you are chasing a specific refresh rate, not enough to rescue a bad price.
You pay a 118% premium for the NVIDIA GeForce RTX 4080 ($1,199 against $549) to get 14% more frames. The extra frames do not keep up with the extra cost — the NVIDIA GeForce RTX 5070 is the better value here.
Two things beyond raw speed: NVIDIA GeForce RTX 4080 carries 16GB against 12GB — the deciding factor at 1440p and above with ray tracing or texture packs, and the reason 12GB cards age badly; and the NVIDIA GeForce RTX 5070 draws 70W less (250W versus 320W), which can mean a smaller PSU and a quieter case; and the NVIDIA GeForce RTX 5070 is 3 years newer (2025 versus 2022), so it gets longer driver support and newer upscaling and frame generation features.