65 FPS at 1440p across 9 games (anchored estimate) · solid 60+ FPS 1440p gaming, with headroom in lighter titles
65 Avg FPS Includes estimates
Treat these as estimates. They are interpolated from our own measured results on comparable silicon, and they carry the estimate label until this card is benched. Across our 9-game suite at 1440p, NVIDIA GeForce RTX 2070 averages 65 FPS. The ceiling is Counter-Strike 2 at 108 FPS; the floor is Alan Wake 2 at 41 FPS, a 2.6x spread that tells you how engine-dependent this class of card is. In practical terms that means solid 60+ FPS 1440p gaming, with headroom in lighter titles. Under load we logged a peak of 168W against the 175W rating.
| Benchmark | Result | 1% Low | VRAM Used | Power | Data |
|---|---|---|---|---|---|
| Cyberpunk 2077 1440p · Ultra | 48 FPS | 38 | 6.75 GB | 166 W | Estimated |
| Red Dead Redemption 2 1440p · Favor Quality | 55 FPS | 43 | 6.57 GB | 163 W | Estimated |
| Alan Wake 2 1440p · High | 41 FPS | 34 | 7.36 GB | 161 W | Estimated |
| Baldur's Gate 3 1440p · Ultra | 79 FPS | 65 | 7 GB | 168 W | Estimated |
| Black Desert Online 1440p · Remastered | 82 FPS | 52 | 5.29 GB | 166 W | Estimated |
| Counter-Strike 2 1440p · Very High | 108 FPS | 58 | 7.13 GB | 140 W | Estimated |
| Diablo 4 1440p · Ultra | 68 FPS | 52 | 7.36 GB | 164 W | Estimated |
| Escape from Tarkov 1440p · Ultra | 58 FPS | 43 | 7.36 GB | 150 W | Estimated |
| The Witcher 3 1440p · Ultra | 45 FPS | 38 | 7.05 GB | 166 W | Estimated |
| Architecture | Turing (TU106) |
| CUDA cores | 2,304 |
| VRAM | 8GB GDDR6 |
| Memory bus | 256-bit |
| Memory bandwidth | 448 GB/s |
| Boost clock | 1,620 MHz |
| TDP | 175 W |
| Process | 12nm FFN |
| Interface | PCIe 3.0 x16 |
| Release date | 2018-10-17 |
| Launch MSRP | $499 |
NVIDIA GeForce RTX 2070 averages 65 FPS at 1440p on our suite. Treat these as estimates. They are interpolated from our own measured results on comparable silicon, and they carry the estimate label until this card is benched.
100% = this card, Gaming headline metric (Avg FPS). #49 of 82 desktop cards in this vertical.
| GPU | Relative | % | Avg FPS |
|---|---|---|---|
| NVIDIA GeForce RTX 3060 Ti | 108% | 70 | |
| NVIDIA GeForce RTX 5060 | 106% | 69 | |
| NVIDIA GeForce RTX 3060 | 103% | 67 | |
| AMD Radeon RX 7600 XT | 102% | 66 | |
| NVIDIA GeForce RTX 2070 | 100% | 65 | |
| Intel Arc A770 Limited Edition | 100% | 65 | |
| AMD Radeon RX 6700 | 97% | 63 | |
| AMD Radeon RX 7600 | 97% | 63 | |
| NVIDIA GeForce RTX 5050 | 97% | 63 |
Same card, other workloads: NVIDIA GeForce RTX 2070 AI benchmarks, LLM tokens/sec, image generation & VRAM fit
| Transistors | 10,800 million |
| Die size | 445 mm² |
| Fabricated by | TSMC |
| Transistor density | 24.3 million per mm² |
Denser than 21% of the 76 cards we have silicon data for. Density is the clearest measure of what a process node bought: a card that gained it without growing the die got its speed from the fab rather than the architecture.
Silicon figures from Wikipedia (CC BY-SA 4.0). Benchmarks on this page are our own. Compare every chip.
Whole-job timings, composed from our measured per-model results on this card.
Can't run: 60-second AI short film (needs Qwen3 32B), 60-second AI short film, narrated (needs Qwen3 32B), 10 short social clips (needs Qwen3 32B), 40-product photo shoot (needs FLUX.1 Kontext dev), 6-panel comic page (needs Qwen3 32B), 20 long-form articles (needs Llama 3.3 70B), Full codebase review (needs Qwen2.5-Coder 14B), Character sheet, 12 poses (needs FLUX.1 dev), 100-photo restoration batch (needs FLUX.1 Kontext dev), 24-frame storyboard (needs Qwen3 32B), 100-photo restore and enlarge (needs FLUX.1 Kontext dev).