NVIDIA GeForce RTX 2070 SUPER vs NVIDIA GeForce RTX 3070 Ti Mobile Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce RTX 3070 Ti Mobile leads with an average benchmark score of 23,518, while the NVIDIA GeForce RTX 2070 SUPER scores 20,282. That is a 16% advantage for the 3070 Ti Mobile in aggregate performance.

Q: What is the largest single-test margin between the two cards?

A: In 3DMark Steel Nomad (DX12), the RTX 3070 Ti Mobile scores 2,572 versus 1,651 for the RTX 2070 SUPER, a 55.8% lead. This is the largest delta in any recorded test.

Q: Does the RTX 2070 SUPER win any benchmark tests?

A: Yes, it wins five tests: DirectX 10, DirectX 9, G2D, G3D, and GPU Compute. Its largest win is DirectX 9, where it scores 223 versus 196, a 12.1% margin.

Q: How do the two cards compare in the 69th and 65th percentile rankings?

A: The RTX 3070 Ti Mobile sits at the 69th percentile among all GPUs, while the RTX 2070 SUPER sits at the 65th percentile. The newer mobile part ranks higher overall, consistent with its higher average score.

Q: What are the nearest rivals for each card according to the database?

A: For the RTX 3070 Ti Mobile, the closest rivals are the RTX 3080 Mobile (average score 23,628, 0.5% higher) and AMD Radeon AI PRO R9700 (23,315, 0.9% lower). For the RTX 2070 SUPER, the closest rivals are the Quadro M4000M (20,480, 1.0% lower) and RTX 3070 Mobile (20,534, 1.2% higher).

Q: Do both cards use the same memory configuration?

A: Yes, both have 8 GB of GDDR6 memory on a 256-bit bus, with 448.0 GB/s bandwidth and 14 Gbps effective memory speed.

Architecture Differences

The two GPUs come from different NVIDIA architectures and process nodes. The RTX 3070 Ti Mobile is built on the Ampere architecture using the GA104 chip, manufactured on Samsung's 8 nm process. It houses 17,400 million transistors on a 392 mm² die, yielding a transistor density of 44.4 million per mm².

The RTX 2070 SUPER uses the Turing architecture with the TU104 chip, fabricated by TSMC on a 12 nm process. It packs 13,600 million transistors across a much larger 545 mm² die, giving a transistor density of only 25.0 million per mm². The density difference is stark: Ampere crams 77.6% more transistors per area, a direct consequence of the newer process.

Shader resources differ dramatically. The RTX 3070 Ti Mobile has 5,888 shading units, 184 texture mapping units (TMUs), and 96 render output units (ROPs). The RTX 2070 SUPER has 2,560 shading units, 160 TMUs, and 64 ROPs. The mobile Ampere part has 130% more shaders and 50% more ROPs, though its TMU count is only 15% higher.

Ray tracing and tensor core counts also diverge. The RTX 3070 Ti Mobile carries 46 RT cores and 184 tensor cores. The RTX 2070 SUPER has 40 RT cores but 320 tensor cores. The older card has 73.9% more tensor cores, but the newer one has 15% more RT cores.

The compute throughput numbers reflect these differences. The RTX 3070 Ti Mobile delivers 16.60 TFLOPS of FP32 and 16.60 TFLOPS of FP16 (1:1 ratio). The RTX 2070 SUPER delivers 9.062 TFLOPS FP32 and 18.12 TFLOPS FP16 (2:1 ratio). In raw FP32, Ampere is 83.2% faster; in FP16, Turing is 9.2% faster.

Other architectural differences include pixel and texture rates. The RTX 3070 Ti Mobile reaches 135.4 GPixel/s and 259.4 GTexel/s. The RTX 2070 SUPER reaches 113.3 GPixel/s and 283.2 GTexel/s. The mobile part has a 19.5% higher pixel rate, but the desktop card has a 9.2% higher texture rate.

The bus interface differs as well: PCIe 4.0 x16 on the RTX 3070 Ti Mobile versus PCIe 3.0 x16 on the RTX 2070 SUPER. The RTX 2070 SUPER is a dual-slot desktop card with 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C outputs, while the RTX 3070 Ti Mobile has portable device dependent outputs.

The Verdict

The data points to a clear but nuanced choice. The RTX 3070 Ti Mobile wins the overall average score by 15.9% (23,518 versus 20,282) and ranks higher at the 69th percentile versus the 65th. It dominates in modern workloads: 55.8% ahead in 3DMark Steel Nomad, 27.2% ahead in Geekbench OpenCL, and 10.4% ahead in Geekbench Vulkan. For gamers targeting current DX12 titles or compute-heavy workloads, the RTX 3070 Ti Mobile is the stronger pick.

The RTX 2070 SUPER, however, is not obsolete. It wins in legacy DirectX 9 and 10 tests, and it holds a 2.4% edge in Passmark G3D and a 0.7% edge in GPU Compute. It also wins the G2D test by 10.8%. For users running older APIs or needing desktop connectivity with three DisplayPort outputs and USB Type-C, the RTX 2070 SUPER remains competitive.

The tie in head-to-head wins (5 each) reveals that neither card is universally superior. The RTX 3070 Ti Mobile wins the modern tests by large margins, while the RTX 2070 SUPER wins legacy and some compute tests by smaller margins. The RTX 3070 Ti Mobile is the better overall performer, but only if your workload aligns with its strengths in DX12 and Vulkan. If you prioritize raw texture rate (283.2 versus 259.4 GTexel/s) or FP16 compute (18.12 versus 16.60 TFLOPS), the RTX 2070 SUPER has arguments in its favor.

Specification Differences

| Specification | RTX 3070 Ti Mobile | RTX 2070 SUPER |

|---|---|---|

| Architecture | Ampere | Turing |

| Chip | GA104 | TU104 |

| Process node | 8 nm | 12 nm |

| Foundry | Samsung | TSMC |

| Transistors | 17,400 million | 13,600 million |

| Die size | 392 mm² | 545 mm² |

| Transistor density | 44.4M / mm² | 25.0M / mm² |

| Base clock | 915 MHz | 1605 MHz |

| Boost clock | 1410 MHz | 1770 MHz |

| Shading units | 5888 | 2560 |

| TMUs | 184 | 160 |

| ROPs | 96 | 64 |

| RT cores | 46 | 40 |

| Tensor cores | 184 | 320 |

| Pixel rate | 135.4 GPixel/s | 113.3 GPixel/s |

| Texture rate | 259.4 GTexel/s | 283.2 GTexel/s |

| FP32 | 16.60 TFLOPS | 9.062 TFLOPS |

| FP16 | 16.60 TFLOPS (1:1) | 18.12 TFLOPS (2:1) |

| TDP | 115 W | 215 W |

| Bus interface | PCIe 4.0 x16 | PCIe 3.0 x16 |

| Slot width | Not specified | Dual-slot |

| Power connectors | None | 1x 6-pin + 1x 8-pin |

| Suggested PSU | Not specified | 550 W |

| Display outputs | Portable Device Dependent | 1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C |

| Dimensions | Not specified | 267 mm (10.5 in) length, 116 mm (4.6 in) height, 35 mm (1.4 in) width |

| Release date | 2022-01-03 | 2019-07-08 |

| Launch MSRP | Not provided | 499 USD |

Head-to-Head Benchmarks

3DMark Steel Nomad (DX12)

The RTX 3070 Ti Mobile delivers a dominant 2,572 score versus 1,651 for the RTX 2070 SUPER. The 55.8% lead is the largest margin in any test. This indicates the Ampere architecture's shader count (5,888 versus 2,560) and newer process provide a massive advantage in modern DX12 rendering.

Geekbench OpenCL

The RTX 3070 Ti Mobile scores 106,022 against 83,358 for the RTX 2070 SUPER, a 27.2% advantage. This compute-heavy test rewards the higher FP32 throughput (16.60 versus 9.062 TFLOPS) and the 96 ROPs versus 64.

Geekbench Vulkan

A 10.4% win for the RTX 3070 Ti Mobile: 100,027 versus 90,637. The margin is smaller than OpenCL, suggesting the Turing card's 320 tensor cores help narrow the gap in Vulkan compute workloads.

Passmark DirectX 10

The RTX 2070 SUPER wins 132 versus 124, a 6.1% margin. This is a modest win for the older architecture, likely benefiting from its higher clock speeds (1,770 MHz boost versus 1,410 MHz).

Passmark DirectX 11

The RTX 3070 Ti Mobile narrowly wins 153 versus 151, just 1.3% ahead. This near-tie shows that in older DX11 titles, the two cards are effectively equal despite the architectural gap.

Passmark DirectX 12

The RTX 3070 Ti Mobile wins 73 versus 67, a 9% margin. The Ampere card's advantage in DX12 is consistent with the 3DMark result, though the absolute scores are low.

Passmark DirectX 9

The RTX 2070 SUPER wins by the biggest margin in its favor: 223 versus 196, a 12.1% lead. Legacy DX9 workloads clearly favor the Turing card's higher clock speeds and texture rate.

Passmark G2D

The RTX 2070 SUPER wins 878 versus 783, a 10.8% margin. This 2D graphics test may reflect the desktop card's dedicated display outputs and higher base clock of 1,605 MHz versus 915 MHz.

Passmark G3D

The RTX 2070 SUPER wins 18,169 versus 17,727, a 2.4% margin. This is a close result, and it means the older desktop card still edges out the newer mobile part in this aggregate 3D test.

Passmark GPU Compute

The RTX 2070 SUPER wins 7,557 versus 7,504, a 0.7% margin. This is essentially a statistical tie, showing that compute performance is balanced between the two despite the architectural differences.

The overall picture is one of a split decision. The RTX 3070 Ti Mobile wins the modern, high-level benchmarks (Steel Nomad, OpenCL, Vulkan) with margins from 10.4% to 55.8%. The RTX 2070 SUPER wins the legacy and some 2D/compute tests with margins from 0.7% to 12.1%. The average score of 23,518 versus 20,282, however, gives the RTX 3070 Ti Mobile the overall crown.