NVIDIA GeForce RTX 2060 SUPER vs NVIDIA GeForce RTX 3070 Mobile Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce RTX 3070 Mobile leads with an average benchmark score of 20534, compared to 18093 for the NVIDIA GeForce RTX 2060 SUPER, a margin of roughly 13.5% based on the recorded data.

Q: How do the two cards compare in the 3DMark Steel Nomad DX12 test?

A: The RTX 3070 Mobile scores 2380, which is 18.3% higher than the RTX 2060 SUPER's 2011, making it the clear winner in this modern DX12 workload.

Q: Is there any benchmark where the RTX 2060 SUPER wins?

A: Yes. The RTX 2060 SUPER wins in PassMark DirectX 9 (218 vs 160, a 26.6% advantage), PassMark G2D (854 vs 641, a 24.9% advantage), and PassMark G3D (16462 vs 15309, a 7% advantage).

Q: What is the transistor density difference between the two GPUs?

A: The RTX 3070 Mobile packs 44.4 million transistors per square millimeter on Samsung's 8 nm process, while the RTX 2060 SUPER has 24.3 million per square millimeter on TSMC's 12 nm node.

Q: Do both GPUs support the same DirectX version?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 according to the specification data.

Q: Which GPU has more RT cores?

A: The RTX 3070 Mobile has 40 RT cores, while the RTX 2060 SUPER has 34 RT cores, a difference of 6 cores in favor of the Ampere part.

Architecture Differences

The NVIDIA GeForce RTX 3070 Mobile is built on the Ampere architecture using the GA104 chip, fabricated on Samsung's 8 nm process. The RTX 2060 SUPER uses the Turing architecture with the TU106 chip on TSMC's 12 nm node. This process shrink allows the RTX 3070 Mobile to integrate 17,400 million transistors into a 392 mm² die, achieving a transistor density of 44.4 million per square millimeter. The RTX 2060 SUPER contains 10,800 million transistors on a larger 445 mm² die, with a density of 24.3 million per square millimeter. The density advantage of the RTX 3070 Mobile is substantial and directly enables its higher compute throughput.

The shading unit count differs dramatically. The RTX 3070 Mobile has 5120 shading units, 160 texture mapping units, and 80 ROPs, while the RTX 2060 SUPER has 2176 shading units, 136 TMUs, and 64 ROPs. The RT core counts also differ: 40 versus 34. In tensor cores, the RTX 2060 SUPER actually has more, with 272 compared to 160 on the RTX 3070 Mobile, though the Ampere implementation is architecturally newer. The FP32 throughput reflects the shading unit disparity: 15.97 TFLOPS for the RTX 3070 Mobile versus 7.181 TFLOPS for the RTX 2060 SUPER. FP16 performance is 15.97 TFLOPS (1:1) on the Ampere card, while the Turing card delivers 14.36 TFLOPS (2:1).

Clock behavior also distinguishes the two. The RTX 3070 Mobile has a base clock of 1110 MHz and a boost clock of 1560 MHz, whereas the RTX 2060 SUPER runs at a higher base of 1470 MHz and boost of 1650 MHz. Despite the lower clocks, the RTX 3070 Mobile achieves higher pixel and texture rates: 124.8 GPixel/s and 249.6 GTexel/s versus 105.6 GPixel/s and 224.4 GTexel/s on the RTX 2060 SUPER. The memory subsystem is identical in capacity, type, bus width, and bandwidth: 8 GB GDDR6 on a 256 bit bus delivering 448.0 GB/s. Both run memory at 1750 MHz with 14 Gbps effective speed.

Power and physical specifications diverge sharply. The RTX 3070 Mobile is rated at 115 W with no power connectors, being a portable device dependent solution. The RTX 2060 SUPER is a dual-slot desktop card rated at 175 W, requiring a single 8-pin connector and a 450 W suggested PSU. The RTX 2060 SUPER measures 229 mm in length, 113 mm in height, and 35 mm in width, with display outputs including 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C. The RTX 3070 Mobile's display outputs are listed as portable device dependent. Bus interface also differs: PCIe 4.0 x16 for the mobile card versus PCIe 3.0 x16 for the desktop card.

The Verdict

The data points firmly toward the NVIDIA GeForce RTX 3070 Mobile as the stronger performer in modern workloads. It wins 7 of the 10 head-to-head benchmark comparisons, including the most demanding modern test, 3DMark Steel Nomad DX12, where it leads by 18.3%. Its average benchmark score of 20534 places it in the 65th percentile of all GPUs, while the RTX 2060 SUPER sits at 18093 and the 62nd percentile. The RTX 3070 Mobile also holds a 20.8% lead in Geekbench OpenCL and a 12.1% lead in Geekbench Vulkan, indicating broad compute and API-level superiority.

However, the RTX 2060 SUPER is not without its niches. It wins PassMark G3D by 7%, which is the classic gaming-oriented aggregate score, and it dominates in legacy DirectX 9 workloads by 26.6%. It also has a 24.9% advantage in PassMark G2D, which reflects 2D and desktop compositing performance. For users prioritizing maximum performance in current DX12 and Vulkan titles, the RTX 3070 Mobile is the clear choice. For users running older DirectX 9 applications or those who weigh the PassMark G3D metric heavily, the RTX 2060 SUPER retains relevance despite its lower overall average score.

The architectural gap is decisive in compute-heavy tasks. The RTX 3070 Mobile delivers more than double the FP32 throughput (15.97 TFLOPS vs 7.181 TFLOPS) and nearly double the pixel rate, with a 115 W power envelope compared to 175 W on the older card. The RTX 2060 SUPER's higher base and boost clocks do not compensate for its older, less dense architecture and lower core counts. The verdict is straightforward: the RTX 3070 Mobile is the superior GPU for modern and compute-intensive workloads, while the RTX 2060 SUPER remains competitive only in specific legacy or 2D scenarios.

Specification Differences

| Specification | NVIDIA GeForce RTX 3070 Mobile | NVIDIA GeForce RTX 2060 SUPER |

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

| Architecture | Ampere | Turing |

| Chip | GA104 | TU106 |

| Process Node | 8 nm | 12 nm |

| Foundry | Samsung | TSMC |

| Transistors | 17,400 million | 10,800 million |

| Die Size | 392 mm² | 445 mm² |

| Transistor Density | 44.4M / mm² | 24.3M / mm² |

| Base Clock | 1110 MHz | 1470 MHz |

| Boost Clock | 1560 MHz | 1650 MHz |

| Shading Units | 5120 | 2176 |

| TMUs | 160 | 136 |

| ROPs | 80 | 64 |

| RT Cores | 40 | 34 |

| Tensor Cores | 160 | 272 |

| Pixel Rate | 124.8 GPixel/s | 105.6 GPixel/s |

| Texture Rate | 249.6 GTexel/s | 224.4 GTexel/s |

| FP32 | 15.97 TFLOPS | 7.181 TFLOPS |

| FP16 | 15.97 TFLOPS (1:1) | 14.36 TFLOPS (2:1) |

| TDP | 115 W | 175 W |

| Slot Width | Not specified | Dual-slot |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | Not specified | 450 W |

| Bus Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |

| Display Outputs | Portable Device Dependent | 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, 1x USB Type-C |

| Dimensions | Not specified | 229 mm x 113 mm x 35 mm |

| Release Date | 2021-01-11 | 2019-07-08 |

| Predecessor | GeForce 20 Mobile | GeForce 10 |

| Successor | None | GeForce 30 |

| Launch MSRP | Not specified | 399 USD |

Head-to-Head Benchmarks

The RTX 3070 Mobile wins decisively in the most modern and compute-oriented tests. In 3DMark Steel Nomad DX12, it scores 2380 against 2011, a 18.3% lead that signals stronger DirectX 12 performance. Geekbench OpenCL shows a 20.8% advantage, with scores of 92939 versus 76957, and Geekbench Vulkan adds a 12.1% margin, 86768 versus 77402. These are the largest wins for the RTX 3070 Mobile and reflect its Ampere architecture's raw compute advantage.

The PassMark suite tells a more nuanced story. The RTX 3070 Mobile wins DirectX 10 by a narrow 1.8% (113 vs 111) and DirectX 11 by 6.2% (138 vs 130). DirectX 12 shows a 4.9% lead (64 vs 61). GPU compute is nearly tied, with the RTX 3070 Mobile ahead by just 1.6% (6827 vs 6721). These margins are far smaller than the Geekbench and 3DMark deltas, indicating that the PassMark tests do not fully exploit the RTX 3070 Mobile's hardware.

The RTX 2060 SUPER claims the remaining three tests. Its largest win is PassMark DirectX 9, where it scores 218 versus 160, a 26.6% advantage. PassMark G2D shows a 24.9% lead (854 vs 641), and PassMark G3D gives it a 7% edge (16462 vs 15309). The G3D result is notable because it is a common gaming performance indicator, and despite losing the overall average score comparison, the RTX 2060 SUPER outperforms in this aggregate metric. The DirectX 9 result indicates that legacy API workloads favor the Turing architecture's higher clocks and older driver optimization path.

Where Each One Wins

The NVIDIA GeForce RTX 3070 Mobile wins in modern API workloads, compute-heavy applications, and efficiency. Its 18.3% lead in 3DMark Steel Nomad DX12 makes it the better choice for current DirectX 12 games and future titles built on that API. Geekbench OpenCL and Vulkan wins of 20.8% and 12.1% respectively show strong performance in general-purpose GPU compute, including machine learning inference, video encoding, and cross-platform rendering. The 115 W power envelope, despite higher performance, gives it a clear efficiency advantage over the 175 W RTX 2060 SUPER. Its PCIe 4.0 x16 interface and 8 nm process with 44.4M transistors per square millimeter also position it as the more future-proof option. The RTX 3070 Mobile also wins all PassMark DirectX 10, 11, and 12 tests, albeit by smaller margins, meaning it handles the full modern DirectX stack better.

The NVIDIA GeForce RTX 2060 SUPER wins in legacy DirectX 9 scenarios, 2D workloads, and one aggregate gaming metric. Its 26.6% DirectX 9 advantage makes it the preferable card for older games or applications that still rely on that API, where the higher base and boost clocks (1470 MHz and 1650 MHz) help overcome its lower core counts. The 24.9% G2D win indicates stronger 2D compositing and desktop rendering performance, which may matter for productivity workflows with many windows or 2D graphics tools. The 7% PassMark G3D win is the most interesting result: despite losing the overall average benchmark score, the RTX 2060 SUPER performs better in this specific gaming aggregate, so users who prioritize that metric might find it competitive. Its higher tensor core count (272 vs 160) also suggests potentially different characteristics in AI workloads, though the benchmark data does not directly measure that. The RTX 2060 SUPER also has a launch MSRP of 399 USD, which was its official price at introduction. For users on legacy software stacks or those who weigh PassMark G3D heavily, the RTX 2060 SUPER retains specific advantages, but the RTX 3070 Mobile dominates the broader performance picture.