AMD Radeon 780M vs NVIDIA GeForce RTX 2070 SUPER Comparison

# FAQ

Q: How does the NVIDIA GeForce RTX 2070 SUPER compare to the AMD Radeon 780M in the 3DMark Steel Nomad DX12 test?

A: The RTX 2070 SUPER scores 1651, while the Radeon 780M scores 480. This gives the NVIDIA part a 244% advantage in that specific test.

Q: Which GPU has the higher average benchmark score across the database?

A: The RTX 2070 SUPER averages 20282 points, placing it in the 65th percentile of all GPUs. The Radeon 780M averages 17588 points, placing it in the 61st percentile. The difference is 2694 points in favor of the NVIDIA card.

Q: What is the Geekbench Vulkan performance gap between these two?

A: In Geekbench Vulkan, the RTX 2070 SUPER posts 90637 points versus 33683 for the Radeon 780M, a 169.1% difference in favor of NVIDIA.

Q: Are there any benchmark tests where the AMD Radeon 780M wins?

A: No. Across all three recorded head-to-head tests (3DMark Steel Nomad, Geekbench OpenCL, and Geekbench Vulkan), the RTX 2070 SUPER wins all three. The win tally is 3 for NVIDIA and 0 for AMD.

Q: How do the nearest rivals for each card compare in average score?

A: The RTX 2070 SUPER sits within 1.5% of the Intel Arc A750 (20582), Intel Arc B570 (20556), and NVIDIA RTX 3070 Mobile (20534). The Radeon 780M trades within 0.5% of the AMD Radeon Pro 560 (17551) and AMD Radeon Pro 460 (17509), and within 0.4% of the AMD Radeon HD 7790 (17666).

Q: What are the power requirements and form factors?

A: The RTX 2070 SUPER has a 215 W TDP, requires a dual-slot cooler, and uses 1x 6-pin plus 1x 8-pin power connectors with a 550 W suggested PSU. The Radeon 780M is an integrated GPU (IGP) with a 15 W TDP, no power connectors, and no suggested PSU listed.

Architecture Differences

The two GPUs represent fundamentally different design philosophies. The NVIDIA GeForce RTX 2070 SUPER is a discrete, high-power desktop card built on the Turing architecture, using the TU104 chip fabricated at TSMC on a 12 nm process. It packs 13,600 million transistors across a 545 mm² die, yielding a transistor density of 25.0M / mm². In contrast, the AMD Radeon 780M is an integrated graphics processor, part of the Phoenix APU, built on RDNA 3.0 architecture. It uses a 4 nm TSMC process and integrates 25,390 million transistors on a much smaller 178 mm² die, achieving a far higher density of 142.6M / mm².

The compute resources differ dramatically. The RTX 2070 SUPER features 2560 shading units, 160 texture mapping units, and 64 ROPs. It also includes dedicated hardware: 40 ray tracing cores and 320 tensor cores. The Radeon 780M, by comparison, has 768 shading units, 48 TMUs, and 32 ROPs. It includes 12 ray tracing cores but has no tensor cores listed in the database.

Clock behavior also reflects their different roles. The RTX 2070 SUPER runs at a base clock of 1605 MHz and boosts to 1770 MHz. The Radeon 780M has a much lower base of 800 MHz but a far higher boost of 2900 MHz, reflecting the efficiency of the 4 nm process. Memory configurations are entirely different as well: the RTX 2070 SUPER uses 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth, while the Radeon 780M relies on system shared memory, with system-dependent bandwidth.

The feature sets are nearly identical at the API level. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. However, the underlying implementations diverge. The NVIDIA card uses a PCIe 3.0 x16 interface, while the AMD IGP uses PCIe 4.0 x8. The RTX 2070 SUPER offers dedicated display outputs (1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C), whereas the Radeon 780M's outputs are motherboard dependent.

Head-to-Head Benchmarks

The recorded head-to-head data shows a clean sweep for the NVIDIA GeForce RTX 2070 SUPER across all three benchmark tests. The largest margin appears in Geekbench OpenCL, where the NVIDIA card scores 83358 against the AMD's 18602. That represents a 348.1% delta, meaning the RTX 2070 SUPER delivers more than four times the raw compute throughput in this particular OpenCL workload.

The 3DMark Steel Nomad DX12 test shows a 244% advantage for NVIDIA, with scores of 1651 versus 480. This test likely stresses sustained graphics rendering and shader throughput, where the RTX 2070 SUPER's 9.062 TFLOPS FP32 performance and dedicated 8 GB frame buffer give it a decisive edge over the Radeon 780M's 8.909 TFLOPS FP32 output, which is bottlenecked by shared system memory.

The closest margin, though still substantial, comes in Geekbench Vulkan. Here the RTX 2070 SUPER scores 90637, and the Radeon 780M scores 33683, a 169.1% difference. Vulkan's lower-level API can sometimes narrow gaps between architectures, but the sheer difference in shading units (2560 versus 768) and memory bandwidth (448 GB/s versus system dependent) keeps NVIDIA firmly ahead.

It is worth remembering the Radeon 780M's FP32 figure of 8.909 TFLOPS is remarkably close to the RTX 2070 SUPER's 9.062 TFLOPS on paper. Yet the benchmark results show a wide gulf. This suggests that the Radeon 780M's performance is heavily constrained by its shared memory subsystem and lower pixel and texture rates. The RTX 2070 SUPER produces 113.3 GPixel/s and 283.2 GTexel/s, compared to 92.80 GPixel/s and 139.2 GTexel/s for the AMD part. The fill rate advantage, combined with dedicated VRAM, translates into the large score deltas observed.

The average benchmark scores corroborate the head-to-head results. The RTX 2070 SUPER's average of 20282 places it just 1% below the NVIDIA Quadro M4000M (20480) and 1.2% below the RTX 3070 Mobile (20534). The Radeon 780M's average of 17588 sits 0.2% above the AMD Radeon Pro 560 (17551) and 0.3% below the NVIDIA GeForce RTX 4060 (17639). In percentile terms, the RTX 2070 SUPER ranks higher globally (65th versus 61st), but both cards occupy similar tiers relative to the entire GPU landscape.

The Verdict

The data presents a straightforward conclusion: the NVIDIA GeForce RTX 2070 SUPER outperforms the AMD Radeon 780M in every recorded benchmark. The margins are not subtle. In OpenCL, the NVIDIA card leads by 348.1%; in 3DMark Steel Nomad, by 244%; and in Vulkan, by 169.1%. For anyone prioritizing raw graphics performance, the RTX 2070 SUPER is the clear choice based on the database results.

However, the Radeon 780M is not without its own rationale. Its 15 W TDP versus 215 W TDP represents a massive efficiency advantage. It requires no power connectors, occupies no expansion slot, and its performance, while lower, still lands in the 61st percentile of all GPUs. It is competitive with the NVIDIA GeForce RTX 4060 in average score (17639 versus 17588, a mere 0.3% gap). The Radeon 780M is an integrated solution, meaning the host system provides memory and cooling, which simplifies system design.

The RTX 2070 SUPER, on the other hand, is an end-of-life discrete card with a launch MSRP of 499 USD. It demands a dual-slot cooler, a 550 W power supply, and a PCIe 3.0 x16 slot. Its transistor count is lower (13,600 million versus 25,390 million) and its process node is older (12 nm versus 4 nm), yet its dedicated memory and higher fill rates deliver the benchmark wins.

For a desktop user with an existing power supply and case space, the RTX 2070 SUPER is the performance pick. For a compact or mobile system where power draw and space are at a premium, the Radeon 780M offers a reasonable level of graphics capability without any add-in card requirements. The database shows that one GPU wins on raw speed; the other wins on integration and power economy.

Specification Differences

| Specification | NVIDIA GeForce RTX 2070 SUPER | AMD Radeon 780M |

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

| Architecture | Turing | RDNA 3.0 |

| Process node | 12 nm | 4 nm |

| Die size | 545 mm² | 178 mm² |

| Transistors | 13,600 million | 25,390 million |

| Transistor density | 25.0M / mm² | 142.6M / mm² |

| Base clock | 1605 MHz | 800 MHz |

| Boost clock | 1770 MHz | 2900 MHz |

| Memory size | 8 GB | System Shared |

| Memory type | GDDR6 | System Shared |

| Memory bus width | 256 bit | System Shared |

| Memory bandwidth | 448.0 GB/s | System Dependent |

| Shading units | 2560 | 768 |

| TMUs | 160 | 48 |

| ROPs | 64 | 32 |

| RT cores | 40 | 12 |

| Tensor cores | 320 | None |

| Pixel rate | 113.3 GPixel/s | 92.80 GPixel/s |

| Texture rate | 283.2 GTexel/s | 139.2 GTexel/s |

| FP32 performance | 9.062 TFLOPS | 8.909 TFLOPS |

| FP16 performance | 18.12 TFLOPS (2:1) | 8.909 TFLOPS (1:1) |

| TDP | 215 W | 15 W |

| Slot width | Dual-slot | IGP |

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

| Suggested PSU | 550 W | None |

| Bus interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

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

| Production status | End-of-life | Active |

| Release date | 2019-07-08 | 2024-01-30 |

| Predecessor | GeForce 10 | Navi II IGP |

| Successor | GeForce 30 | Navi III IGP |