Intel Arc A770M vs NVIDIA GeForce RTX 2070 SUPER Comparison

The NVIDIA GeForce RTX 2070 SUPER and the Intel Arc A770M represent two distinctly different approaches to high-performance graphics. The RTX 2070 SUPER is a desktop card from NVIDIA’s Turing generation, built on a mature 12 nm process, while the Arc A770M is a mobile-focused part from Intel’s Xe-HPG architecture, manufactured on a denser 6 nm node. The recorded benchmark data shows a clear split: the RTX 2070 SUPER wins seven of ten head-to-head tests, while the Arc A770M takes three, including a decisive victory in the modern DirectX 12 workload. This analysis walks through the numbers, architecture, and use-case implications without relying on outside specifications.

Where Each One Wins

The RTX 2070 SUPER dominates the legacy and compute-oriented tests. In Passmark’s DirectX 10 and DirectX 11 suites, it scores 132 and 151, respectively, compared to the Arc A770M’s 56 and 69. That is a lead of 135.7% in DirectX 10 and 118.8% in DirectX 11. The older DirectX 9 test also favors NVIDIA, with a score of 223 versus 178, a 25.3% advantage. The gap widens further in general 3D performance: the RTX 2070 SUPER posts 18169 in Passmark G3D, while the Arc A770M manages only 11774, a 54.3% difference. Compute workloads follow suit, with the NVIDIA card scoring 7557 in Passmark GPU Compute versus 4778 for Intel, a 58.2% edge. Even in the 2D test, the RTX 2070 SUPER leads with 878 against 711, a 23.5% margin.

The Arc A770M, however, wins the tests that matter for future-facing workloads. In 3DMark Steel Nomad (DirectX 12), it scores 2278 against the RTX 2070 SUPER’s 1651, which is a 27.5% advantage for Intel. This is the single largest win for either card in absolute percentage terms among the head-to-head results. The Arc A770M also takes the OpenCL compute test, scoring 89494 versus 83358, a 6.9% lead. Finally, in Passmark DirectX 12, it edges out NVIDIA with 70 versus 67, a narrow 4.3% margin. These three wins cluster around modern API and compute workloads, suggesting the Arc A770M’s architecture is better optimized for newer rendering paths, even if its overall driver maturity or legacy support lags behind.

The use-case split is therefore straightforward: the RTX 2070 SUPER is the stronger choice for older DirectX titles, general 3D rendering, and GPU compute tasks that rely on established code paths. The Arc A770M excels in DirectX 12-native applications and OpenCL-heavy workflows, where its higher raw compute throughput can be leveraged.

Architecture Differences

The two GPUs are built on different process nodes and have fundamentally different transistor budgets. The RTX 2070 SUPER uses TSMC’s 12 nm process, packing 13,600 million transistors into a 545 mm² die, yielding a transistor density of 25.0 million per square millimeter. The Arc A770M, also from TSMC but on a 6 nm node, fits 21,700 million transistors into a smaller 406 mm² die, achieving a density of 53.4 million per square millimeter. This density advantage is substantial, meaning Intel packs nearly 60% more transistors into a physically smaller chip.

The compute resources differ sharply. The Arc A770M has 4096 shading units, 256 texture mapping units, and 128 render output units. The RTX 2070 SUPER has 2560 shading units, 160 TMUs, and 64 ROPs. Intel’s card also has 32 ray tracing cores, while NVIDIA’s has 40. Notably, the RTX 2070 SUPER includes 320 tensor cores, which the Arc A770M lacks entirely. The memory subsystems diverge as well: both use GDDR6 on a 256-bit bus, but the Arc A770M has 16 GB of memory with a bandwidth of 512.0 GB/s, whereas the RTX 2070 SUPER offers 8 GB with 448.0 GB/s. Clock speeds favor Intel too: the Arc A770M runs at a base of 1650 MHz and boosts to 2050 MHz, while the RTX 2070 SUPER runs at 1605 MHz base and 1770 MHz boost.

The power profiles are inverted. The RTX 2070 SUPER has a 215 W TDP, requires a 1x 6-pin plus 1x 8-pin power connector, and suggests a 550 W PSU. The Arc A770M has a 120 W TDP and is classified as an IGP (integrated graphics processor) for portable devices, meaning it draws significantly less power and has no external power connector. The interface also differs: Intel uses PCIe 4.0 x16, while NVIDIA uses PCIe 3.0 x16. The RTX 2070 SUPER is a dual-slot desktop card measuring 267 mm in length, while the Arc A770M has no listed dimensions, consistent with its mobile integration. Both support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, so API-level feature parity exists despite the internal differences.

Head-to-Head Benchmarks

The most striking result is the 3DMark Steel Nomad DirectX 12 test. The Arc A770M scores 2278, while the RTX 2070 SUPER scores 1651. The difference is 27.5% in favor of Intel. This is not a marginal edge; it is a clear generational improvement in modern API efficiency. The RTX 2070 SUPER’s older Turing architecture appears to struggle with the more demanding geometry and ray tracing workloads in Steel Nomad, while the Arc A770M’s Xe-HPG design handles them more effectively.

In OpenCL, the Arc A770M again leads, scoring 89494 against 83358. The 6.9% gap is modest but consistent with its higher FP32 throughput (16.79 TFLOPS versus 9.062 TFLOPS). The Vulkan test reverses the trend: the RTX 2070 SUPER scores 90637, while the Arc A770M scores 74422, giving NVIDIA a 21.8% advantage. This suggests that Vulkan driver optimization is stronger on the NVIDIA side, despite Intel’s architectural advantages.

The Passmark suite reveals the RTX 2070 SUPER’s dominance in legacy APIs. The DirectX 10 test shows a 135.7% lead (132 versus 56), and DirectX 11 shows a 118.8% lead (151 versus 69). These are enormous margins, indicating that Intel’s drivers do not optimize for older DirectX feature levels. DirectX 9 is closer but still favors NVIDIA: 223 versus 178, a 25.3% edge. The DirectX 12 Passmark test is the exception, with Intel winning 70 to 67, a 4.3% difference that is within noise but still recorded as a win.

The general 3D performance test (Passmark G3D) is lopsided: 18169 for NVIDIA versus 11774 for Intel, a 54.3% advantage. This aggregate score likely reflects a mix of workloads, and the RTX 2070 SUPER’s superior legacy and Vulkan performance pulls it ahead. The GPU Compute test similarly favors NVIDIA, 7557 versus 4778, a 58.2% lead. Finally, the 2D test (Passmark G2D) goes to NVIDIA 878 to 711, a 23.5% margin. Out of ten tests, the RTX 2070 SUPER wins seven, but the three Intel wins include the most modern DirectX 12 benchmark, which may be more indicative of future game performance.

The Verdict

The data supports a clear recommendation based on workload. For users prioritizing legacy DirectX games, general 3D rendering, or compute tasks that rely on mature libraries, the RTX 2070 SUPER is the stronger card. Its 54.3% lead in Passmark G3D and 58.2% lead in GPU Compute are decisive. The Vulkan performance is also significantly better, with a 21.8% advantage, which matters for many modern cross-platform titles.

The Arc A770M, however, is the better choice for DirectX 12-native applications. Its 27.5% win in 3DMark Steel Nomad is the largest single-test margin in the comparison, and it also leads in OpenCL by 6.9%. The 16 GB memory capacity and 512.0 GB/s bandwidth provide headroom for texture-heavy workloads, even if the benchmark scores do not always reflect that advantage. The 120 W TDP also makes it a far more power-efficient option for portable devices, though that is a system-level consideration rather than a direct performance metric.

The average benchmark scores corroborate the split. The RTX 2070 SUPER has an average score of 20282, placing it at the 65th percentile among all GPUs, while the Arc A770M averages 18383, at the 62nd percentile. The RTX 2070 SUPER’s nearest rivals include the Intel Arc B570 and Arc A750, both within 1.5% of its score. The Arc A770M’s nearest rival is the AMD Radeon RX 460, which is essentially tied at 0.1% difference. This suggests the RTX 2070 SUPER sits in a slightly higher performance tier overall, but the Arc A770M is not far behind, and its specific strengths in DirectX 12 could make it preferable for future-proofing.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 2070 SUPER has an average score of 20282, compared to the Intel Arc A770M’s 18383. The RTX 2070 SUPER also has a higher percentile ranking at 65 versus 62.

Q: How much faster is the Arc A770M in 3DMark Steel Nomad?

A: The Arc A770M scores 2278, while the RTX 2070 SUPER scores 1651. This is a 27.5% advantage for Intel in this DirectX 12 workload.

Q: Does the RTX 2070 SUPER have any ray tracing advantage?

A: The RTX 2070 SUPER has 40 ray tracing cores, while the Arc A770M has 32. However, the benchmark data does not isolate ray tracing performance, so no direct comparison is available from the recorded tests.

Q: What is the memory capacity difference?

A: The Arc A770M has 16 GB of GDDR6 memory, while the RTX 2070 SUPER has 8 GB. Both use a 256-bit bus, but the Arc A770M offers 512.0 GB/s bandwidth versus 448.0 GB/s for NVIDIA.

Q: Which card performs better in Vulkan?

A: The RTX 2070 SUPER scores 90637 in Geekbench Vulkan, while the Arc A770M scores 74422. That is a 21.8% lead for NVIDIA.

Q: Is the Arc A770M more power efficient?

A: Yes, the Arc A770M has a 120 W TDP, whereas the RTX 2070 SUPER has a 215 W TDP. The Intel card is classified as an IGP with no external power connectors, while the NVIDIA card requires a 6-pin and 8-pin connector.

Specification Differences

The following fields differ between the two GPUs, based solely on the recorded data:

| Specification | NVIDIA GeForce RTX 2070 SUPER | Intel Arc A770M |

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

| Process Node | 12 nm | 6 nm |

| Transistors | 13,600 million | 21,700 million |

| Die Size | 545 mm² | 406 mm² |

| Transistor Density | 25.0M / mm² | 53.4M / mm² |

| Base Clock | 1605 MHz | 1650 MHz |

| Boost Clock | 1770 MHz | 2050 MHz |

| Memory Clock | 1750 MHz (14 Gbps effective) | 2000 MHz (16 Gbps effective) |

| Memory Size | 8 GB | 16 GB |

| Memory Bandwidth | 448.0 GB/s | 512.0 GB/s |

| Shading Units | 2560 | 4096 |

| TMUs | 160 | 256 |

| ROPs | 64 | 128 |

| RT Cores | 40 | 32 |

| Tensor Cores | 320 | None (null) |

| Pixel Rate | 113.3 GPixel/s | 262.4 GPixel/s |

| Texture Rate | 283.2 GTexel/s | 524.8 GTexel/s |

| FP32 Performance | 9.062 TFLOPS | 16.79 TFLOPS |

| FP16 Performance | 18.12 TFLOPS (2:1) | 33.59 TFLOPS (2:1) |

| TDP | 215 W | 120 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 x16 |

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

| Dimensions | 267 mm length, 116 mm height, 35 mm width | Not listed |

| Release Date | 2019-07-08 | Not listed |

| Launch MSRP | 499 USD | Not listed |

The differences in die size, transistor count, and clock speeds explain the performance split. The Arc A770M’s smaller die with more transistors and higher clocks gives it a theoretical throughput advantage, but the RTX 2070 SUPER’s mature driver stack and tensor cores (which the Arc lacks) compensate in many real-world tests. The power and form factor differences also reflect their intended markets: one is a desktop card, the other a mobile IGP.