GPU Comparison

The Intel Arc A350M and NVIDIA GeForce RTX 2080 SUPER occupy completely different tiers of the GPU market, and the benchmark data reflects that gap clearly. Across the two shared head-to-head tests, the RTX 2080 SUPER wins both by massive margins, with the Arc A350M trailing by 75.3% in Geekbench OpenCL and 77.8% in Geekbench Vulkan. The RTX 2080 SUPER’s average benchmark score of 24170 places it at the 69th percentile of all GPUs, while the Arc A350M’s 24647 average sits at the 70th percentile, a statistical tie in overall standing, but the individual workloads tell a very different story.

Head-to-Head Benchmarks

The Geekbench OpenCL test is the closest of the two comparisons, yet it remains a decisive NVIDIA victory. The RTX 2080 SUPER scores 99226, while the Arc A350M manages 24546. That is a 75.3% deficit for Intel’s part. In raw terms, the NVIDIA GPU delivers over four times the OpenCL performance. This workload typically scales with shading unit count and memory bandwidth, and the RTX 2080 SUPER has 3072 shading units versus 768 on the Arc A350M, along with 495.9 GB/s of bandwidth versus 112.0 GB/s.

The Geekbench Vulkan result widens the gap further. The RTX 2080 SUPER posts 111284, against 24747 for the Arc A350M. That translates to a 77.8% lead for NVIDIA. Vulkan is often more sensitive to driver overhead and architecture efficiency, and here the Turing-based card pulls even further ahead than in OpenCL. The delta between the two tests, 75.3% versus 77.8%, shows that NVIDIA’s advantage grows slightly when moving to the lower-level API.

No benchmark favors the Arc A350M. The head-to-head table records zero wins for Intel and two for NVIDIA. The closest the Arc A350M comes to competitiveness is in its own nearest-rival list, where it sits just 0.4% behind the AMD Radeon RX 590 and 0.5% behind the NVIDIA RTX A5000 Mobile. Meanwhile, the RTX 2080 SUPER’s nearest rivals include the GTX 780 Ti (0.3% behind) and the GTX 1630 (0.4% behind), indicating it competes in a similar performance band despite its much higher absolute scores.

Where Each One Wins

The RTX 2080 SUPER wins everywhere in direct comparison. Its 8 GB GDDR6 memory on a 256-bit bus delivers 495.9 GB/s bandwidth, compared to the Arc A350M’s 4 GB on a 64-bit bus with 112.0 GB/s. That bandwidth advantage alone, roughly 4.4 times, explains much of the performance gap in memory-intensive workloads. The RTX 2080 SUPER also has 64 ROPs versus 24 on the Arc A350M, giving it a 116.2 GPixel/s pixel rate against 52.80 GPixel/s. For rasterization-heavy scenes, that is a fundamental throughput advantage.

Texture performance follows the same pattern. The RTX 2080 SUPER’s 192 TMUs produce 348.5 GTexel/s, while the Arc A350M’s 48 TMUs yield 105.6 GTexel/s. In compute, the FP32 throughput of 11.15 TFLOPS on NVIDIA dwarfs the 3.379 TFLOPS on Intel. The RTX 2080 SUPER also brings 384 tensor cores and 48 RT cores, while the Arc A350M has 6 RT cores and no tensor core count listed. Any workload touching ray tracing or AI acceleration will decisively favor NVIDIA.

The Arc A350M does have one clear advantage: power efficiency. At 25 W TDP versus 250 W for the RTX 2080 SUPER, the Intel part uses one-tenth the power. For thin-and-light laptops where the RTX 2080 SUPER’s dual-slot cooler and 600 W suggested PSU are impossible, the Arc A350M is the only viable option. The RTX 2080 SUPER is a desktop-class card with 267 mm length, while the Arc A350M is an integrated graphics package (IGP) with no dedicated cooler. If the use case is a portable device with minimal thermal headroom, the Arc A350M wins by default.

Architecture Differences

The two GPUs come from different architectural eras. The Arc A350M uses Intel’s Xe-HPG architecture on the DG2-128 chip, built on a 6 nm TSMC process. The RTX 2080 SUPER uses NVIDIA’s Turing architecture on the TU104 chip, fabricated on a 12 nm TSMC process. Intel’s newer node allows a much denser design: 7,200 million transistors in 157 mm², yielding a transistor density of 45.9M/mm². NVIDIA packs 13,600 million transistors into 545 mm², for just 25.0M/mm². The Intel chip is smaller and denser, but that density does not translate into performance here.

Core counts differ by a factor of four. The Arc A350M has 768 shading units, 48 TMUs, and 24 ROPs. The RTX 2080 SUPER has 3072 shading units, 192 TMUs, and 64 ROPs. Ray tracing hardware also differs: 6 RT cores on Intel versus 48 on NVIDIA. The RTX 2080 SUPER adds 384 tensor cores, which the Arc A350M lacks entirely. Memory configuration is equally divergent, 4 GB GDDR6 on a 64-bit bus versus 8 GB GDDR6 on a 256-bit bus.

Clock behavior is notable. The Arc A350M boosts to 2200 MHz from a 1150 MHz base, a much higher boost ratio than the RTX 2080 SUPER’s 1650 MHz base to 1815 MHz boost. Despite the Intel part’s higher boost clock, its far lower core count and memory bandwidth cap its performance. Memory clocks differ slightly: 1750 MHz (14 Gbps effective) on Intel versus 1937 MHz (15.5 Gbps effective) on NVIDIA. The bus interface also differs, PCIe 4.0 x8 on Intel versus PCIe 3.0 x16 on NVIDIA. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

The Verdict

The data is unambiguous: the RTX 2080 SUPER is the faster GPU by every measured metric. It wins both head-to-head benchmarks with margins exceeding 75%, and its architectural advantages in core count, memory bandwidth, and specialized hardware are overwhelming. Anyone choosing between these two for gaming, rendering, or compute should pick the RTX 2080 SUPER without hesitation. The 69th percentile ranking versus the 70th for the Arc A350M might suggest parity, but that percentile is computed against different rival sets and does not reflect direct comparison.

The Arc A350M’s path to preference is narrow but real. Its 25 W TDP makes it suitable for ultra-portable devices where the RTX 2080 SUPER cannot physically fit, the latter requires a dual-slot cooler, 267 mm length, and a 600 W suggested PSU. The Arc A350M is an IGP, meaning it requires no separate power connectors and fits in any chassis with a PCIe 4.0 x8 slot. For a compact, low-power system where the absolute fastest performance is not the goal, the Arc A350M is the only option that exists.

The RTX 2080 SUPER’s launch MSRP was 699 USD. That price point, combined with its end-of-life status, suggests it was a high-end part in its day. The Arc A350M, with no listed MSRP, appears aimed at a different market segment entirely. The verdict from the data: pick the RTX 2080 SUPER for performance, pick the Arc A350M only when power and size constraints eliminate NVIDIA’s card.

FAQ

Q: Which GPU wins the Geekbench OpenCL benchmark?

A: The NVIDIA GeForce RTX 2080 SUPER wins with a score of 99226, while the Intel Arc A350M scores 24546, giving NVIDIA a 75.3% lead.

Q: How large is the Vulkan performance gap between the two?

A: The RTX 2080 SUPER scores 111284 in Geekbench Vulkan versus 24747 for the Arc A350M, a 77.8% advantage for NVIDIA.

Q: What is the memory bandwidth difference?

A: The RTX 2080 SUPER has 495.9 GB/s bandwidth from its 256-bit bus and 8 GB GDDR6, while the Arc A350M has 112.0 GB/s from a 64-bit bus and 4 GB GDDR6.

Q: Does the Arc A350M have tensor cores?

A: No tensor core count is listed for the Intel Arc A350M, whereas the RTX 2080 SUPER has 384 tensor cores.

Q: What are the TDP ratings for each GPU?

A: The Intel Arc A350M has a TDP of 25 W, while the NVIDIA GeForce RTX 2080 SUPER has a TDP of 250 W.

Q: Which GPU has a higher boost clock?

A: The Intel Arc A350M boosts to 2200 MHz, which is higher than the RTX 2080 SUPER’s 1815 MHz boost clock.

Specification Differences

| Specification | Intel Arc A350M | NVIDIA GeForce RTX 2080 SUPER |

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

| Architecture | Xe-HPG | Turing |

| Process Node | 6 nm | 12 nm |

| Transistors | 7,200 million | 13,600 million |

| Die Size | 157 mm² | 545 mm² |

| Transistor Density | 45.9M / mm² | 25.0M / mm² |

| Base Clock | 1150 MHz | 1650 MHz |

| Boost Clock | 2200 MHz | 1815 MHz |

| Memory Size | 4 GB | 8 GB |

| Memory Type | GDDR6 | GDDR6 |

| Memory Bus Width | 64 bit | 256 bit |

| Memory Bandwidth | 112.0 GB/s | 495.9 GB/s |

| Memory Clock | 1750 MHz (14 Gbps effective) | 1937 MHz (15.5 Gbps effective) |

| Shading Units | 768 | 3072 |

| TMUs | 48 | 192 |

| ROPs | 24 | 64 |

| RT Cores | 6 | 48 |

| Tensor Cores | Not listed | 384 |

| Pixel Rate | 52.80 GPixel/s | 116.2 GPixel/s |

| Texture Rate | 105.6 GTexel/s | 348.5 GTexel/s |

| FP32 Performance | 3.379 TFLOPS | 11.15 TFLOPS |

| FP16 Performance | 6.758 TFLOPS (2:1) | 22.30 TFLOPS (2:1) |

| TDP | 25 W | 250 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | Not listed | 1x 6-pin + 1x 8-pin |

| Suggested PSU | Not listed | 600 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 3.0 x16 |

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

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

| Release Date | 2022-03-29 | 2019-07-22 |

| Launch MSRP | Not listed | 699 USD |