Intel Arc A380 vs NVIDIA GeForce GTX 850M Comparison

# NVIDIA GeForce GTX 850M vs Intel Arc A380

The NVIDIA GeForce GTX 850M is a Maxwell-era mobile GPU from 2014, while the Intel Arc A380 is an Alchemist-generation desktop part from 2022. In the two direct benchmark comparisons available, the Intel Arc A380 wins both, with the GTX 850M finishing 74.3% behind in Geekbench OpenCL and 76.1% behind in Geekbench Vulkan. The GTX 850M holds a 46th percentile ranking among all GPUs, while the Arc A380 sits at the 44th percentile, a narrow gap that belies the magnitude of the per-test differences. The GTX 850M's average benchmark score is 9302, whereas the Arc A380 averages 8558, meaning the older NVIDIA part actually posts a higher average despite losing every head-to-head test.

Where Each One Wins

The Intel Arc A380 dominates in raw compute throughput and modern API performance. It wins both Geekbench OpenCL and Vulkan tests decisively, with scores of 38224 and 36736 respectively, versus 9821 and 8782 for the GTX 850M. The Arc A380 also brings dedicated ray tracing hardware with 8 RT cores, a feature the GTX 850M entirely lacks, making it the clear choice for workloads that leverage DirectX 12 Ultimate features or ray-traced effects. Its Passmark suite results further show strength across different API levels: DirectX 9 scores 73, DirectX 10 scores 37, DirectX 11 scores 38, DirectX 12 scores 35, and the 3DMark Steel Nomad DX12 test yields 808 points.

The GTX 850M, by contrast, offers no scenario in the data where it wins a direct benchmark. However, its average benchmark score of 9302 is 8.7% higher than the Arc A380's 8558, suggesting that in aggregate across a broader test suite, the older NVIDIA card holds its own relative to the Intel part's mixed results. The GTX 850M also has a higher percentile ranking at 46 versus 44, indicating it outperforms a slightly larger fraction of the GPU population. For users constrained to a 45 W TDP with integrated form factor and no external power connectors, the GTX 850M is the only viable option among the two, as the Arc A380 requires a dual-slot cooler, one 8-pin connector, and a 250 W suggested PSU.

Architecture Differences

The architectural gap between these two GPUs spans nearly a decade of development. The GTX 850M uses NVIDIA's GM107 chip built on Maxwell architecture, fabricated on TSMC's 28 nm process with 1,870 million transistors on a 148 mm² die, yielding a transistor density of 12.6 million transistors per mm². The Arc A380 uses Intel's DG2-128 chip on the Xe-HPG architecture, built on TSMC's 6 nm process with 7,200 million transistors on a 157 mm² die, achieving 45.9 million transistors per mm² — a 3.6x density advantage.

The GTX 850M's memory subsystem is modest: 2 GB of DDR3 on a 128-bit bus delivering 32.03 GB/s of bandwidth. The Arc A380 ships with 6 GB of GDDR6 on a 96-bit bus for 186.0 GB/s, a 5.8x bandwidth improvement. The Intel part's memory clock runs at 1937 MHz (15.5 Gbps effective), while the GTX 850M's memory runs at 1001 MHz (2 Gbps effective). The Arc A380 also doubles the shading units (1024 vs 640), TMUs (64 vs 40), and ROPs (32 vs 16). The Intel GPU's pixel rate of 65.60 GPixel/s and texture rate of 131.2 GTexel/s dwarf the GTX 850M's 14.43 GPixel/s and 36.08 GTexel/s.

Compute throughput follows suit: the Arc A380 delivers 4.198 TFLOPS of FP32 versus 1,154.6 GFLOPS for the NVIDIA part, and the Intel GPU also supports FP16 at 8.397 TFLOPS (2:1 ratio), a capability the GTX 850M lacks. The Arc A380's base clock is 2000 MHz with a 2050 MHz boost, whereas the GTX 850M has no listed base or boost clock. API support differs meaningfully: the GTX 850M supports DirectX 12 (11_0), while the Arc A380 supports DirectX 12 Ultimate (12_2), along with OpenGL 4.6 and Vulkan 1.4 on both. The bus interfaces differ as well — the GTX 850M uses PCIe 3.0 x16, while the Arc A380 uses PCIe 4.0 x8.

Head-to-Head Benchmarks

The Geekbench OpenCL test shows the Arc A380 at 38224 points against the GTX 850M's 9821, a 74.3% deficit for the NVIDIA part. This is a 3.9x raw score advantage for Intel, reflecting the massive difference in shading units, memory bandwidth, and clock speeds. The GTX 850M's nearest rivals in this range include the GeForce GTX 465 (9294), Radeon R7 M380 (9313), and GeForce GTX 960 (9273), all within 0.3% of its score. The Arc A380's closest competitors are the FirePro W5170M (8595), Radeon HD 8870M (8462), and GeForce MX330 (8458), with the Intel card leading them by 0.4% to 1.4%.

In Geekbench Vulkan, the Arc A380 scores 36736 versus the GTX 850M's 8782, a 76.1% gap that translates to a 4.2x advantage. This wider margin in Vulkan compared to OpenCL suggests the Arc A380's newer architecture scales better with modern low-level APIs. The GTX 850M's Vulkan score of 8782 is actually lower than its OpenCL score of 9821, while the Arc A380's Vulkan score of 36736 is only slightly below its OpenCL result of 38224, indicating more consistent performance across API types for the Intel GPU. The deltaPct values in the head-to-head data are negative from the GTX 850M's perspective, confirming that NVIDIA's part trails in both tests without a single win.

The Passmark results for the Arc A380 provide additional context, though no comparable GTX 850M data exists in the pack. The Intel GPU scores 6252 in Passmark G3D, 2762 in GPU Compute, and 610 in G2D. Its DirectX-specific Passmark scores are notably low — 73 for DX9, 37 for DX10, 38 for DX11, and 35 for DX12 — which is unusual given its strong Geekbench performance. This discrepancy suggests the Arc A380's driver stack may perform unevenly across different benchmark suites, though this cannot be verified against the GTX 850M without matching data.

FAQ

Q: Which GPU has better raw compute performance?

A: The Intel Arc A380 massively outperforms the GTX 850M in compute. It delivers 4.198 TFLOPS of FP32 versus 1,154.6 GFLOPS, and its Geekbench OpenCL score of 38224 is 74.3% higher than the GTX 850M's 9821.

Q: Does the GTX 850M win any benchmark in this comparison?

A: No. In the two head-to-head tests available (Geekbench OpenCL and Vulkan), the Arc A380 wins both. However, the GTX 850M has a higher average benchmark score of 9302 versus 8558 and a higher percentile ranking (46 vs 44).

Q: What are the memory specifications for each card?

A: The GTX 850M has 2 GB of DDR3 on a 128-bit bus with 32.03 GB/s bandwidth. The Arc A380 has 6 GB of GDDR6 on a 96-bit bus with 186.0 GB/s bandwidth, running at 1937 MHz (15.5 Gbps effective) versus 1001 MHz (2 Gbps effective).

Q: Which GPU supports ray tracing?

A: Only the Intel Arc A380 has dedicated ray tracing hardware, featuring 8 RT cores. The GTX 850M has no RT cores and its DirectX 12 (11_0) support does not include the Ultimate feature set that the Arc A380's DirectX 12 Ultimate (12_2) provides.

Q: What is the difference in power and physical requirements?

A: The GTX 850M has a 45 W TDP and is an integrated form factor with no power connectors, while the Arc A380 has a 75 W TDP, requires a dual-slot cooler with one 8-pin connector, and has a suggested PSU of 250 W. The Arc A380 also has physical dimensions of 222 mm length, 114 mm height, and 42 mm width.

Q: How do the nearest rivals compare for each GPU?

A: The GTX 850M's closest rival is the AMD Radeon R7 M380 at 9313 (0.1% lower), followed by the GeForce GTX 465 at 9294 (0.1% higher) and GeForce GTX 960 at 9273 (0.3% higher). The Arc A380's closest rival is the AMD FirePro W5170M at 8595 (0.4% higher), with the Radeon HD 8870M at 8462 (1.1% lower) and GeForce MX330 at 8458 (1.2% lower).

Specification Differences

| Specification | NVIDIA GeForce GTX 850M | Intel Arc A380 |

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

| Architecture | Maxwell | Xe-HPG |

| Generation | GeForce 800M | Alchemist (Arc 3) |

| Process Node | 28 nm | 6 nm |

| Transistors | 1,870 million | 7,200 million |

| Die Size | 148 mm² | 157 mm² |

| Transistor Density | 12.6M / mm² | 45.9M / mm² |

| Base Clock | — | 2000 MHz |

| Boost Clock | — | 2050 MHz |

| Memory Clock | 1001 MHz (2 Gbps) | 1937 MHz (15.5 Gbps) |

| Memory Size | 2 GB | 6 GB |

| Memory Type | DDR3 | GDDR6 |

| Memory Bus | 128 bit | 96 bit |

| Memory Bandwidth | 32.03 GB/s | 186.0 GB/s |

| Shading Units | 640 | 1024 |

| TMUs | 40 | 64 |

| ROPs | 16 | 32 |

| RT Cores | — | 8 |

| Pixel Rate | 14.43 GPixel/s | 65.60 GPixel/s |

| Texture Rate | 36.08 GTexel/s | 131.2 GTexel/s |

| FP32 Performance | 1,154.6 GFLOPS | 4.198 TFLOPS |

| FP16 Performance | — | 8.397 TFLOPS (2:1) |

| TDP | 45 W | 75 W |

| Slot Width | IGP | Dual-slot |

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

| Suggested PSU | — | 250 W |

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

| Display Outputs | Portable Device Dependent | 1x HDMI 2.13x DisplayPort 2.0 |

| DirectX Support | 12 (11_0) | 12 Ultimate (12_2) |

| Release Date | 2014-03-11 | 2022-06-13 |

| Predecessor | GeForce 700M | Xe Graphics |

| Successor | GeForce 900M | Battlemage |