Intel Arc A380 vs NVIDIA GeForce GTX 650 Ti Comparison

# FAQ

Q: Which GPU is faster in compute workloads according to the benchmark data?

A: The Intel Arc A380 dominates in compute. In Geekbench OpenCL, it scores 38,224 versus the GTX 650 Ti's 7,877, a 385.3% advantage. In Geekbench Vulkan, the Arc A380 posts 36,736 against 8,229, a 346.4% lead.

Q: How do the two cards compare in overall average benchmark score?

A: The Arc A380 has an average benchmark score of 8,558, while the GTX 650 Ti averages 8,053. The Arc A380 sits at the 44th percentile of all GPUs, and the GTX 650 Ti is at the 42nd percentile.

Q: What are the nearest rivals for each card based on average score?

A: For the Arc A380, the closest rival is the AMD FirePro W5170M (8,595, -0.4% delta), followed by the AMD Radeon HD 8870M (8,462, +1.1% delta). For the GTX 650 Ti, the nearest rivals are the NVIDIA GRID K2 (8,080, -0.3% delta) and the GTX 650 Ti Boost (8,067, -0.2% delta).

Q: Do both GPUs support the same DirectX version?

A: No. The Arc A380 supports DirectX 12 Ultimate (12_2), while the GTX 650 Ti supports only DirectX 12 (11_0). This is a significant API-level difference.

Q: Which card has more memory and bandwidth?

A: The Arc A380 has 6 GB of GDDR6 memory on a 96-bit bus, delivering 186.0 GB/s bandwidth. The GTX 650 Ti has 1024 MB of GDDR5 on a 128-bit bus, delivering 86.40 GB/s.

Q: What are the launch MSRPs of the two cards?

A: Both cards launched at a launch MSRP of 149 USD. The Arc A380 was released in June 2022, while the GTX 650 Ti launched in October 2012.

# Architecture Differences

The Intel Arc A380 is built on the DG2-128 chip using the Xe-HPG architecture, part of the Alchemist generation (Arc 3). It is fabricated on a 6 nm process at TSMC, with 7,200 million transistors packed into a 157 mm² die. That yields a transistor density of 45.9M per mm².

The NVIDIA GeForce GTX 650 Ti uses the GK106S chip with the Kepler architecture, part of the GeForce 600 generation. It is built on a 28 nm process at TSMC, with 2,540 million transistors on a larger 221 mm² die, resulting in a much lower transistor density of 11.5M per mm².

The Arc A380 features 1,024 shading units, 64 TMUs, and 32 ROPs. It also includes 8 dedicated ray tracing cores, a feature entirely absent from the GTX 650 Ti. The GTX 650 Ti has 768 shading units, 64 TMUs, and 16 ROPs, with no ray tracing capability.

Memory architecture differs substantially. The Arc A380 uses 6 GB of GDDR6 across a 96-bit bus, achieving 186.0 GB/s bandwidth. The GTX 650 Ti uses 1024 MB of GDDR5 on a 128-bit bus, achieving only 86.40 GB/s. The Arc A380's memory clock runs at 1937 MHz (15.5 Gbps effective), while the GTX 650 Ti's runs at 1350 MHz (5.4 Gbps effective).

Process technology gives the Arc A380 a massive efficiency advantage. Its 6 nm node allows a TDP of just 75 W, while the GTX 650 Ti draws 110 W despite being far slower. The Arc A380 requires a dual-slot cooler and a 1x 8-pin power connector, with a suggested PSU of 250 W. The GTX 650 Ti is a single-slot card with a 1x 6-pin connector and a suggested PSU of 300 W.

API support is another generation gap. The Arc A380 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GTX 650 Ti supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. Display outputs also differ: the Arc A380 has 1x HDMI 2.1 and 3x DisplayPort 2.0, while the GTX 650 Ti has 2x DVI and 1x mini-HDMI 1.4a.

# Head-to-Head Benchmarks

The head-to-head data contains two compute-focused benchmarks, and the Intel Arc A380 wins both decisively. In Geekbench OpenCL, the Arc A380 scores 38,224 versus the GTX 650 Ti's 7,877. That represents a 385.3% delta — the Arc A380 delivers nearly five times the OpenCL compute performance.

In Geekbench Vulkan, the Arc A380 scores 36,736 against 8,229 for the GTX 650 Ti. The delta is 346.4%, meaning the Arc A380 is about 4.5 times faster. Both benchmarks tell the same story: the architectural leap from Kepler to Xe-HPG is enormous in raw compute throughput.

The FP32 compute figures corroborate the benchmark results. The Arc A380 delivers 4.198 TFLOPS of FP32 performance, while the GTX 650 Ti manages only 1,425.4 GFLOPS. The Arc A380's pixel rate is 65.60 GPixel/s versus 14.85 GPixel/s for the GTX 650 Ti, and texture rates are 131.2 GTexel/s versus 59.39 GTexel/s.

The GTX 650 Ti has no wins in the head-to-head comparison. The score is 2-0 in favor of the Arc A380. Even in average benchmark score, the Arc A380 leads 8,558 to 8,053. The percentile ranking confirms the gap: 44th versus 42nd percentile across all GPUs.

It is worth remembering the GTX 650 Ti lacks any Passmark benchmark results in the data, while the Arc A380 has Passmark DirectX 9 (73), DirectX 10 (37), DirectX 11 (38), DirectX 12 (35), G2D (610), G3D (6252), and GPU compute (2762) scores. This suggests the GTX 650 Ti may not have been tested under those workloads, but the available data still favors the Arc A380 across every comparable metric.

# Specification Differences

| Specification | Intel Arc A380 | NVIDIA GeForce GTX 650 Ti |

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

| Architecture | Xe-HPG | Kepler |

| Generation | Alchemist (Arc 3) | GeForce 600 |

| Process Node | 6 nm | 28 nm |

| Transistors | 7,200 million | 2,540 million |

| Die Size | 157 mm² | 221 mm² |

| Transistor Density | 45.9M / mm² | 11.5M / mm² |

| Base Clock | 2000 MHz | N/A |

| Boost Clock | 2050 MHz | N/A |

| Memory Clock | 1937 MHz / 15.5 Gbps effective | 1350 MHz / 5.4 Gbps effective |

| Memory Size | 6 GB | 1024 MB |

| Memory Type | GDDR6 | GDDR5 |

| Memory Bus | 96 bit | 128 bit |

| Memory Bandwidth | 186.0 GB/s | 86.40 GB/s |

| Shading Units | 1024 | 768 |

| TMUs | 64 | 64 |

| ROPs | 32 | 16 |

| RT Cores | 8 | N/A |

| Pixel Rate | 65.60 GPixel/s | 14.85 GPixel/s |

| Texture Rate | 131.2 GTexel/s | 59.39 GTexel/s |

| FP32 | 4.198 TFLOPS | 1,425.4 GFLOPS |

| FP16 | 8.397 TFLOPS (2:1) | N/A |

| TDP | 75 W | 110 W |

| Slot Width | Dual-slot | Single-slot |

| Power Connectors | 1x 8-pin | 1x 6-pin |

| Suggested PSU | 250 W | 300 W |

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

| Display Outputs | 1x HDMI 2.1, 3x DisplayPort 2.0 | 2x DVI, 1x mini-HDMI 1.4a |

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

| Vulkan | 1.4 | 1.2.175 |

| Release Date | 2022-06-13 | 2012-10-08 |

| Length | 222 mm (8.7 inches) | 145 mm (5.7 inches) |

# The Verdict

The benchmark data is unambiguous: the Intel Arc A380 is the vastly superior GPU. It wins both head-to-head benchmarks by margins exceeding 345%, and its average benchmark score of 8,558 is 6.3% higher than the GTX 650 Ti's 8,053. The Arc A380 also ranks higher in the overall percentile distribution at 44th versus 42nd.

The architectural differences explain the performance gap. The Arc A380 benefits from a 6 nm process, 7,200 million transistors, and a modern Xe-HPG design with ray tracing support and DirectX 12 Ultimate. The GTX 650 Ti uses a 28 nm process with 2,540 million transistors and a Kepler architecture that predates many modern graphics features.

Memory capacity is another decisive factor. The Arc A380 offers 6 GB of GDDR6, six times the GTX 650 Ti's 1024 MB of GDDR5. The bandwidth advantage is equally stark: 186.0 GB/s versus 86.40 GB/s. For modern workloads, the GTX 650 Ti's memory configuration is a severe bottleneck.

The Arc A380 is also more power-efficient. It draws 75 W versus 110 W for the GTX 650 Ti, despite delivering far higher performance. The suggested PSU requirement is lower for the Arc A380 (250 W versus 300 W), and it uses a more modern PCIe 4.0 x8 interface compared to the GTX 650 Ti's PCIe 3.0 x16.

However, the GTX 650 Ti is not without merit. It is a single-slot card at just 145 mm (5.7 inches) in length, making it far more compact than the Arc A380's 222 mm (8.7 inches) dual-slot design. For users with severe space constraints or older systems with limited expansion slots, the GTX 650 Ti's form factor could be a practical consideration.

# Where Each One Wins

Intel Arc A380 wins in: Compute performance (OpenCL and Vulkan benchmarks by 385.3% and 346.4% respectively), memory capacity and bandwidth (6 GB GDDR6 at 186.0 GB/s versus 1024 MB GDDR5 at 86.40 GB/s), FP32 throughput (4.198 TFLOPS versus 1,425.4 GFLOPS), pixel fill rate (65.60 GPixel/s versus 14.85 GPixel/s), texture fill rate (131.2 GTexel/s versus 59.39 GTexel/s), ray tracing (8 RT cores versus none), modern API support (DirectX 12 Ultimate and Vulkan 1.4), power efficiency (75 W versus 110 W), and overall average benchmark score (8,558 versus 8,053).

NVIDIA GeForce GTX 650 Ti wins in: Physical dimensions (145 mm length versus 222 mm, single-slot versus dual-slot), memory bus width (128-bit versus 96-bit), and it has a lower suggested PSU requirement relative to its performance class. The GTX 650 Ti also uses a PCIe 3.0 x16 interface, which may be more compatible with older motherboards.

For modern gaming and compute workloads, the Arc A380 is the clear choice. The data shows a generational leap in every measurable performance category. For users building a small-form-factor system with minimal power delivery and no need for ray tracing or high-resolution textures, the GTX 650 Ti's compact size is its only real advantage. But given that both cards launched at the same MSRP of 149 USD, the Arc A380 delivers dramatically more performance per dollar spent.