GPU Comparison

The benchmark data presents a stark generational and performance-class divide. The NVIDIA GeForce GTX 1050, a 2016 entry-level Pascal card, is pitted against the Intel Arc Pro B60, a 2025 professional workstation GPU built on the Xe2-HPG architecture. Across every single head-to-head benchmark recorded, the Intel Arc Pro B60 emerges victorious, but the magnitude of those wins and the context of each card's intended workload reveal a more nuanced story than a simple tally of victories.

Where Each One Wins

The data is unambiguous: the Intel Arc Pro B60 wins in every workload category measured. The head-to-head table shows a clean sweep, with the Intel card taking all 8 benchmark comparisons. However, the nature of those wins tells a tale of specialization versus legacy compatibility.

The GTX 1050’s only "wins" are not in performance, but in its relative standing among its own peers. Its average benchmark score of 3629 places it in the 21st percentile of all GPUs, with its closest rival being the NVIDIA GeForce GT 735M (avg score 3616, within 0.4%). This indicates the GTX 1050 is a card that was competitive in its own era, sitting just above the GeForce GT 545 (avg 3594, 1% slower). It is a legacy part, now end-of-life, whose performance profile is anchored to the DirectX 9 era where it scores 83 in the passmark test, its highest DirectX score.

The Arc Pro B60, in contrast, is a modern compute and rendering monster. Its 3DMark Steel Nomad DX12 score of 2646 is a massive leap, and its Passmark G3D score of 14580 is nearly triple that of the GTX 1050's 5028. The B60's strength lies in modern, heavily threaded workloads. Its Passmark GPU Compute score of 7029 is over three times higher than the GTX 1050's 2091, highlighting a clear advantage for professional applications, AI inference, and any task that leverages raw shader throughput. The B60 wins where the GTX 1050 cannot compete: modern API performance, compute, and raw rasterization power.

Architecture Differences

The architectural gap between these two GPUs is the primary driver of the performance chasm. The GTX 1050 is built on NVIDIA's Pascal architecture, using a 14 nm process at Samsung's foundry. It packs 3,300 million transistors onto a 132 mm² die, resulting in a transistor density of 25.0M / mm². The Intel Arc Pro B60 is a fundamentally newer design, built on the Xe2-HPG architecture, fabricated on a 5 nm process at TSMC. It integrates 19,600 million transistors on a 272 mm² die, achieving a much higher density of 72.1M / mm². This represents a generational shift in manufacturing capability and design efficiency.

The core configurations are equally divergent. The GTX 1050 features 640 shading units, 40 texture mapping units (TMUs), and 32 raster output units (ROPs). It has no dedicated ray tracing or tensor cores, reflecting its 2016 design. The Arc Pro B60, by contrast, is equipped with 2560 shading units, 160 TMUs, and 80 ROPs. It also includes 20 dedicated ray tracing cores, a feature entirely absent from the older NVIDIA card. This hardware difference explains the B60's support for DirectX 12 Ultimate (12_2), while the GTX 1050 is limited to DirectX 12 (12_1). The B60's FP32 throughput is rated at 12.29 TFLOPS versus the GTX 1050's 1.862 TFLOPS, a 6.6x difference in raw compute potential.

Memory architecture also tells a story of different targets. The GTX 1050 uses 2 GB of GDDR5 on a 128-bit bus, delivering 112.1 GB/s of bandwidth. The Arc Pro B60 is equipped with 24 GB of GDDR6 on a 192-bit bus, providing a massive 456.0 GB/s of bandwidth. This 24 GB framebuffer is clearly aimed at professional visualization, large datasets, and AI model loading, whereas the GTX 1050's 2 GB is sufficient only for older games and basic desktop tasks at its intended resolutions.

Head-to-Head Benchmarks

The data from the head-to-head comparisons shows a consistent and often dramatic margin of victory for the Intel Arc Pro B60. The largest relative win comes in the 3DMark Steel Nomad DX12 test. Here, the B60 scores 2646 against the GTX 1050's 122, a delta of -95.4% (meaning the GTX 1050 is 95.4% slower). This is a modern, demanding test, and the result shows the GTX 1050 is simply not a viable option for contemporary DirectX 12 gaming or rendering.

In the Passmark suite, the B60's advantage is consistent but varies by API generation. In the legacy DirectX 9 test, the B60 scores 179 versus the GTX 1050's 83, a -53.6% delta. This is the closest margin, suggesting that the GTX 1050's architecture was relatively stronger in older, less complex workloads. However, the gap widens significantly with newer APIs. In DirectX 10, the B60 scores 61 versus 24 (-60.7%); in DirectX 11, it scores 122 versus 38 (-68.9%); and in DirectX 12, it scores 76 versus 20 (-73.7%). This trend indicates that the Intel card's performance scales much better with modern, multi-threaded graphics APIs, while the older NVIDIA card falls further behind as driver overhead and feature requirements increase.

The compute and general performance metrics reinforce this narrative. In Passmark G3D (a general 3D graphics benchmark), the B60 scores 14580 versus 5028, a -65.5% delta. In Passmark GPU Compute, the B60 scores 7029 versus 2091, a -70.3% delta. Even in the 2D test, the B60 wins with 763 versus 457, a -40.1% delta, showing an advantage in basic desktop and interface rendering. The only benchmark where the GTX 1050 has a competitive showing is in its Geekbench scores, but those are not part of the head-to-head table. The data is clear: the B60 is a comprehensively faster and more capable GPU.

FAQ

Q: Which GPU is faster in 3DMark Steel Nomad DX12?

A: The Intel Arc Pro B60 is decisively faster, scoring 2646 compared to the GTX 1050's 122. The delta is -95.4%, indicating the GTX 1050 is over 20 times slower in this modern test.

Q: Does the GTX 1050 win any benchmark?

A: No. In the provided head-to-head data, the Intel Arc Pro B60 wins all 8 recorded benchmarks. The GTX 1050's only advantage is its relative standing among its own peers, where it performs close to other older cards like the GeForce GT 735M.

Q: What is the memory size difference?

A: The Intel Arc Pro B60 has 24 GB of GDDR6 memory, while the NVIDIA GeForce GTX 1050 has 2 GB of GDDR5. The B60's memory is also on a wider 192-bit bus, providing 456.0 GB/s of bandwidth versus 112.1 GB/s.

Q: Which card supports ray tracing?

A: Only the Intel Arc Pro B60 has dedicated ray tracing cores (20 of them). The NVIDIA GeForce GTX 1050 has no ray tracing cores, which is consistent with its older Pascal architecture.

Q: How do their compute scores compare?

A: The Intel Arc Pro B60 scores 7029 in Passmark GPU Compute, while the GTX 1050 scores 2091. The B60 is roughly 3.4 times faster in this compute-focused test.

Q: What is the difference in their FP32 performance?

A: The Intel Arc Pro B60 is rated at 12.29 TFLOPS FP32, whereas the NVIDIA GeForce GTX 1050 is rated at 1.862 TFLOPS. This represents a significant raw compute advantage for the Intel card.

The Verdict

The data-driven verdict is unequivocal for performance: the Intel Arc Pro B60 is the superior product in every measurable way. Its benchmark scores are consistently and dramatically higher, its memory subsystem is vastly more capable, and its architecture supports modern features like ray tracing that the GTX 1050 lacks entirely. The B60's 24 GB of memory and higher compute throughput position it for professional workloads, 3D rendering, video editing, and machine learning inference, where the GTX 1050 would fail outright.

The NVIDIA GeForce GTX 1050, however, is not without a purpose according to the data. It is an end-of-life product with a 21st percentile ranking, meaning it is slower than 79% of all GPUs in the database. Its performance is only comparable to other legacy mobile or entry-level desktop parts from its era. It should only be considered for legacy systems, basic 2D desktop use, or very old games from its DirectX 9/10 era, where its performance deficit is smallest. For any modern application, the choice is clear.

Specification Differences

| Specification | NVIDIA GeForce GTX 1050 | Intel Arc Pro B60 |

| :--- | :--- | :--- |

| Architecture | Pascal | Xe2-HPG |

| Process Node | 14 nm | 5 nm |

| Transistors | 3,300 million | 19,600 million |

| Die Size | 132 mm² | 272 mm² |

| Base Clock | 1354 MHz | 2000 MHz |

| Boost Clock | 1455 MHz | 2400 MHz |

| Memory Size | 2 GB | 24 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus Width | 128 bit | 192 bit |

| Memory Bandwidth | 112.1 GB/s | 456.0 GB/s |

| Shading Units | 640 | 2560 |

| TMUs | 40 | 160 |

| ROPs | 32 | 80 |

| RT Cores | None | 20 |

| Pixel Rate | 46.56 GPixel/s | 192.0 GPixel/s |

| Texture Rate | 58.20 GTexel/s | 384.0 GTexel/s |

| FP32 Performance | 1.862 TFLOPS | 12.29 TFLOPS |

| FP16 Performance | 29.10 GFLOPS (1:64) | 24.58 TFLOPS (2:1) |

| TDP | 75 W | 200 W |

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

| Suggested PSU | 250 W | 550 W |

| Bus Interface | PCIe 3.0 x16 | PCIe 5.0 x8 |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a | 4x mini-DisplayPort 2.1 |

| DirectX Support | 12 (12_1) | 12 Ultimate (12_2) |

| Length | 145 mm | 167 mm |

| Height | 111 mm | 69 mm |

| Production Status | End-of-life | Active |