GPU Comparison

Intel Arc Pro B60 and NVIDIA GeForce 820A occupy opposite ends of the GPU spectrum, separated by over a decade of architectural evolution and a 13x gap in average benchmark scores. The Arc Pro B60, built on TSMC's 5 nm process with the Xe2-HPG architecture, delivers 12.29 TFLOPS of FP32 compute, while the Fermi 2.0-based 820A, a 28 nm part from 2014, manages just 240.0 GFLOPS. The data shows a complete generational mismatch, yet each card finds its niche based on workload and platform constraints.

Head-to-Head Benchmarks

The benchmark disparity is stark, though direct comparison is limited by the fact that the two cards share no common test results in the database. The Arc Pro B60 has eight recorded benchmarks across 3DMark and Passmark suites, while the GeForce 820A has only a single Geekbench OpenCL result. This lack of overlap means the comparison relies on aggregated averages and percentile positioning.

The Arc Pro B60's average benchmark score is 3182, placing it in the 20th percentile of all GPUs. Its strongest result comes from Passmark G3D, where it scores 14580, a figure that dwarfs its Passmark GPU Compute score of 7029. In DirectX-specific tests, the Arc shows a peculiar pattern: Passmark DirectX 9 yields 179, while DirectX 11 drops to 122, and DirectX 12 falls further to 76. The 3DMark Steel Nomad DX12 test produces 2646, and Passmark G2D scores 763. These numbers position the Arc Pro B60 within 0.6% of the NVIDIA Quadro P1000 (3163) and 3.5% ahead of the NVIDIA GeForce RTX 5080 SUPER (3075), though it trails the GeForce GT 640 by 0.9% and the GeForce 920M by 3.2%.

The GeForce 820A's solitary Geekbench OpenCL score of 2983 gives it an average of 2983, placing it in the 19th percentile. Its nearest rivals include the GeForce GTX 860M (2967, 0.5% behind), the GeForce GTX 750 Ti (2953, 1% behind), the Quadro P600 (2923, 2.1% behind), and the GeForce RTX 4060 Ti 8 GB (2913, 2.4% behind). Notably, the 820A's single score actually exceeds several higher-tier NVIDIA parts, though this reflects the specific OpenCL workload rather than general capability.

The aggregate numbers tell the real story: the Arc Pro B60's average score of 3182 is 6.7% higher than the 820A's 2983. The Arc Pro B60 also holds a 1-point percentile advantage (20th vs 19th). In raw compute terms, the Arc Pro B60's 12.29 TFLOPS FP32 is roughly 51x the 820A's 240.0 GFLOPS, and its pixel rate of 192.0 GPixel/s compares to just 2.500 GPixel/s for the older card. Texture rate similarly favors the Arc at 384.0 GTexel/s versus 10.00 GTexel/s.

FAQ

Q: Which card has the higher average benchmark score?

A: The Intel Arc Pro B60 scores 3182 on average, which is 6.7% higher than the NVIDIA GeForce 820A's 2983 average.

Q: How do the two cards compare in memory bandwidth?

A: The Arc Pro B60 offers 456.0 GB/s of bandwidth from 24 GB of GDDR6 on a 192-bit bus, while the 820A provides 16.02 GB/s from 1024 MB of DDR3 on a 64-bit bus. That is roughly a 28x difference in bandwidth.

Q: What is the production status of each GPU?

A: The Intel Arc Pro B60 is listed as Active, while the NVIDIA GeForce 820A is End-of-life. The 820A was released on 2014-03-16 and has a predecessor (GeForce 700A) and successor (GeForce 900A), whereas the Arc Pro B60 was released on 2025-09-04 with no listed predecessor or successor.

Q: Do both cards support the same DirectX version?

A: No. The Arc Pro B60 supports DirectX 12 Ultimate (12_2), while the 820A supports DirectX 12 (11_0). The Arc Pro B60 also supports Vulkan 1.4, but the 820A has no Vulkan support listed.

Q: Which card has a higher transistor density?

A: The Arc Pro B60 has a transistor density of 72.1M per mm², based on 19,600 million transistors on a 272 mm² die. The 820A has 5.0M per mm², based on 585 million transistors on a 116 mm² die.

Q: What are the power connector requirements?

A: The Arc Pro B60 requires a single 8-pin power connector and a 550 W suggested PSU, with a 200 W TDP. The 820A has no power connectors and a 15 W TDP, with no suggested PSU listed.

Architecture Differences

The architectural gap between these two GPUs spans three generations of NVIDIA technology and a fundamental shift in Intel's approach. The Arc Pro B60 uses Intel's Xe2-HPG architecture on the BMG-G21 chip, built on TSMC's 5 nm process. This is Intel's Battlemage generation for professional workstations, integrating 19,600 million transistors across a 272 mm² die. The architecture includes 20 ray tracing cores, a feature entirely absent from the 820A, and supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6.

The NVIDIA GeForce 820A uses the Fermi 2.0 architecture on the GF117 chip, fabricated on TSMC's 28 nm process. This is a mobile-oriented part with just 585 million transistors on a 116 mm² die. Fermi 2.0 predates several key modern features: it has no ray tracing cores, no tensor cores, and no Vulkan support. Its DirectX support is limited to 12 (11_0), a partial implementation that lacks many DirectX 12 features. The 820A does support OpenGL 4.6, matching the Arc Pro B60 in that regard.

The transistor counts illustrate the scale difference: the Arc Pro B60 packs 19,600 million transistors, which is 33.5x more than the 820A's 585 million. Transistor density tells a similar story with 72.1M per mm² for the Arc versus 5.0M per mm² for the 820A, a 14.4x difference. The process node advantage (5 nm vs 28 nm) enables this density and the corresponding performance leap.

Memory architectures differ fundamentally. The Arc Pro B60 uses 24 GB of GDDR6 with a 192-bit bus and 456.0 GB/s bandwidth. The 820A uses 1024 MB of DDR3 with a 64-bit bus and 16.02 GB/s bandwidth. The Arc's memory clock is 2375 MHz (19 Gbps effective), while the 820A's is 1001 MHz (2 Gbps effective). The 24x capacity and 28x bandwidth advantages of the Arc Pro B60 reflect its professional workstation positioning versus the 820A's low-power mobile origins.

Specification Differences

The specification sheets diverge on nearly every measurable field. The Arc Pro B60 has 2560 shading units, 160 TMUs, and 80 ROPs, compared to the 820A's 96 shading units, 16 TMUs, and 8 ROPs. The Arc Pro B60 includes 20 ray tracing cores; the 820A has none. FP32 compute is 12.29 TFLOPS for the Arc versus 240.0 GFLOPS for the 820A, a 51x difference. FP16 is only listed for the Arc Pro B60 at 24.58 TFLOPS (2:1 ratio).

The Arc Pro B60's pixel rate is 192.0 GPixel/s and texture rate is 384.0 GTexel/s, while the 820A manages 2.500 GPixel/s and 10.00 GTexel/s. Power consumption differs dramatically: the Arc Pro B60 has a 200 W TDP and requires a 550 W suggested PSU with a single 8-pin connector. The 820A has a 15 W TDP, no power connectors, and no suggested PSU. The Arc Pro B60 is a dual-slot card measuring 167 mm (6.6 inches) in length, 69 mm (2.7 inches) in height, and 40 mm (1.6 inches) in width. The 820A is an IGP (integrated graphics processor) with no physical dimensions listed.

Bus interfaces also differ: the Arc Pro B60 uses PCIe 5.0 x8, while the 820A uses PCIe 2.0 x16. Display outputs are 4x mini-DisplayPort 2.1 for the Arc Pro B60, while the 820A's outputs are described as "Portable Device Dependent." The Arc Pro B60 has a launch MSRP of 499 USD; the 820A has no launch MSRP listed. Release dates are 2025-09-04 for the Arc Pro B60 and 2014-03-16 for the 820A. The 820A has a predecessor (GeForce 700A) and successor (GeForce 900A), while the Arc Pro B60 has neither.

Where Each One Wins

The Arc Pro B60 wins decisively in every performance-oriented scenario. Its 12.29 TFLOPS of FP32 compute, 456.0 GB/s memory bandwidth, and 24 GB of VRAM make it suitable for professional 3D rendering, compute workloads, and modern gaming at high resolutions. The 20 ray tracing cores enable hardware-accelerated ray tracing, a feature the 820A cannot offer. Its PCIe 5.0 x8 interface and four mini-DisplayPort 2.1 outputs support high-bandwidth display configurations. The Active production status and 2025 release date mean ongoing driver support and modern API compatibility, including DirectX 12 Ultimate and Vulkan 1.4.

The GeForce 820A wins in power efficiency and platform flexibility. Its 15 W TDP requires no power connectors, making it suitable for legacy or low-power systems where the Arc Pro B60's 200 W TDP and 8-pin connector are impractical. As an IGP with "Portable Device Dependent" outputs, it targets embedded or mobile designs where space and thermal constraints dominate. The 820A's single Geekbench OpenCL score of 2983 shows it can handle basic compute tasks, and its position within 2.4% of the GeForce RTX 4060 Ti 8 GB in that specific test suggests it remains competitive for certain OpenCL workloads. The 820A's End-of-life status and 2014 release date mean it is a legacy part, but its predecessor and successor lineage shows a defined product lifecycle.

The benchmark data supports a clear split: the Arc Pro B60 is for users who need raw performance and modern features, while the 820A serves systems where power draw and physical footprint are the primary constraints. The Arc Pro B60's 6.7% average score advantage and 1-point percentile edge confirm its overall superiority, but the 820A's 15 W TDP versus 200 W TDP represents a 13x power efficiency advantage that cannot be ignored in battery-powered or passively cooled designs.