Intel Arc Pro B370 vs NVIDIA GeForce RTX 5080 Comparison

Where Each One Wins

The Intel Arc Pro B370 and NVIDIA GeForce RTX 5080 occupy entirely different segments of the graphics hardware spectrum, and the recorded benchmark data reflects that separation clearly. The Arc Pro B370 carries no benchmark entries in the database, meaning there are no measured scores to compare directly against the RTX 5080's extensive test results. The RTX 5080, by contrast, has ten recorded benchmarks spanning DirectX 9 through DirectX 12, OpenCL, Vulkan, and compute workloads, with an average benchmark score of 56083.

The RTX 5080 wins every measurable category, but the nature of those wins varies by workload. In synthetic 3D Mark Steel Nomad DX12, the card posts 8637 points. In Geekbench OpenCL it scores 235901, and in Geekbench Vulkan it reaches 255450. Passmark results show 36565 in G3D, 21789 in GPU compute, 1415 in G2D, and API-specific scores of 389 in DirectX 9, 324 in DirectX 11, 208 in DirectX 10, and 151 in DirectX 12. The spread between the highest and lowest Passmark API scores is substantial, with DirectX 9 nearly 2.6 times the DirectX 12 result, indicating that the architecture's raw throughput does not translate uniformly across legacy and modern rendering paths.

The Arc Pro B370, with no benchmarks in the database, cannot be positioned on any performance curve. Its percentile rank of 50 places it at the median of all GPUs in the database, but that rank is based on an average benchmark score of zero, which reflects the absence of recorded data rather than measured performance. The RTX 5080 sits at the 87th percentile, a position supported by its average score of 56083. The delta between the 50th and 87th percentiles represents the gap between a GPU with no recorded results and one that outperforms 87 percent of all tracked graphics hardware.

Architecture Differences

The two GPUs diverge fundamentally at every architectural layer. The Arc Pro B370 uses Intel's Panther Lake chip with Xe3-LPG architecture, built on a 3 nm process at Intel's foundry. The RTX 5080 uses NVIDIA's GB203 chip with Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC. Transistor counts and die sizes are unknown for the Intel part, while the RTX 5080 carries 45,600 million transistors across a 378 mm² die, for a density of 120.6 million transistors per square millimeter.

Clock behavior differs sharply. The Arc Pro B370 runs a 300 MHz base clock and boosts to 2400 MHz. The RTX 5080 runs a 2295 MHz base clock and boosts to 2617 MHz, with memory clocked at 1875 MHz or 30 Gbps effective. The Intel part uses system shared memory, with bandwidth described as system dependent, while the RTX 5080 uses 16 GB of GDDR7 on a 256-bit bus delivering 960.0 GB/s of bandwidth. This memory configuration difference alone separates the two in scenarios where bandwidth dominates.

Compute resources show a wide gap. The Arc Pro B370 has 1280 shading units, 40 texture mapping units, 20 raster output units, and 10 ray tracing cores. The RTX 5080 has 10752 shading units, 336 TMUs, 112 ROPs, 84 RT cores, and 336 tensor cores. Pixel rate for the Intel part is 48.00 GPixel/s, versus 293.1 GPixel/s for NVIDIA. Texture rate is 96.00 GTexel/s versus 879.3 GTexel/s. FP32 throughput is 6.144 TFLOPS versus 56.28 TFLOPS. FP16 performance is 12.29 TFLOPS at a 2:1 ratio for Intel, and 56.28 TFLOPS at 1:1 for NVIDIA, meaning the RTX 5080 does not gain a ratio advantage in half-precision workloads but still delivers over 4.5 times the raw FP16 throughput.

Power and physical characteristics reinforce the positioning. The Arc Pro B370 is an integrated graphics processor with a 25 W TDP, no power connectors, IGP bus interface, and slot width listed as IGP. The RTX 5080 is a dual-slot discrete card with a 360 W TDP, one 16-pin power connector, a suggested 750 W PSU, PCIe 5.0 x16 interface, and dimensions of 304 mm length, 137 mm height, and 40 mm width. Display outputs for the Intel part are portable device dependent, while the RTX 5080 provides 1x HDMI 2.1b and 3x DisplayPort 2.1b. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity exists despite the enormous hardware gap.

Release timing places the Arc Pro B370 at 2026-01-26, with the RTX 5080 released nearly a year earlier on 2025-01-29. The Intel part succeeds HD Graphics-WM, while the RTX 5080 succeeds GeForce 40 and has a successor listed as GeForce 60. The production status for both is Active.

The Verdict

The data supports only one conclusion for performance: the RTX 5080 is the substantially faster GPU across every recorded benchmark. Its average score of 56083 places it at the 87th percentile, and its nearest rivals in the database are AMD Radeon 8060S at 55757 (0.6 percent behind), AMD Radeon RX 6750 GRE 12 GB at 55698 (0.7 percent behind), AMD Radeon Pro W5700X at 54828 (2.3 percent behind), and AMD Radeon RX 9070 GRE at 57367 (2.2 percent ahead). These deltas show that the RTX 5080 sits in a tightly contested cluster, with the largest measured gap among its nearest rivals being just 2.3 percent either way.

The Arc Pro B370 has no recorded benchmarks, so any performance comparison between the two would be speculative. The architecture differences, however, are stark enough to indicate the intended use cases. The Arc Pro B370 is an integrated part with a 25 W TDP, no discrete memory, and no power connectors, designed for portable devices where the display outputs are device dependent. The RTX 5080 is a 360 W discrete card with dedicated GDDR7 memory, a 16-pin power connector, and a suggested 750 W PSU, built for high-end desktop systems with PCIe 5.0 x16 bandwidth.

Users with a system that can physically accommodate the RTX 5080, including its 304 mm length and dual-slot width, and who need maximum recorded performance, should select that card based on the benchmark data. Users constrained to integrated graphics with minimal power draw, such as portable devices or compact systems without a PCIe slot for a discrete card, will use the Arc Pro B370 by necessity. The RTX 5080's launch MSRP is 999 USD, listed once here as reference. No pricing data exists for the Arc Pro B370 in the database.

The RTX 5080's nearest rivals in the database, all AMD parts, sit within a narrow band of 2.3 percent of its average score. This indicates that the RTX 5080 does not dominate its immediate competitive set by a wide margin, even though it sits at the 87th percentile overall. The Arc Pro B370's 50th percentile rank, based on zero recorded average score, cannot be interpreted as competitive positioning.

FAQ

Q: Does the Intel Arc Pro B370 have any recorded benchmark scores?

A: No. The database contains no benchmark entries for the Arc Pro B370, and its average benchmark score is recorded as 0.

Q: What is the RTX 5080's average benchmark score and percentile rank?

A: The RTX 5080 has an average benchmark score of 56083 and sits at the 87th percentile among all GPUs in the database.

Q: How does the RTX 5080 compare to its nearest rivals in the database?

A: The nearest rivals are AMD Radeon 8060S at 55757 (0.6 percent behind), AMD Radeon RX 6750 GRE 12 GB at 55698 (0.7 percent behind), AMD Radeon Pro W5700X at 54828 (2.3 percent behind), and AMD Radeon RX 9070 GRE at 57367 (2.2 percent ahead).

Q: What memory configurations do the two GPUs use?

A: The Arc Pro B370 uses system shared memory with system dependent bandwidth. The RTX 5080 uses 16 GB of GDDR7 on a 256-bit bus with 960.0 GB/s bandwidth.

Q: What are the power requirements for each GPU?

A: The Arc Pro B370 has a 25 W TDP with no power connectors. The RTX 5080 has a 360 W TDP with one 16-pin power connector and a suggested 750 W PSU.

Q: Which GPU supports ray tracing hardware?

A: Both support DirectX 12 Ultimate (12_2). The Arc Pro B370 has 10 ray tracing cores, while the RTX 5080 has 84 RT cores.

Head-to-Head Benchmarks

No head-to-head benchmark entries exist in the database for these two GPUs, and the Arc Pro B370 has zero recorded benchmarks. The comparison therefore relies entirely on the RTX 5080's individual results and the architectural specifications of both parts.

The RTX 5080's Passmark G3D score of 36565 represents its strongest synthetic gaming-oriented result among the Passmark suite. Its Passmark GPU compute score of 21789 indicates substantial compute throughput. The Geekbench OpenCL score of 235901 and Vulkan score of 255450 show that the card performs well in both compute APIs, with Vulkan slightly ahead of OpenCL by about 8.3 percent. The 3D Mark Steel Nomad DX12 score of 8637 provides a modern DX12 workload result.

The Passmark API-specific results reveal an interesting pattern. DirectX 9 scores 389, DirectX 11 scores 324, DirectX 10 scores 208, and DirectX 12 scores 151. The DirectX 9 score is 2.6 times the DirectX 12 score, and DirectX 11 is 2.1 times DirectX 12. This ordering suggests that the Blackwell 2.0 architecture's DX12 path, while feature-rich, does not produce the highest raw Passmark scores in that API. The G2D score of 1415 reflects 2D graphics performance that is independent of the 3D API results.

The Arc Pro B370's theoretical output figures, while unverified by benchmarks, indicate a part operating in a different class. Its 6.144 TFLOPS FP32 is roughly 9.2 percent of the RTX 5080's 56.28 TFLOPS. Its 48.00 GPixel/s is about 16.4 percent of the RTX 5080's 293.1 GPixel/s. Its 96.00 GTexel/s is about 10.9 percent of the RTX 5080's 879.3 GTexel/s. These ratios, derived strictly from the recorded specifications, show that even the theoretical peak of the Intel part falls far below the NVIDIA card's measured capabilities.

The memory bandwidth comparison is equally lopsided. The RTX 5080's 960.0 GB/s is a fixed figure, while the Arc Pro B370's bandwidth is system dependent, meaning it shares memory with the CPU and its bandwidth varies by platform. The RTX 5080's 16 GB GDDR7 on a 256-bit bus provides dedicated capacity and deterministic bandwidth, whereas the Intel part has no dedicated VRAM.

The RTX 5080's nearest rivals in the database, all within 2.3 percent of its average score, suggest that its competitive position is strong but not dominant. The AMD Radeon RX 9070 GRE leads the RTX 5080 by 2.2 percent, while the AMD Radeon 8060S trails by 0.6 percent. These margins are small enough that workload-specific results could shift the ordering. The Arc Pro B370, with no benchmark data, cannot be placed on this competitive curve at all.