Intel Arc B390 vs NVIDIA GeForce RTX 5070 Mobile Comparison

Where Each One Wins

The recorded benchmark data splits cleanly between these two mobile graphics solutions, though the comparison is heavily skewed by the available test coverage. The Intel Arc B390 has exactly one benchmark entry in the database: a 3DMark Steel Nomad DX12 result of 1482 points. The NVIDIA GeForce RTX 5070 Mobile has nine entries spanning OpenCL, Vulkan, PassMark DirectX 9 through 12, 2D, 3D, and compute workloads.

The RTX 5070 Mobile wins every category where both have comparable data, but direct head-to-head comparisons are limited because the test suites overlap only partially. The Arc B390's sole result places it in the 9th percentile among all GPUs in the database, while the RTX 5070 Mobile sits in the 75th percentile. That percentile gap of 66 points represents a substantial positioning difference: the Intel part ranks alongside legacy NVIDIA entries like the GT 520MX and GT 710, while the NVIDIA part competes with desktop-class cards from the previous generation.

For compute-oriented workloads, the RTX 5070 Mobile's Geekbench OpenCL score of 122238 and Vulkan score of 116960 demonstrate strong general-purpose throughput. Its PassMark GPU Compute score of 8279 reinforces this pattern. The Arc B390 has no equivalent compute benchmark in the database, so any compute comparison must rely on architectural inference rather than measured results.

In legacy DirectX workloads, the RTX 5070 Mobile shows an unusual pattern: DirectX 9 scores 214, DirectX 10 scores 129, DirectX 11 scores 192, and DirectX 12 scores 93. The DirectX 12 result being the lowest among the four is notable, though the Arc B390's only benchmark is a DirectX 12 test. The Intel part's 1482 points in 3DMark Steel Nomad cannot be directly compared to PassMark scores, as they use different methodologies and scales.

The RTX 5070 Mobile also records a PassMark G3D score of 20355 and a PassMark G2D score of 896, providing a fuller picture of its rasterization and 2D capabilities. The Arc B390 lacks equivalent entries, meaning its 2D and general 3D performance outside the single 3DMark test remains unmeasured in this database.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. The Intel Arc B390 uses the Panther Lake chip built on Xe3-LPG architecture, manufactured on a 3 nm process at Intel's own foundry. The NVIDIA GeForce RTX 5070 Mobile uses the GB206 chip on Blackwell 2.0 architecture, produced on a 5 nm process at TSMC. The process node difference gives Intel a manufacturing advantage in transistor density potential, though the database does not list transistor counts for the Arc B390, making direct density comparisons impossible.

The RTX 5070 Mobile's transistor budget is documented at 21,900 million on a 181 mm² die, yielding a density of 121.0M per mm². The Arc B390's transistor count and die size are both listed as unknown, so no direct comparison of silicon complexity can be made from recorded data.

Memory architecture differs completely. The Arc B390 uses system shared memory with system-dependent bandwidth, meaning its performance scales with the host laptop's memory configuration. The RTX 5070 Mobile carries 8 GB of dedicated GDDR7 memory on a 128-bit bus, delivering 384.0 GB/s of bandwidth. This dedicated memory arrangement provides predictable performance independent of system RAM, whereas the Intel solution's bandwidth is variable.

Core configuration shows a significant resource gap. The Arc B390 has 1536 shading units, 48 texture mapping units, 24 raster output units, and 12 ray tracing cores. The RTX 5070 Mobile more than doubles these counts: 4608 shading units, 144 TMUs, 48 ROPs, and 36 RT cores. The NVIDIA part also includes 144 tensor cores, while the Arc B390 lists no tensor core count.

Clock behavior differs substantially. The Arc B390 has a 300 MHz base clock boosting to 2500 MHz, a wide boost range suggesting aggressive power scaling. The RTX 5070 Mobile runs at a 907 MHz base and 1425 MHz boost, a much narrower range with a lower peak. Despite lower clocks, the RTX 5070 Mobile achieves higher throughput due to its larger core count.

Power envelopes are inverted relative to performance. The Arc B390 has an 80 W TDP, while the RTX 5070 Mobile is rated at 50 W. The lower-power NVIDIA part delivers higher benchmark scores, indicating superior performance-per-watt in the recorded data.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries between the Arc B390 and RTX 5070 Mobile. The comparison must therefore be assembled from their respective benchmark suites and nearest rival relationships.

The Arc B390's 3DMark Steel Nomad score of 1482 sits just 1.3% above the NVIDIA GeForce GT 520MX's average of 1463, 1.5% above the GeForce 800M's 1460, 2.5% above the GT 625 OEM's 1446, and 2.7% above the GT 710's 1443. These deltas show the Intel part clustered tightly with decade-old entry-level NVIDIA mobile parts, all within a 3% performance band.

The RTX 5070 Mobile's average benchmark score of 29928 places it essentially level with the NVIDIA GeForce RTX 3070 Ti, which averages 29945, a delta of -0.1%. It sits 0.5% above the RTX 2080 Ti's 29783, 0.6% below the AMD Radeon RX 6800's 30095, and 1.7% below the RX 6700's 30433. These deltas indicate the mobile RTX 5070 performs on par with high-end desktop GPUs from the Ampere and RDNA2 generations.

Scaling from the rival comparisons, the gap between the Arc B390 and RTX 5070 Mobile is enormous. The Intel part competes with GPUs scoring around 1450 in its benchmark, while the NVIDIA part matches GPUs scoring around 30000. The nearest rival clusters are separated by roughly 20x in average score, though the different benchmark suites prevent a precise multiplicative comparison.

The RTX 5070 Mobile's PassMark DirectX results show workload-specific performance. Its DirectX 9 score of 214 is the highest of its DX tests, followed by DirectX 11 at 192, DirectX 10 at 129, and DirectX 12 at 93. This pattern suggests the architecture handles older APIs more efficiently relative to its own capabilities, though absolute performance across all these tests is strong compared to the Arc B390's single DX12 data point.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce RTX 5070 Mobile has an average benchmark score of 29928, while the Intel Arc B390 has an average benchmark score of 1482.

Q: How does the Arc B390 compare to its nearest rivals?

A: The Arc B390 scores 1.3% above the GeForce GT 520MX, 1.5% above the GeForce 800M, 2.5% above the GT 625 OEM, and 2.7% above the GT 710.

Q: What memory configurations do these GPUs use?

A: The Arc B390 uses system shared memory with system-dependent bandwidth. The RTX 5070 Mobile has 8 GB of GDDR7 memory on a 128-bit bus with 384.0 GB/s bandwidth.

Q: Which GPU has more shading units?

A: The RTX 5070 Mobile has 4608 shading units, three times the Arc B390's 1536 shading units.

Q: What are the TDP ratings for each GPU?

A: The Arc B390 has an 80 W TDP, while the RTX 5070 Mobile has a 50 W TDP.

Q: How does the RTX 5070 Mobile rank against desktop GPUs?

A: It scores 0.1% below the RTX 3070 Ti, 0.5% above the RTX 2080 Ti, 0.6% below the RX 6800, and 1.7% below the RX 6700.

The Verdict

The recorded data presents a stark performance hierarchy. The RTX 5070 Mobile operates in the 75th percentile of all GPUs, matching desktop cards like the RTX 3070 Ti and RX 6800 within a 1% margin. It delivers 13.13 TFLOPS of FP32 performance and 384.0 GB/s of dedicated memory bandwidth, all within a 50 W power envelope.

The Arc B390, by contrast, sits in the 9th percentile, clustering with GPUs from over a decade ago. Its 7.680 TFLOPS of FP32 performance and system-shared memory place it in an entirely different performance class. The 80 W TDP is higher than the RTX 5070 Mobile's 50 W, yet the Intel part delivers roughly 42% lower FP32 throughput.

For applications requiring dedicated graphics memory, the RTX 5070 Mobile is the only viable option of the two, as the Arc B390 depends entirely on system memory with variable bandwidth. For ray tracing workloads, the RTX 5070 Mobile's 36 RT cores dwarf the Arc B390's 12. For tensor-based workloads, the RTX 5070 Mobile includes 144 tensor cores while the Arc B390 lists none.

The data indicates the RTX 5070 Mobile is positioned for demanding gaming and compute workloads, while the Arc B390 is positioned as an integrated-class solution with modest expectations. Users requiring high frame rates, large texture budgets, or compute acceleration should select the RTX 5070 Mobile based on the recorded benchmark evidence.

Specification Differences

| Specification | Intel Arc B390 | NVIDIA GeForce RTX 5070 Mobile |

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

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Unknown | 21,900 million |

| Die Size | Unknown | 181 mm² |

| Transistor Density | Not listed | 121.0M / mm² |

| Base Clock | 300 MHz | 907 MHz |

| Boost Clock | 2500 MHz | 1425 MHz |

| Memory Size | System Shared | 8 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 384.0 GB/s |

| Shading Units | 1536 | 4608 |

| Texture Mapping Units | 48 | 144 |

| Raster Output Units | 24 | 48 |

| Ray Tracing Cores | 12 | 36 |

| Tensor Cores | Not listed | 144 |

| Pixel Rate | 60.00 GPixel/s | 68.40 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 205.2 GTexel/s |

| FP32 Performance | 7.680 TFLOPS | 13.13 TFLOPS |

| FP16 Performance | 15.36 TFLOPS (2:1) | 13.13 TFLOPS (1:1) |

| TDP | 80 W | 50 W |

| Bus Interface | IGP | PCIe 5.0 x16 |

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

| OpenGL Support | 4.6 | 4.6 |

| Vulkan Support | 1.4 | 1.4 |

| Release Date | 2026-01-26 | 2025-04-14 |

| Percentile vs All GPUs | 9 | 75 |

| Average Benchmark Score | 1482 | 29928 |