Intel Arc Pro B390 vs NVIDIA GeForce RTX 4070 Comparison

Head-to-Head Benchmarks

The recorded data does not include any direct head-to-head benchmark results between the Intel Arc Pro B390 and the NVIDIA GeForce RTX 4070. The head-to-head benchmark array is empty, and the win counts for both products are zero. Consequently, a direct per-test comparison cannot be constructed from the database.

However, the database does contain a full benchmark suite for the NVIDIA GeForce RTX 4070, which establishes a reference performance profile. The Intel Arc Pro B390 has no benchmark entries, making a numerical comparison of application-level performance impossible. The RTX 4070 records an average benchmark score of 37,648 across its suite, with a percentile rank of 81 among all GPUs. Its nearest rivals in the database include the NVIDIA Tesla P4 (average score 37,628, a 0.1% difference), the AMD Radeon RX Vega 56 (average score 37,507, a 0.4% difference), the NVIDIA GeForce RTX 4080 Mobile (average score 38,135, a 1.3% difference in favor of the mobile part), and the AMD Radeon PRO W6400 (average score 37,157, a 1.3% difference).

The RTX 4070’s individual benchmark results show strong consistency. In 3DMark Steel Nomad DX12, it scores 3,854. Geekbench OpenCL yields 154,858, and Geekbench Vulkan yields 174,152. Passmark scores include 26,927 in G3D, 14,720 in GPU Compute, 1,164 in G2D, 320 in DirectX 9, 244 in DirectX 11, 139 in DirectX 10, and 103 in DirectX 12. These figures place the RTX 4070 firmly in the upper-middle tier of the database’s GPU rankings, roughly 31 percentile points above the midpoint.

For the Intel Arc Pro B390, the database only records a percentile rank of 50 against all GPUs, with an average benchmark score of 0 due to absent test data. This percentile figure suggests the part is positioned near the median of the database, but no measured workload results substantiate that placement. The absence of benchmark entries means the Arc Pro B390 cannot be directly compared to the RTX 4070 on any recorded test. The data simply does not support a head-to-head numerical analysis, and any claim of superiority in either direction would be unsupported by measurements.

FAQ

Q: What is the average benchmark score for the NVIDIA GeForce RTX 4070?

A: The RTX 4070 records an average benchmark score of 37,648 across its database entries, with a percentile rank of 81 among all GPUs.

Q: Does the Intel Arc Pro B390 have any recorded benchmark scores?

A: No. The Arc Pro B390 has an empty benchmark array and an average benchmark score of 0. Its percentile rank is 50, but no individual test results are present in the database.

Q: Which GPU has a higher pixel rate?

A: The RTX 4070 has a pixel rate of 158.4 GPixel/s, whereas the Arc Pro B390 has a pixel rate of 60.00 GPixel/s. The RTX 4070 is 2.64 times higher in this metric.

Q: How do the texture rates compare?

A: The RTX 4070 delivers 455.4 GTexel/s, while the Arc Pro B390 delivers 120.0 GTexel/s. The RTX 4070 is approximately 3.8 times higher.

Q: What is the difference in floating-point performance (FP32)?

A: The RTX 4070 achieves 29.15 TFLOPS FP32, while the Arc Pro B390 achieves 7.680 TFLOPS FP32. The RTX 4070 is roughly 3.8 times higher in raw FP32 throughput.

Q: What are the production statuses of the two GPUs?

A: The Arc Pro B390 is listed as Active, while the RTX 4070 is listed as End-of-life. The RTX 4070 has a successor (GeForce 50), while the Arc Pro B390 has no successor listed.

Architecture Differences

The two GPUs use fundamentally different architectures and process technologies. The Intel Arc Pro B390 is built on the Xe3-LPG architecture using Intel’s own 3 nm process node and is part of the Arc Graphics-WM (Panther Lake) generation. The NVIDIA GeForce RTX 4070 uses the Ada Lovelace architecture on TSMC’s 5 nm process and belongs to the GeForce 40-series. The foundry differs as well: Intel fabricates the Arc Pro B390, while TSMC fabricates the RTX 4070.

The RTX 4070’s chip, AD104, contains 35,800 million transistors on a 294 mm² die, yielding a transistor density of 121.8M per mm². The Arc Pro B390’s transistor count and die size are recorded as unknown, so a density comparison is not possible. The Intel part is an integrated GPU (IGP), meaning it shares system memory and has no dedicated memory bus. The RTX 4070 is a discrete GPU with 12 GB of GDDR6X memory on a 192-bit bus, providing 504.2 GB/s of bandwidth. The Arc Pro B390’s memory bandwidth is system-dependent, with no fixed figure.

Compute resource counts differ substantially. The Arc Pro B390 has 1,536 shading units, 48 texture mapping units, 24 raster output units, and 12 ray tracing cores. The RTX 4070 has 5,888 shading units, 184 TMUs, 64 ROPs, and 46 RT cores. The Intel part does not list tensor cores, while the RTX 4070 includes 184 tensor cores. In terms of API support, both GPUs expose DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Power and physical characteristics diverge sharply. The Arc Pro B390 has a TDP of 80 W, uses no power connectors, and is integrated into a portable device as an IGP. The RTX 4070 has a TDP of 200 W, uses a single 16-pin power connector, requires a 550 W suggested PSU, and is a dual-slot card with dimensions of 240 mm length, 110 mm height, and 40 mm width. The Intel part has no fixed dimensions or slot width beyond its integrated nature, and its display outputs are portable-device dependent. The RTX 4070 provides 1x HDMI 2.1 and 3x DisplayPort 1.4a outputs.

Specification Differences

The following specifications differ between the two products, based solely on the recorded data:

• Process node: Intel Arc Pro B390 uses 3 nm; RTX 4070 uses 5 nm.
• Foundry: Intel for the Arc Pro B390; TSMC for the RTX 4070.
• Transistors: Unknown for the Arc Pro B390; 35,800 million for the RTX 4070.
• Die size: Unknown for the Arc Pro B390; 294 mm² for the RTX 4070.
• Transistor density: Not listed for the Arc Pro B390; 121.8M per mm² for the RTX 4070.
• Base clock: 300 MHz for the Arc Pro B390; 1,920 MHz for the RTX 4070.
• Boost clock: 2,500 MHz for the Arc Pro B390; 2,475 MHz for the RTX 4070.
• Memory clock: System shared for the Arc Pro B390; 1,313 MHz (21 Gbps effective) for the RTX 4070.
• Memory size: System shared for the Arc Pro B390; 12 GB for the RTX 4070.
• Memory type: System shared for the Arc Pro B390; GDDR6X for the RTX 4070.
• Memory bus width: System shared for the Arc Pro B390; 192 bit for the RTX 4070.
• Memory bandwidth: System dependent for the Arc Pro B390; 504.2 GB/s for the RTX 4070.
• Shading units: 1,536 for the Arc Pro B390; 5,888 for the RTX 4070.
• TMUs: 48 for the Arc Pro B390; 184 for the RTX 4070.
• ROPs: 24 for the Arc Pro B390; 64 for the RTX 4070.
• RT cores: 12 for the Arc Pro B390; 46 for the RTX 4070.
• Tensor cores: Not listed for the Arc Pro B390; 184 for the RTX 4070.
• Pixel rate: 60.00 GPixel/s for the Arc Pro B390; 158.4 GPixel/s for the RTX 4070.
• Texture rate: 120.0 GTexel/s for the Arc Pro B390; 455.4 GTexel/s for the RTX 4070.
• FP32: 7.680 TFLOPS for the Arc Pro B390; 29.15 TFLOPS for the RTX 4070.
• FP16: 15.36 TFLOPS (2:1) for the Arc Pro B390; 29.15 TFLOPS (1:1) for the RTX 4070.
• TDP: 80 W for the Arc Pro B390; 200 W for the RTX 4070.
• Slot width: IGP for the Arc Pro B390; dual-slot for the RTX 4070.
• Power connectors: None for the Arc Pro B390; 1x 16-pin for the RTX 4070.
• Suggested PSU: Not listed for the Arc Pro B390; 550 W for the RTX 4070.
• Bus interface: IGP for the Arc Pro B390; PCIe 4.0 x16 for the RTX 4070.
• Display outputs: Portable device dependent for the Arc Pro B390; 1x HDMI 2.1, 3x DisplayPort 1.4a for the RTX 4070.
• Dimensions: Not listed for the Arc Pro B390; 240 mm length, 110 mm height, 40 mm width for the RTX 4070.
• Production status: Active for the Arc Pro B390; End-of-life for the RTX 4070.
• Release date: 2026-01-26 for the Arc Pro B390; 2023-04-11 for the RTX 4070.
• Predecessor: HD Graphics-WM for the Arc Pro B390; GeForce 30 for the RTX 4070.
• Successor: None for the Arc Pro B390; GeForce 50 for the RTX 4070.
• Launch MSRP: Not listed for the Arc Pro B390; 599 USD for the RTX 4070.

The Verdict

The recorded data draws a clear distinction between these two products, primarily based on architecture and measured performance. The NVIDIA GeForce RTX 4070 is a discrete, high-throughput GPU with substantial compute resources and a full set of recorded benchmark results. Its average score of 37,648 and 81st percentile ranking place it well above the database midpoint. The RTX 4070’s raw specifications, including 29.15 TFLOPS FP32, 158.4 GPixel/s pixel rate, and 455.4 GTexel/s texture rate, indicate a part designed for demanding rendering and compute workloads. Its 12 GB of GDDR6X memory on a 192-bit bus delivers 504.2 GB/s of bandwidth, supporting high-resolution textures and large data sets. The presence of 184 tensor cores and 46 RT cores further extends its capability in accelerated workloads.

The Intel Arc Pro B390, by contrast, is an integrated GPU with no dedicated memory, no discrete power connectors, and a 80 W TDP. It has 1,536 shading units, 24 ROPs, and 12 RT cores, with a peak FP32 throughput of 7.680 TFLOPS. Its pixel and texture rates are 60.00 GPixel/s and 120.0 GTexel/s, respectively. The database contains no benchmark results for this part, so its practical performance cannot be quantified. Its 50th percentile rank, absent any test scores, is a placeholder rather than a measured outcome. The part is classified as Active and was released in early 2026, indicating a current-generation integrated solution.

Based on the data, the RTX 4070 is the only product with measurable performance evidence. Its benchmark scores and percentile rank show a GPU that sits near the 81st percentile, with nearest rivals within 1.3% in either direction. The Arc Pro B390 has no such evidence, making any performance claim impossible. For users selecting a GPU for tasks that require high throughput, the RTX 4070 is the supported choice. For users constrained to integrated graphics in a portable device with minimal power draw, the Arc Pro B390 offers a functional but unquantified alternative. The RTX 4070 is End-of-life, while the Arc Pro B390 is Active, which may influence availability, but the performance data favors the NVIDIA part unequivocally.