Intel Arc Pro B390 vs NVIDIA GeForce RTX 4010 Comparison

Head-to-Head Benchmarks

The Intel Arc Pro B390 and NVIDIA GeForce RTX 4010 occupy opposite ends of the performance spectrum, and the recorded data confirms a fundamental gap in compute capabilities. The Intel Arc Pro B390 delivers a FP32 throughput of 7.680 TFLOPS, which is 2.84 times the 2.706 TFLOPS produced by the RTX 4010. This translates into a raw arithmetic advantage that shows up across every shading workload, though the RTX 4010 has a single recorded benchmark result to anchor its real-world position.

In the 3DMark Steel Nomad DX12 test, the RTX 4010 scores 2893 points. Its nearest rivals cluster tightly around this figure: the RTX 4060 Ti 16 GB scores 2907, the RTX PRO 4000 Blackwell SFF scores 2910, the RTX 4060 Ti 8 GB scores 2913, and the Quadro P600 scores 2923. The RTX 4010 trails its closest competitor by only 0.5%, and the entire rival group sits within a 1% band, indicating that the RTX 4010's performance level is consistent with a narrow segment of the database. The Intel Arc Pro B390 has no benchmark entries, so its percentile standing of 50 against all GPUs comes from a different source of measurement than the RTX 4010's 18th percentile, which is derived from an average benchmark score of 2893.

The texture and pixel throughput figures reinforce the Intel part's lead. The Arc Pro B390 reaches 120.0 GTexel/s and 60.00 GPixel/s, while the RTX 4010 manages 42.29 GTexel/s and 28.19 GPixel/s. These numbers point to a 2.84x advantage in texture fill and a 2.13x advantage in pixel fill, meaning the Intel part can sustain higher geometry and fragment throughput under load. The RTX 4010's FP16 output is identical to its FP32 at 2.706 TFLOPS (1:1), whereas the Intel part doubles its FP16 rate to 15.36 TFLOPS (2:1), a 5.68x gap that matters in workloads with mixed-precision math.

Clock speeds tell a different story. The RTX 4010 runs a base clock of 1417 MHz and boosts to 1762 MHz, while the Arc Pro B390 idles at a 300 MHz base and boosts to 2500 MHz. The Intel part's higher boost clock partially explains its throughput advantage despite having only twice the shading units, 1536 versus 768. The RTX 4010's lower clocks are offset by its dedicated GDDR6 memory subsystem, which provides 96.00 GB/s of bandwidth over a 64-bit bus. The Intel part uses system shared memory, making its bandwidth system dependent, so no direct comparison is possible from the recorded data.

Architecture Differences

The two GPUs come from different manufacturing generations and design philosophies. Intel builds the Arc Pro B390 on a 3 nm process at its own foundry, using the Panther Lake chip with the Xe3-LPG architecture. NVIDIA's RTX 4010 uses the GA107 chip on an 8 nm Samsung process with the older Ampere architecture. The process node gap of 5 nm gives Intel a substantial density and efficiency advantage, though the RTX 4010's discrete design compensates with a dedicated memory interface.

Transistor budgets differ sharply. The RTX 4010 packs 8,700 million transistors into a 200 mm² die, yielding a density of 43.5 million transistors per mm². The Intel part's transistor count and die size are recorded as unknown, so no density comparison is possible. What is clear is the architectural feature set: the Arc Pro B390 uses Xe3-LPG, which supports 12 RT cores and a 2:1 FP16 path, while the RTX 4010 uses Ampere with 6 RT cores and 24 tensor cores, plus a 1:1 FP16 rate. Tensor cores are absent from the Intel specification, which means AI acceleration relies on the shading units alone.

Memory architecture is the largest differentiator. The RTX 4010 has 4 GB of GDDR6 on a 64-bit bus with 96.00 GB/s of bandwidth, a fixed and predictable resource for graphics workloads. The Arc Pro B390 shares system memory, with its bus width and bandwidth described as system shared and system dependent. This makes the Intel part's memory performance entirely dependent on the host platform, whereas the RTX 4010 offers deterministic bandwidth.

The feature-level APIs are identical: both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 4010's display outputs are four mini-DisplayPort 1.4a connectors, while the Intel part's outputs are portable device dependent, reflecting its integrated nature. Power delivery also diverges: the RTX 4010 is rated at 50 W TDP with a 250 W suggested PSU and no power connectors, while the Arc Pro B390 draws 80 W and requires none, being an IGP.

The Verdict

The data indicates two separate market positions. The Intel Arc Pro B390 is an integrated processor graphics solution, built for portable devices where its 80 W TDP and system memory sharing are acceptable trade-offs. Its compute throughput, 7.680 TFLOPS FP32 and 120.0 GTexel/s, is high for an IGP and its 50th percentile standing against all GPUs places it in the middle of the database. The RTX 4010, by contrast, is a discrete single-slot card with a 50 W TDP, a 163 mm length, and a fixed 4 GB GDDR6 memory pool. Its 18th percentile and 2893 average benchmark score place it below the median, and its nearest rivals all score within 1% of it, confirming that it sits at a specific performance tier.

For portable or integrated systems, the Arc Pro B390 is the only option of the two, since the RTX 4010 requires a PCIe 4.0 x8 slot and a single-slot bracket. For discrete desktop builds with a 250 W PSU budget, the RTX 4010 offers a dedicated memory subsystem and a known quantity in 3DMark Steel Nomad, scoring 2893. Neither part leads in every category: the Intel part wins on raw compute, fill rates, and process node, while the RTX 4010 wins on memory determinism, tensor core availability, and a recorded benchmark result.

Specification Differences

The following fields differ between the two parts in the database:

• Process node: Intel Arc Pro B390 uses 3 nm, NVIDIA GeForce RTX 4010 uses 8 nm.
• Foundry: Intel for the Arc Pro B390, Samsung for the RTX 4010.
• Transistors: unknown for the Arc Pro B390, 8,700 million for the RTX 4010.
• Die size: unknown for the Arc Pro B390, 200 mm² for the RTX 4010.
• Transistor density: not recorded for the Arc Pro B390, 43.5M / mm² for the RTX 4010.
• Base clock: 300 MHz for the Arc Pro B390, 1417 MHz for the RTX 4010.
• Boost clock: 2500 MHz for the Arc Pro B390, 1762 MHz for the RTX 4010.
• Memory clock: system shared for the Arc Pro B390, 1500 MHz 12 Gbps effective for the RTX 4010.
• Memory size: system shared for the Arc Pro B390, 4 GB for the RTX 4010.
• Memory type: system shared for the Arc Pro B390, GDDR6 for the RTX 4010.
• Bus width: system shared for the Arc Pro B390, 64 bit for the RTX 4010.
• Bandwidth: system dependent for the Arc Pro B390, 96.00 GB/s for the RTX 4010.
• Shading units: 1536 for the Arc Pro B390, 768 for the RTX 4010.
• TMUs: 48 for the Arc Pro B390, 24 for the RTX 4010.
• ROPs: 24 for the Arc Pro B390, 16 for the RTX 4010.
• RT cores: 12 for the Arc Pro B390, 6 for the RTX 4010.
• Tensor cores: not recorded for the Arc Pro B390, 24 for the RTX 4010.
• Pixel rate: 60.00 GPixel/s for the Arc Pro B390, 28.19 GPixel/s for the RTX 4010.
• Texture rate: 120.0 GTexel/s for the Arc Pro B390, 42.29 GTexel/s for the RTX 4010.
• FP32: 7.680 TFLOPS for the Arc Pro B390, 2.706 TFLOPS for the RTX 4010.
• FP16: 15.36 TFLOPS (2:1) for the Arc Pro B390, 2.706 TFLOPS (1:1) for the RTX 4010.
• TDP: 80 W for the Arc Pro B390, 50 W for the RTX 4010.
• Slot width: IGP for the Arc Pro B390, single-slot for the RTX 4010.
• Power connectors: none for both, but the RTX 4010 lists a 250 W suggested PSU.
• Bus interface: IGP for the Arc Pro B390, PCIe 4.0 x8 for the RTX 4010.
• Display outputs: portable device dependent for the Arc Pro B390, 4x mini-DisplayPort 1.4a for the RTX 4010.
• Dimensions: not recorded for the Arc Pro B390, 163 mm length and 69 mm height for the RTX 4010.
• Release date: 2026-01-26 for the Arc Pro B390, 2024-04-15 for the RTX 4010.
• Predecessor: HD Graphics-WM for the Arc Pro B390, GeForce 30 for the RTX 4010.
• Successor: not recorded for the Arc Pro B390, GeForce 50 for the RTX 4010.
• Percentile: 50 for the Arc Pro B390, 18 for the RTX 4010.
• Average benchmark score: 0 for the Arc Pro B390, 2893 for the RTX 4010.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The Intel Arc Pro B390 delivers 7.680 TFLOPS FP32, which is 2.84 times the 2.706 TFLOPS of the NVIDIA GeForce RTX 4010.

Q: How does the RTX 4010 compare to its nearest rivals in the database?

A: The RTX 4010 scores 2893 in 3DMark Steel Nomad DX12, trailing the RTX 4060 Ti 16 GB (2907) by 0.5%, the RTX PRO 4000 Blackwell SFF (2910) by 0.6%, the RTX 4060 Ti 8 GB (2913) by 0.7%, and the Quadro P600 (2923) by 1%.

Q: What memory configuration does each GPU use?

A: The RTX 4010 uses 4 GB of GDDR6 on a 64-bit bus with 96.00 GB/s bandwidth. The Arc Pro B390 uses system shared memory, with its bandwidth described as system dependent.

Q: Do the two GPUs support the same graphics APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the power draw for each part?

A: The Intel Arc Pro B390 has an 80 W TDP, and the NVIDIA GeForce RTX 4010 has a 50 W TDP with a 250 W suggested PSU.

Q: Which GPU has a recorded benchmark score?

A: Only the NVIDIA GeForce RTX 4010 has a benchmark entry, scoring 2893 in 3DMark Steel Nomad DX12 and holding an 18th percentile ranking. The Intel Arc Pro B390 has no benchmark entries but holds a 50th percentile ranking.