Intel Arc B390 vs NVIDIA N1X 40SM Comparison

# Intel Arc B390 vs NVIDIA N1X 40SM

The database contains two very different integrated graphics solutions: the Intel Arc B390, built on Intel's 3 nm Panther Lake process with Xe3-LPG architecture, and the NVIDIA N1X 40SM, a Blackwell 2.0 IGP on TSMC's 5 nm node. The Intel part has a recorded 3DMark Steel Nomad DX12 score of 1482, placing it at the 9th percentile of all GPUs, while the NVIDIA part has no benchmark entries and sits at the 50th percentile by default classification. The data reveals a fundamental split: one chip is measured, the other is specified but untested in the database.

Where Each One Wins

The Intel Arc B390 wins in every category where actual measured performance exists. Its sole benchmark result, 3DMark Steel Nomad DX12, yields a score of 1482. That score places it 1.3% ahead of the NVIDIA GeForce GT 520MX (1463), 1.5% ahead of the NVIDIA GeForce 800M (1460), 2.5% ahead of the NVIDIA GeForce GT 625 OEM (1446), and 2.7% ahead of the NVIDIA GeForce GT 710 (1443). These are narrow margins, but they represent real, recorded performance deltas. The NVIDIA N1X 40SM has zero benchmark scores in the database, so it cannot claim a single measured win.

The NVIDIA N1X 40SM wins on raw specifications. It carries 5120 shading units against the Arc B390's 1536, a 3.33x advantage. Its FP32 throughput of 24.02 TFLOPS exceeds the Intel part's 7.680 TFLOPS by a factor of 3.13. The texture rate of 750.7 GTexel/s versus 120.0 GTexel/s is a 6.26x gap. The pixel rate of 93.84 GPixel/s versus 60.00 GPixel/s is a 1.56x lead. The memory subsystem is also categorically different: 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth, versus the Intel part's system-shared memory with system-dependent bandwidth. These specifications suggest the NVIDIA part should dominate in any compute-heavy or memory-bound workload, but the database has no benchmark to confirm that hypothesis.

Architecture Differences

The Intel Arc B390 uses the Xe3-LPG architecture on Intel's 3 nm process, fabricated by Intel. It belongs to the Arc Graphics-M (Panther Lake) generation and supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The chip includes 12 ray tracing cores and no listed tensor cores. Its memory is entirely system-shared, with no dedicated VRAM, no fixed bus width, and bandwidth that varies with the host system. The base clock is 300 MHz with a boost clock of 2500 MHz. The TDP is 80 W, and the slot width is listed as IGP with no power connectors and no suggested PSU. The bus interface is also IGP, meaning it is integrated into the processor package.

The NVIDIA N1X 40SM uses the Blackwell 2.0 architecture on TSMC's 5 nm process, with a die size of 382 mm². It belongs to the Blackwell IGP (N1x) generation. The chip includes 40 ray tracing cores and 160 tensor cores, which the Intel part lacks entirely. The memory is dedicated: 128 GB of LPDDR5X on a 256-bit bus, delivering 273.2 GB/s of bandwidth. The memory clock is listed as 1067 MHz with 8.5 Gbps effective. The base clock is 741 MHz with a boost clock of 2346 MHz. The TDP is listed as unknown. The bus interface is PCIe 5.0 x16, which differs from the Intel part's IGP interface. Display output is limited to 1x HDMI. The NVIDIA part lists DirectX, OpenGL, and Vulkan as N/A, which is unusual and suggests either a proprietary API stack or incomplete data entry.

The process node difference is notable: 3 nm versus 5 nm. The Intel part uses a newer process but a smaller chip, while the NVIDIA part uses an older process with a 382 mm² die. The transistor counts and densities are unknown for both, so no direct comparison is possible from the data.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries for these two parts. The only measured result belongs to the Intel Arc B390: a 3DMark Steel Nomad DX12 score of 1482. That score is the sole data point for either GPU. The NVIDIA N1X 40SM has an average benchmark score of 0, which is consistent with having no recorded tests.

Comparing the Intel part to its nearest rivals provides context. The Arc B390's 1482 score is 1.3% higher than the GT 520MX's 1463, 1.5% higher than the GeForce 800M's 1460, 2.5% higher than the GT 625 OEM's 1446, and 2.7% higher than the GT 710's 1443. These deltas are small, indicating that the Arc B390 sits at the very bottom of the performance spectrum among all recorded GPUs, at the 9th percentile. The NVIDIA N1X 40SM, by contrast, is classified at the 50th percentile, but that classification is not supported by any benchmark score. The 50th percentile figure appears to be a placeholder or derived from its specification tier, not from measured data.

The FP32 comparison is stark: 24.02 TFLOPS versus 7.680 TFLOPS. The NVIDIA part offers 3.13x the raw compute. The texture rate of 750.7 GTexel/s versus 120.0 GTexel/s is a 6.26x advantage. The pixel rate gap is smaller but still significant: 93.84 GPixel/s versus 60.00 GPixel/s, a 1.56x lead. The memory bandwidth of 273.2 GB/s versus system-dependent bandwidth is not directly comparable, but the NVIDIA part's dedicated 128 GB pool is a categorical advantage. These specification differences suggest that in any workload that scales with compute or memory throughput, the NVIDIA part would likely outperform the Intel part by a wide margin, but the database has no measured evidence to confirm or refute that expectation.

FAQ

Q: Which GPU has a higher recorded benchmark score?

A: Only the Intel Arc B390 has a recorded score. It achieved 1482 in 3DMark Steel Nomad DX12. The NVIDIA N1X 40SM has no benchmark entries and an average score of 0.

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

A: The Arc B390's 1482 score is 1.3% higher than the GT 520MX (1463), 1.5% higher than the GeForce 800M (1460), 2.5% higher than the GT 625 OEM (1446), and 2.7% higher than the GT 710 (1443).

Q: What are the memory configurations of the two parts?

A: The Intel Arc B390 uses system-shared memory with system-dependent bandwidth and no dedicated VRAM. The NVIDIA N1X 40SM uses 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth.

Q: Which GPU has more shading units and tensor cores?

A: The NVIDIA N1X 40SM has 5120 shading units and 160 tensor cores. The Intel Arc B390 has 1536 shading units and no listed tensor cores.

Q: What is the FP32 compute difference?

A: The NVIDIA N1X 40SM delivers 24.02 TFLOPS FP32, while the Intel Arc B390 delivers 7.680 TFLOPS, a 3.13x advantage for the NVIDIA part.

Q: Do both GPUs support the same APIs?

A: No. The Intel Arc B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA N1X 40SM lists DirectX, OpenGL, and Vulkan as N/A.

Specification Differences

| Specification | Intel Arc B390 | NVIDIA N1X 40SM |

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

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Process Node | 3 nm (Intel) | 5 nm (TSMC) |

| Die Size | unknown | 382 mm² |

| Base Clock | 300 MHz | 741 MHz |

| Boost Clock | 2500 MHz | 2346 MHz |

| Memory Size | System Shared | 128 GB |

| Memory Type | System Shared | LPDDR5X |

| Bus Width | System Shared | 256 bit |

| Bandwidth | System Dependent | 273.2 GB/s |

| Shading Units | 1536 | 5120 |

| TMUs | 48 | 320 |

| ROPs | 24 | 40 |

| RT Cores | 12 | 40 |

| Tensor Cores | none listed | 160 |

| Pixel Rate | 60.00 GPixel/s | 93.84 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 750.7 GTexel/s |

| FP32 | 7.680 TFLOPS | 24.02 TFLOPS |

| FP16 | 15.36 TFLOPS (2:1) | 24.02 TFLOPS (1:1) |

| TDP | 80 W | unknown |

| Bus Interface | IGP | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 1x HDMI |

| DirectX | 12 Ultimate (12_2) | N/A |

| OpenGL | 4.6 | N/A |

| Vulkan | 1.4 | N/A |

| Release Date | 2026-01-26 | 2026-05-31 |

| Benchmark Score | 1482 | 0 |

| Percentile | 9 | 50 |

The Verdict

The data supports a clear split. The Intel Arc B390 is a measured, low-end integrated GPU. Its 1482 Steel Nomad score places it at the 9th percentile, just barely ahead of several legacy NVIDIA discrete GPUs from the GT 520MX and GT 710 era. It consumes 80 W, uses system-shared memory, and has a modest 1536 shading units. It is a functional DirectX 12 Ultimate part with Vulkan 1.4 support, suitable for basic 3D workloads.

The NVIDIA N1X 40SM is an untested specification. It has no benchmark scores, no nearest rivals, and an average score of 0. Its 50th percentile classification is not backed by any measured data. What the database does show is a massively more capable specification sheet: 5120 shading units, 320 TMUs, 40 RT cores, 160 tensor cores, 128 GB of LPDDR5X, and 24.02 TFLOPS FP32. The die size of 382 mm² on a 5 nm process suggests a substantial chip. The absence of DirectX, OpenGL, and Vulkan API listings is a notable gap, possibly indicating a non-standard software stack.

For buyers or analysts relying solely on the database, the Intel Arc B390 is the only part with proven performance. The NVIDIA N1X 40SM offers theoretical superiority in every compute and memory metric, but without a benchmark result, that superiority remains unverified. The data indicates that anyone requiring immediate, measured GPU performance should look at the Intel part. Anyone evaluating the NVIDIA part must wait for benchmark entries to validate its specification-level advantages.