Intel Arc Pro B390 vs NVIDIA N1 16SM Comparison

Head-to-Head Benchmarks

The recorded database contains no direct head-to-head benchmark results for the Intel Arc Pro B390 against the NVIDIA N1 16SM. Both entries show zero benchmark scores, zero wins for either side, and no nearest rival data. This absence of measured performance data means the comparison must rely entirely on architectural specifications and theoretical throughput figures.

The compute throughput comparison shows a clear numerical advantage for the NVIDIA part. The N1 16SM delivers 9.609 TFLOPS of FP32 performance, while the Intel Arc Pro B390 provides 7.680 TFLOPS. This translates to the NVIDIA GPU holding approximately a 25% lead in raw single-precision floating-point throughput. The delta is substantial enough to suggest meaningfully faster general compute workloads, though the lack of real-world benchmarks prevents confirmation.

In FP16 throughput, the situation reverses in terms of ratio. The Intel part achieves 15.36 TFLOPS with a 2:1 rate, doubling its FP32 figure. The NVIDIA part maintains a 1:1 ratio, holding at 9.609 TFLOPS for FP16 as well. This gives Intel a 60% advantage in half-precision compute when the hardware acceleration path is used. Applications that leverage FP16 tensor operations or shader math could see significant benefits on the Intel architecture.

Texture processing heavily favors NVIDIA. The N1 16SM outputs 300.3 GTexel/s compared to the Intel part's 120.0 GTexel/s, a 2.5x difference. This stems from the NVIDIA GPU having 128 texture mapping units versus 48 on the Intel chip. Pixel throughput is nearly identical: Intel posts 60.00 GPixel/s against NVIDIA's 56.30 GPixel/s, a marginal 6.6% advantage for Intel that comes from its slightly higher boost clock.

Clock behavior differs meaningfully. The Intel Arc Pro B390 has a base clock of 300 MHz and a boost clock of 2500 MHz, indicating a wide dynamic range. The NVIDIA N1 16SM runs at 741 MHz base and 2346 MHz boost, with a narrower operating window. The Intel part's boost clock is 6.6% higher, which explains its small pixel rate lead despite having the same 24 ROPs.

Memory architecture presents a stark contrast. The NVIDIA N1 16SM integrates 128 GB of LPDDR5X on a 256-bit bus, delivering 273.2 GB/s of bandwidth. The Intel Arc Pro B390 uses system shared memory with bandwidth described as system dependent, providing no fixed figure. The NVIDIA part's dedicated memory allocation and fixed bandwidth offer predictable performance, while the Intel solution's performance varies with the host system's memory configuration.

Architecture Differences

The two processors come from fundamentally different design lineages. Intel's Arc Pro B390 uses the Xe3-LPG architecture built on Panther Lake silicon, fabricated on Intel's 3 nm process. NVIDIA's N1 16SM uses the Blackwell 2.0 architecture on GB20B silicon, manufactured by TSMC on a 5 nm process. The die size for the NVIDIA chip is 382 mm², while Intel's die size is listed as unknown in the database.

Shader resources differ substantially. The NVIDIA part carries 2048 shading units, 128 TMUs, 24 ROPs, 16 RT cores, and 64 tensor cores. The Intel part has 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores, with tensor cores listed as null in the database. NVIDIA holds a 33% advantage in shading units, a 2.7x advantage in TMUs, and a 33% advantage in RT cores. The 64 tensor cores on NVIDIA's part versus no listed tensor cores on Intel's side indicates different AI acceleration approaches.

API support creates a major functional divide. The Intel Arc Pro B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA N1 16SM lists all three APIs as N/A, meaning it does not support standard graphics APIs in the conventional sense. This suggests the N1 16SM is not designed for general-purpose Windows gaming or traditional graphics workloads, despite its substantial compute resources.

Memory architecture diverges completely. NVIDIA integrates 128 GB of LPDDR5X with a 256-bit bus and fixed 273.2 GB/s bandwidth. Intel uses system shared memory with no dedicated allocation, type listed as system shared, and bandwidth labeled system dependent. The NVIDIA approach provides a large, fast, dedicated memory pool; the Intel approach relies on whatever system memory is available.

Power and connectivity profiles differ. Intel lists a TDP of 80 W with no power connectors and an IGP bus interface. NVIDIA's TDP is listed as unknown, also with no power connectors and an IGP slot width, but it uses a PCIe 5.0 x16 bus interface. Display outputs also differ: Intel is listed as portable device dependent, while NVIDIA provides 1x HDMI.

Release timing shows a four-month gap. Intel's Arc Pro B390 entered the database with a release date of January 26, 2026. NVIDIA's N1 16SM followed on May 31, 2026. Both are listed as active production parts.

FAQ

Q: Which GPU has higher FP32 compute throughput?

A: The NVIDIA N1 16SM delivers 9.609 TFLOPS versus 7.680 TFLOPS for the Intel Arc Pro B390, giving NVIDIA a 25% lead in single-precision performance.

Q: Does the Intel part support modern graphics APIs?

A: Yes, the Intel Arc Pro B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA N1 16SM lists all three APIs as N/A, indicating no standard graphics API support.

Q: How much memory does each GPU have?

A: The NVIDIA N1 16SM has 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth. The Intel Arc Pro B390 uses system shared memory with bandwidth that is system dependent and no fixed memory size.

Q: What are the boost clock speeds?

A: The Intel Arc Pro B390 boosts to 2500 MHz, while the NVIDIA N1 16SM boosts to 2346 MHz. Intel's boost clock is 6.6% higher.

Q: Which GPU has more texture mapping units?

A: The NVIDIA N1 16SM has 128 TMUs, compared to 48 on the Intel Arc Pro B390, a 2.7x advantage that drives its 300.3 GTexel/s texture rate versus 120.0 GTexel/s.

Q: What process nodes are used?

A: Intel uses its 3 nm process for the Arc Pro B390, while NVIDIA uses TSMC's 5 nm process for the N1 16SM, which has a die size of 382 mm².

The Verdict

The data indicates two products aimed at entirely different roles. The NVIDIA N1 16SM is a compute-oriented part with 2048 shading units, 64 tensor cores, 128 GB of dedicated memory, and no graphics API support. The Intel Arc Pro B390 is a graphics-capable IGP with DirectX 12 Ultimate, Vulkan 1.4, and OpenGL 4.6 support, using system shared memory and drawing only 80 W.

For traditional graphics workloads, the Intel Arc Pro B390 is the only viable choice of the two, since NVIDIA lists all graphics APIs as N/A. The Intel part also wins on FP16 throughput with 15.36 TFLOPS versus 9.609 TFLOPS, and has a slight edge in pixel rate at 60.00 GPixel/s versus 56.30 GPixel/s.

For raw compute throughput, FP32, texture work, and memory bandwidth, the NVIDIA N1 16SM dominates. Its 9.609 TFLOPS FP32, 300.3 GTexel/s texture rate, and 273.2 GB/s bandwidth are all substantially higher than the Intel part's figures. The 64 tensor cores also give NVIDIA a dedicated AI acceleration path that Intel does not list.

The release dates place the Intel part in January 2026 and the NVIDIA part in May 2026. Both are active production parts with no launch MSRP recorded in the database. The choice between them depends entirely on whether the workload requires graphics API compatibility or maximum compute throughput.

Specification Differences

| Specification | Intel Arc Pro B390 | NVIDIA N1 16SM |

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

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | 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 |

| Memory Bus Width | System Shared | 256 bit |

| Memory Bandwidth | System Dependent | 273.2 GB/s |

| Shading Units | 1536 | 2048 |

| TMUs | 48 | 128 |

| ROPs | 24 | 24 |

| RT Cores | 12 | 16 |

| Tensor Cores | null | 64 |

| Pixel Rate | 60.00 GPixel/s | 56.30 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 300.3 GTexel/s |

| FP32 | 7.680 TFLOPS | 9.609 TFLOPS |

| FP16 | 15.36 TFLOPS (2:1) | 9.609 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 |

Where Each One Wins

The Intel Arc Pro B390 wins in graphics API compatibility. It supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, making it suitable for conventional rendering pipelines. Its FP16 throughput of 15.36 TFLOPS doubles its FP32 figure, which is advantageous for workloads that can use half-precision math. The pixel rate of 60.00 GPixel/s edges out NVIDIA's 56.30 GPixel/s, and the 80 W TDP with no power connectors makes it a low-power integrated solution.

The NVIDIA N1 16SM wins in compute density. Its 9.609 TFLOPS FP32 is 25% higher than Intel's. The texture rate of 300.3 GTexel/s is 2.5x higher, driven by 128 TMUs. The 273.2 GB/s fixed memory bandwidth with 128 GB capacity dwarfs Intel's system-dependent shared memory. The 64 tensor cores provide a hardware path for AI workloads that the Intel part lacks entirely. The PCIe 5.0 x16 interface also enables host connectivity beyond Intel's IGP bus.

The use-case split is clean: Intel for graphics-capable systems with standard API demands, NVIDIA for compute-heavy applications that do not require DirectX or Vulkan graphics support.