Intel Arc Pro B390 vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded data shows no direct benchmark comparisons between the Intel Arc Pro B390 and the NVIDIA GeForce RTX 5090 D. The Intel part carries an average benchmark score of zero and holds a 50th percentile position among all GPUs in the database, while the NVIDIA part has a substantial average benchmark score of 77,712 and sits at the 92nd percentile. The NVIDIA GeForce RTX 5090 D therefore occupies a much higher performance tier, but the absence of head-to-head test results means the gap cannot be quantified in specific benchmark deltas.

The NVIDIA GeForce RTX 5090 D does have recorded scores across several tests. In 3DMark Steel Nomad DX12 it scores 14,326. Geekbench OpenCL returns 310,674, and Geekbench Vulkan returns 376,915. PassMark results include 231 for DirectX 10, 371 for DirectX 11, 219 for DirectX 12, 434 for DirectX 9, 1,487 for G2D, 44,065 for G3D, and 28,396 for GPU compute. These numbers reflect a GPU with strong compute and graphics output, but they cannot be directly compared to the Intel Arc Pro B390 because that part has no recorded benchmark entries.

The nearest rivals for the NVIDIA GeForce RTX 5090 D provide useful context. The AMD Radeon RX 6650M XT scores 76,904, which is 1.1% lower than the NVIDIA part. The AMD Radeon RX 6850M XT scores 78,940, which is 1.6% higher. The NVIDIA Tesla P100 PCIe 12 GB scores 79,396, which is 2.1% higher, and the NVIDIA Tesla P100 PCIe 16 GB scores 79,605, which is 2.4% higher. This places the RTX 5090 D in a narrow band around these mobile and data center parts, despite its desktop positioning. The Intel Arc Pro B390 has no nearest rivals listed, so no comparative performance statements can be made for it.

Benchmark results indicate that the RTX 5090 D is competitive with, but not dominant over, its closest scored rivals. The delta percentages are small, ranging from 1.1% below to 2.4% above the NVIDIA part. This suggests that in the database's aggregate scoring, the RTX 5090 D performs near the level of those specific AMD and Tesla products. The Intel part, with no scores, cannot be placed in any such comparison.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. The Intel Arc Pro B390 uses the Panther Lake chip with the Xe3-LPG architecture, built on a 3 nm process at Intel's own foundry. The NVIDIA GeForce RTX 5090 D uses the GB202 chip with the Blackwell 2.0 architecture, built on a 5 nm process at TSMC. The Intel part belongs to the Arc Graphics-WM (Panther Lake) generation, while the NVIDIA part belongs to the GeForce 50 series.

The Intel chip is an integrated graphics processor, indicated by the IGP slot width, bus interface, and power connector fields. It has no dedicated power connectors and consumes 80 W. The NVIDIA part is a dual-slot discrete card with a 16-pin power connector, a 575 W TDP, and a suggested 950 W power supply. The physical format difference is stark: one is an IGP embedded in a system, the other is a large add-in board measuring 304 mm in length, 137 mm in height, and 48 mm in width.

Transistor counts differ wildly. The NVIDIA chip packs 92,200 million transistors on a 750 mm² die, with a transistor density of 122.9M per mm². The Intel chip's transistor count and die size are listed as unknown. The Intel process node is smaller at 3 nm versus 5 nm, but the NVIDIA chip uses a much larger die to house far more compute resources.

Compute resources show a massive disparity. The Intel Arc Pro B390 has 1,536 shading units, 48 texture mapping units, 24 ROPs, and 12 ray tracing cores. The NVIDIA GeForce RTX 5090 D has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The Intel part has no tensor core field, while the NVIDIA part includes a full tensor core array. Pixel rate for Intel is 60.00 GPixel/s, while NVIDIA reaches 423.6 GPixel/s. Texture rate for Intel is 120.0 GTexel/s, while NVIDIA reaches 1,636.8 GTexel/s.

Memory architecture is another major split. The Intel part uses system shared memory, with system dependent bandwidth and no dedicated VRAM. The NVIDIA part has 32 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s of bandwidth. Clock speeds also differ: Intel runs at a 300 MHz base and 2500 MHz boost, while NVIDIA runs at 2017 MHz base and 2407 MHz boost. The NVIDIA boost clock is slightly lower than Intel's, but the NVIDIA part's enormous shading unit count and memory bandwidth make the raw clock comparison secondary.

Floating point throughput shows the scale of difference. The Intel part delivers 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 at a 2:1 ratio. The NVIDIA part delivers 104.8 TFLOPS FP32 and 104.8 TFLOPS FP16 at a 1:1 ratio. That is a roughly 13.6x difference in FP32 throughput, though the lack of direct benchmark data means this is a specification-level observation, not a measured performance result.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs differ: the Intel part is portable device dependent, while the NVIDIA part has 1x HDMI 2.1b and 3x DisplayPort 2.1b. The NVIDIA part also has a release date of January 2025, while the Intel part is dated January 2026. The Intel part's predecessor is HD Graphics-WM, and the NVIDIA part's predecessor is GeForce 40, with a successor listed as GeForce 60.

FAQ

Q: What is the biggest architectural difference between the two GPUs?

A: The Intel Arc Pro B390 is an integrated graphics processor using the Xe3-LPG architecture on a 3 nm process, while the NVIDIA GeForce RTX 5090 D is a discrete dual-slot card using Blackwell 2.0 on a 5 nm process. The Intel part has no dedicated memory and uses system shared memory, while the NVIDIA part has 32 GB of GDDR7.

Q: How do the compute resources compare?

A: The NVIDIA part has 21,760 shading units, 680 TMUs, 176 ROPs, and 170 RT cores. The Intel part has 1,536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores. The NVIDIA part also includes 680 tensor cores, while the Intel part has no tensor core field.

Q: Which GPU has a higher boost clock?

A: The Intel Arc Pro B390 has a boost clock of 2500 MHz, which is higher than the NVIDIA GeForce RTX 5090 D's boost clock of 2407 MHz. However, the NVIDIA part's base clock is 2017 MHz versus Intel's 300 MHz base clock.

Q: What is the power consumption difference?

A: The Intel part has a TDP of 80 W and no power connectors. The NVIDIA part has a TDP of 575 W, requires a 16-pin power connector, and has a suggested power supply of 950 W.

Q: Are there any benchmark scores for the Intel Arc Pro B390?

A: No. The Intel part has an empty benchmark list and an average benchmark score of zero. The NVIDIA part has recorded scores across 3DMark, Geekbench, and PassMark tests, with an average benchmark score of 77,712.

Q: How does the NVIDIA part compare to its nearest rivals?

A: The AMD Radeon RX 6650M XT scores 1.1% lower, the AMD Radeon RX 6850M XT scores 1.6% higher, the NVIDIA Tesla P100 PCIe 12 GB scores 2.1% higher, and the NVIDIA Tesla P100 PCIe 16 GB scores 2.4% higher.

Specification Differences

| Field | Intel Arc Pro B390 | NVIDIA GeForce RTX 5090 D |

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

| Chip | Panther Lake | GB202 |

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Generation | Arc Graphics-WM (Panther Lake) | GeForce 50 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | unknown | 92,200 million |

| Die Size | unknown | 750 mm² |

| Transistor Density | null | 122.9M / mm² |

| Base Clock | 300 MHz | 2017 MHz |

| Boost Clock | 2500 MHz | 2407 MHz |

| Memory Size | System Shared | 32 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 512 bit |

| Memory Bandwidth | System Dependent | 1.79 TB/s |

| Shading Units | 1536 | 21760 |

| TMUs | 48 | 680 |

| ROPs | 24 | 176 |

| RT Cores | 12 | 170 |

| Tensor Cores | null | 680 |

| Pixel Rate | 60.00 GPixel/s | 423.6 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 1,636.8 GTexel/s |

| FP32 | 7.680 TFLOPS | 104.8 TFLOPS |

| FP16 | 15.36 TFLOPS (2:1) | 104.8 TFLOPS (1:1) |

| TDP | 80 W | 575 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 16-pin |

| Suggested PSU | null | 950 W |

| Bus Interface | IGP | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 1x HDMI 2.1b3x DisplayPort 2.1b |

| Dimensions | null | 304 mm x 137 mm x 48 mm |

| Release Date | 2026-01-26 | 2025-01-29 |

| Predecessor | HD Graphics-WM | GeForce 40 |

| Successor | null | GeForce 60 |

| Launch MSRP | null | 2,299 USD |

Where Each One Wins

The NVIDIA GeForce RTX 5090 D wins on nearly every measurable specification. It has far more shading units, TMUs, ROPs, and RT cores. Its memory subsystem is dedicated GDDR7 with 1.79 TB/s bandwidth, whereas the Intel part relies on system shared memory with system dependent bandwidth. The NVIDIA part delivers 104.8 TFLOPS FP32 versus 7.680 TFLOPS for Intel, a 13.6x advantage. Pixel rate and texture rate are also far higher on the NVIDIA part.

The Intel Arc Pro B390 wins on power efficiency. Its 80 W TDP is dramatically lower than the NVIDIA part's 575 W. It requires no power connectors and no suggested PSU rating. The Intel part also uses a smaller 3 nm process node, which typically indicates better transistor efficiency. Its boost clock of 2500 MHz is higher than the NVIDIA part's 2407 MHz, though the NVIDIA part compensates with a much higher base clock and far more compute units.

The Intel part is an integrated graphics processor, which means it occupies no expansion slot and has no physical dimensions listed. It is suitable for systems where a discrete card is not an option. The NVIDIA part is a large dual-slot card that requires a PCIe 5.0 x16 slot and a 950 W power supply. Its physical footprint is substantial.

The Intel part has no benchmark scores, so it cannot claim any measured performance wins. The NVIDIA part has recorded scores in multiple tests and sits at the 92nd percentile, while the Intel part sits at the 50th percentile with an average score of zero. The NVIDIA part also has a launch MSRP of 2,299 USD, while the Intel part has no launch MSRP listed.

The Verdict

The data indicates a clear separation in purpose and capability. The NVIDIA GeForce RTX 5090 D is a high-end discrete GPU with massive compute resources, dedicated 32 GB GDDR7 memory, and an average benchmark score of 77,712. It holds the 92nd percentile position. Its nearest rivals are within a narrow 2.4% range, showing it competes closely with those specific parts. Any system using this GPU must accommodate a 575 W TDP, a 16-pin power connector, and a 950 W suggested PSU.

The Intel Arc Pro B390 is an integrated graphics processor with a 50th percentile position and no recorded benchmarks. Its 80 W TDP and system shared memory make it a low-power option for portable or compact systems. Its 3 nm process node and Xe3-LPG architecture represent a modern design, but the compute resource counts are far lower.

The choice between the two depends entirely on the use case. For workloads requiring maximum FP32 throughput, high memory bandwidth, and dedicated VRAM, the NVIDIA part is the only option with measured data supporting its performance. For systems where power consumption, slot space, and integration matter more than raw compute, the Intel part fits a completely different profile. The absence of benchmark data for the Intel part means no performance claims can be made for it, while the NVIDIA part has a full set of test scores. The recorded data does not support any direct comparison, so any decision must rest on the specification differences and the NVIDIA part's benchmark presence.