Intel Arc Pro B390 vs NVIDIA GeForce RTX 5080 SUPER Comparison

FAQ

Q: What is the Intel Arc Pro B390?

A: The Intel Arc Pro B390 is an integrated graphics processor (IGP) based on the Panther Lake chip, using the Xe3-LPG architecture on a 3 nm process node. It is part of the Arc Graphics-WM (Panther Lake) generation and was released on January 26, 2026.

Q: What is the NVIDIA GeForce RTX 5080 SUPER?

A: The NVIDIA GeForce RTX 5080 SUPER is a discrete dual-slot graphics card from the GeForce 50-series, built on the GB203 chip with Blackwell 2.0 architecture. It uses a 5 nm process from TSMC and was released on December 31, 2025.

Q: How do the two compare in terms of memory?

A: The Intel Arc Pro B390 uses system shared memory with system-dependent bandwidth, while the NVIDIA GeForce RTX 5080 SUPER has 24 GB of GDDR7 memory on a 256-bit bus with 1.02 TB/s of bandwidth.

Q: What are the power requirements for each?

A: The Intel Arc Pro B390 has a TDP of 80 W and requires no power connectors, while the NVIDIA GeForce RTX 5080 SUPER has a TDP of 415 W and requires a single 16-pin power connector.

Q: Which card supports ray tracing and what are the specifications?

A: Both support DirectX 12 Ultimate (12_2). The Intel Arc Pro B390 has 12 ray tracing cores, while the NVIDIA GeForce RTX 5080 SUPER has 84 ray tracing cores and 336 tensor cores.

Q: What is the launch MSRP of the NVIDIA GeForce RTX 5080 SUPER?

A: The launch MSRP of the NVIDIA GeForce RTX 5080 SUPER is 999 USD. The Intel Arc Pro B390 has no recorded launch MSRP.

Architecture Differences

The two products occupy opposite ends of the GPU spectrum in both physical implementation and architectural design. The Intel Arc Pro B390 is an integrated graphics processor built for the Panther Lake chip, using the Xe3-LPG architecture on a 3 nm process node fabricated by Intel. In contrast, the NVIDIA GeForce RTX 5080 SUPER is a discrete add-in board based on the GB203 chip with Blackwell 2.0 architecture, manufactured by TSMC on a 5 nm process.

The Intel part ships with 1,536 shading units, 48 texture mapping units, and 24 raster operation pipelines. Its 12 ray tracing cores provide hardware-accelerated ray tracing within a compact IGP package. The NVIDIA part is dramatically larger in compute scale: 10,752 shading units, 336 TMUs, 112 ROPs, 84 ray tracing cores, and 336 tensor cores. The tensor core count indicates a substantially wider matrix math capability for the NVIDIA product, useful for AI-accelerated workloads.

The process node difference is notable: Intel uses a 3 nm node while NVIDIA uses a 5 nm node. The NVIDIA chip contains 45,600 million transistors on a 378 mm² die, resulting in a transistor density of 120.6M per mm². The Intel chip's transistor count and die size are listed as unknown in the database.

Clock behavior also differs. The Intel Arc Pro B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The NVIDIA GeForce RTX 5080 SUPER runs a base clock of 2295 MHz and a boost clock of 2617 MHz. The NVIDIA card's memory clock is 2000 MHz with 32 Gbps effective data rate, whereas the Intel part relies on system shared memory with no dedicated memory clock.

Feature support from an API standpoint is identical: both list DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. However, the physical form factors could not be more different. The Intel part is an IGP with no slot width, no power connectors, and a portable device dependent display output. The NVIDIA card is a dual-slot board measuring 304 mm in length, 137 mm in height, and 40 mm in width, with one HDMI 2.1b output and three DisplayPort 2.1b outputs, connected via PCIe 5.0 x16.

The Verdict

The recorded data separates these two products into entirely different deployment categories. The Intel Arc Pro B390, with its 80 W TDP and integrated form factor, is designed for systems where space and power are constrained. Its 50th percentile ranking against all GPUs, combined with zero recorded benchmark scores, indicates it is positioned as a general-purpose integrated solution rather than a performance leader.

The NVIDIA GeForce RTX 5080 SUPER, meanwhile, is a high-power discrete card with a 415 W TDP, a dual-slot cooler, and a dedicated 16-pin power connector. Its single recorded benchmark score in 3DMark Steel Nomad DX12 is 3075, placing it in the 19th percentile against all GPUs. The nearest rival data shows it is 2.8% behind the NVIDIA Quadro P1000, 3.4% behind the Intel Arc Pro B60, 3.1% ahead of the NVIDIA GeForce 820A, and 3.6% ahead of the NVIDIA GeForce GTX 860M.

From the data, the Intel Arc Pro B390 suits integrated, low-power systems where the CPU package must handle graphics duties. The NVIDIA GeForce RTX 5080 SUPER suits discrete desktops requiring a dedicated board with substantial memory bandwidth and compute throughput. Users needing 24 GB of dedicated GDDR7 memory, 1.02 TB/s bandwidth, and a large shading unit count should select the NVIDIA product. Users needing a 3 nm integrated solution with no external power requirement should select the Intel product.

Specification Differences

| Specification | Intel Arc Pro B390 | NVIDIA GeForce RTX 5080 SUPER |

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

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Unknown | 45,600 million |

| Die Size | Unknown | 378 mm² |

| Transistor Density | Not recorded | 120.6M / mm² |

| Base Clock | 300 MHz | 2295 MHz |

| Boost Clock | 2500 MHz | 2617 MHz |

| Memory Size | System Shared | 24 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 256 bit |

| Memory Bandwidth | System Dependent | 1.02 TB/s |

| Shading Units | 1536 | 10752 |

| TMUs | 48 | 336 |

| ROPs | 24 | 112 |

| Ray Tracing Cores | 12 | 84 |

| Tensor Cores | Not recorded | 336 |

| Pixel Rate | 60.00 GPixel/s | 293.1 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 879.3 GTexel/s |

| FP32 Performance | 7.680 TFLOPS | 56.28 TFLOPS |

| FP16 Performance | 15.36 TFLOPS (2:1) | 56.28 TFLOPS (1:1) |

| TDP | 80 W | 415 W |

| Slot Width | IGP | Dual-slot |

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

| Bus Interface | IGP | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 1x HDMI 2.1b, 3x DisplayPort 2.1b |

| Dimensions | Not recorded | 304 mm x 137 mm x 40 mm |

| Release Date | 2026-01-26 | 2025-12-31 |

| Launch MSRP | Not recorded | 999 USD |

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries between the Intel Arc Pro B390 and the NVIDIA GeForce RTX 5080 SUPER. However, the recorded measurements for the NVIDIA part allow for meaningful analysis of its standing relative to other GPUs.

The NVIDIA GeForce RTX 5080 SUPER achieves a score of 3075 in the 3DMark Steel Nomad DX12 test. This places it in the 19th percentile against all GPUs, meaning the majority of recorded GPUs score higher. The nearest rival data confirms this middling position. The NVIDIA Quadro P1000 scores 3163, which is 2.8% higher than the RTX 5080 SUPER. The Intel Arc Pro B60 scores 3182, which is 3.4% higher. Both of these rivals outperform the RTX 5080 SUPER in the recorded benchmark.

On the other side, the NVIDIA GeForce 820A scores 2983, which is 3.1% lower than the RTX 5080 SUPER. The NVIDIA GeForce GTX 860M scores 2967, which is 3.6% lower. The RTX 5080 SUPER therefore sits in a narrow band, within roughly 3.6% of all four nearest rivals, with a slight advantage over the two lower-scoring parts and a slight disadvantage against the two higher-scoring parts.

The Intel Arc Pro B390 has no recorded benchmark scores and no nearest rivals in the database. Its percentile rank of 50 against all GPUs places it at the median of the recorded GPU population, but without actual benchmark data, no direct performance comparison can be made against the NVIDIA part.

For the NVIDIA card, the FP32 throughput of 56.28 TFLOPS and texture rate of 879.3 GTexel/s indicate high raw compute capacity, while the Intel part's FP32 of 7.680 TFLOPS and texture rate of 120.0 GTexel/s represent a much smaller compute envelope. The pixel rate difference is similarly large: 293.1 GPixel/s for NVIDIA versus 60.00 GPixel/s for Intel. These figures, combined with the 24 GB versus system shared memory configuration, show that the NVIDIA card is designed for substantially heavier rendering workloads despite its unremarkable benchmark percentile standing.