Intel Arc B390 vs NVIDIA GeForce RTX 5080 SUPER Comparison

Head-to-Head Benchmarks

The only recorded benchmark comparison between the Intel Arc B390 and the NVIDIA GeForce RTX 5080 SUPER is the 3DMark Steel Nomad DX12 test. The results show a decisive victory for the NVIDIA card, scoring 3075 points against 1482 points for the Intel part. The recorded delta is -51.8%, meaning the Intel Arc B390 delivers slightly more than half the performance of the RTX 5080 SUPER in this workload.

Placing both scores within the broader database context clarifies the gap. The Intel Arc B390 sits at the 9th percentile of all GPUs in the database, with an average benchmark score of 1482. Its nearest rivals are older low-end NVIDIA parts: the GeForce GT 520MX at 1463 points (1.3% behind), the GeForce 800M at 1460 points (1.5% behind), the GeForce GT 625 OEM at 1446 points (2.5% behind), and the GeForce GT 710 at 1443 points (2.7% behind). The Arc B390 is therefore clustered tightly with a group of legacy entry-level GPUs, separated by margins under 3%.

The RTX 5080 SUPER, by contrast, reaches the 19th percentile of all GPUs, a modest placement that reflects the unusual composition of the database sample. Its average score of 3075 puts it near the NVIDIA Quadro P1000 at 3163 points (2.8% ahead of the RTX card), the Intel Arc Pro B60 at 3182 points (3.4% ahead), the GeForce 820A at 2983 points (3.1% behind), and the GeForce GTX 860M at 2967 points (3.6% behind). The RTX 5080 SUPER is essentially bracketed by a workstation GPU, a professional Arc part, and two older mobile-origin NVIDIA parts, all within a 3.6% band. Despite this odd company, the RTX 5080 SUPER more than doubles the Arc B390's raw score in the Steel Nomad test.

The win distribution is one-sided: the RTX 5080 SUPER claims the single head-to-head win, while the Arc B390 has none. There is no recorded test where the Intel part leads, either in absolute score or in any submetric. The data shows a straightforward performance hierarchy: the NVIDIA card is faster by a wide margin, and the Intel part operates in a completely different performance tier, one populated by GPUs from roughly a decade earlier.

Architecture Differences

The two parts share several API-level capabilities but diverge sharply in nearly every physical and architectural attribute. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so feature parity at the API level is not the differentiator. The differences lie underneath.

The Intel Arc B390 uses the Xe3-LPG architecture built on Panther Lake, fabricated by Intel on a 3 nm process. The RTX 5080 SUPER uses the Blackwell 2.0 architecture with the GB203 chip, fabricated by TSMC on a 5 nm process. The transistor counts tell a stark story: the NVIDIA chip packs 45,600 million transistors on a 378 mm² die, yielding a transistor density of 120.6 million per mm². The Intel part's transistor count and die size are not recorded in the database.

Memory configuration separates the two fundamentally. The Arc B390 uses system shared memory, with shared type, shared bus width, and bandwidth listed as system dependent. The RTX 5080 SUPER has 24 GB of GDDR7 on a 256-bit bus, delivering 1.02 TB/s of bandwidth. This is a dedicated memory subsystem versus an integrated, shared-memory design. The clock behavior reflects the same divide: the Intel part runs at a 300 MHz base and 2500 MHz boost, while the NVIDIA card runs at 2295 MHz base and 2617 MHz boost, with memory clocked at 2000 MHz (32 Gbps effective).

Compute resources are radically different in scale. The Arc B390 has 1536 shading units, 48 texture mapping units, and 24 raster output units. The RTX 5080 SUPER has 10752 shading units, 336 TMUs, and 112 ROPs. The NVIDIA card carries 84 ray tracing cores and 336 tensor cores; the Intel part has 12 ray tracing cores and no recorded tensor core count. The pixel rate for the RTX 5080 SUPER is 293.1 GPixel/s versus 60.00 GPixel/s for the Arc B390. Texture rate is 879.3 GTexel/s versus 120.0 GTexel/s. FP32 throughput is 56.28 TFLOPS versus 7.680 TFLOPS. The FP16 figures reveal different ratio policies: the Arc B390 delivers 15.36 TFLOPS at a 2:1 ratio, while the RTX 5080 SUPER delivers 56.28 TFLOPS at a 1:1 ratio, meaning the NVIDIA part does not halve its throughput for FP16.

Power and physical design reinforce the gap. The Arc B390 is an integrated graphics processor with an 80 W TDP, no power connectors, an IGP bus interface, and display outputs dependent on the portable device. The RTX 5080 SUPER is a dual-slot discrete card with a 415 W TDP, a single 16-pin power connector, a PCIe 5.0 x16 interface, and fixed display outputs consisting of one HDMI 2.1b port and three DisplayPort 2.1b ports. Its physical dimensions are recorded at 304 mm in length, 137 mm in height, and 40 mm in width. The Intel part has no recorded dimensions.

The process node difference is notable in the opposite direction from the performance gap: Intel uses 3 nm while TSMC uses 5 nm for the NVIDIA chip. The architectural generation also differs, with the Arc B390 belonging to the Arc Graphics-M (Panther Lake) generation and the RTX 5080 SUPER belonging to the GeForce 50 series. Both are listed as active production parts.

FAQ

Q: Which GPU has the higher 3DMark Steel Nomad DX12 score?

A: The NVIDIA GeForce RTX 5080 SUPER scores 3075 points, while the Intel Arc B390 scores 1482 points. The delta is -51.8% for the Intel part.

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

A: The Arc B390 sits at the 9th percentile of all GPUs. Its nearest rivals are the NVIDIA GeForce GT 520MX at 1463 points (1.3% behind), the GeForce 800M at 1460 points (1.5% behind), the GeForce GT 625 OEM at 1446 points (2.5% behind), and the GeForce GT 710 at 1443 points (2.7% behind).

Q: How does the RTX 5080 SUPER compare to its nearest rivals?

A: The RTX 5080 SUPER sits at the 19th percentile of all GPUs. The Intel Arc Pro B60 leads it by 3.4% with 3182 points, the Quadro P1000 leads by 2.8% with 3163 points, the GeForce 820A trails by 3.1% with 2983 points, and the GeForce GTX 860M trails by 3.6% with 2967 points.

Q: What memory configurations do the two GPUs use?

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

Q: What are the TDP and power connector requirements?

A: The Arc B390 has an 80 W TDP and no power connectors, as it is an integrated GPU. The RTX 5080 SUPER has a 415 W TDP and uses a single 16-pin power connector.

Q: Do both GPUs support the same graphics APIs?

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

The Verdict

The recorded data supports a clear separation of roles. The RTX 5080 SUPER is the faster part by a wide margin, delivering more than double the Steel Nomad score of the Arc B390 while also offering far larger compute resources, dedicated GDDR7 memory, and ray tracing and tensor core counts that dwarf the Intel part. Its 10752 shading units, 84 RT cores, 336 tensor cores, and 56.28 TFLOPS of FP32 throughput place it in a performance class the Intel integrated GPU cannot approach. The RTX 5080 SUPER is also a physically substantial discrete card with a 415 W TDP, dual-slot cooler, and PCIe 5.0 x16 interface, appropriate for a desktop system with room for a large 304 mm card.

The Arc B390 is an integrated processor with an 80 W TDP and no power connectors, designed for portable devices. Its 1536 shading units, 12 RT cores, and 7.680 TFLOPS of FP32 throughput are adequate for entry-level workloads, and its nearest rivals in the database are legacy NVIDIA parts from the GT 500, 800, and 700 series. The 3 nm Intel process node and Xe3-LPG architecture represent a modern design, but the performance envelope is fundamentally constrained by its integrated nature and shared memory subsystem.

The data indicates that the RTX 5080 SUPER is the choice for workloads requiring high compute throughput, dedicated memory bandwidth, and discrete card expandability. The Arc B390 is the choice for systems where an integrated GPU with low power draw is required, and where performance expectations align with the 9th percentile tier. Neither part competes in the other's intended segment. The RTX 5080 SUPER's nearest rivals include professional and workstation parts within a 3.6% band, while the Arc B390's nearest rivals are all entry-level parts within a 2.7% band. The performance gap between the two is not a matter of tuning or driver maturity; it is a structural difference in design class.

Specification Differences

| Specification | Intel Arc 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 |

| RT 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 | 7.680 TFLOPS | 56.28 TFLOPS |

| FP16 | 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 |

| 3DMark Steel Nomad DX12 | 1482 | 3075 |

| Percentile vs All GPUs | 9 | 19 |

| Launch MSRP | not recorded | 999 USD |

The specification table confirms the architectural narrative. Every compute, memory, and power figure favors the RTX 5080 SUPER except for the process node, where Intel's 3 nm process is smaller than TSMC's 5 nm process. The launch MSRP of 999 USD applies only to the RTX 5080 SUPER; no MSRP is recorded for the Arc B390, consistent with its integrated nature. The release dates are close, with the RTX 5080 SUPER released on 2025-12-31 and the Arc B390 on 2026-01-26. Both parts are active in production.