AMD Radeon RX 7800 XT vs Intel Arc B390 Comparison

Head-to-Head Benchmarks

The recorded data contains a single shared benchmark between the AMD Radeon RX 7800 XT and the Intel Arc B390: 3DMark Steel Nomad (DX12). The AMD card scores 4157, while the Intel part scores 1482. That is a delta of 180.5%, meaning the Radeon delivers roughly 2.8 times the frame throughput in this test. The margin is enormous, and it aligns with the broader positioning of the two products: the RX 7800 XT is a discrete desktop GPU with a 263 W TDP, while the Arc B390 is an integrated processor graphics solution with an 80 W TDP.

The AMD card's 3DMark result places it in the 51st percentile among all GPUs in the database. Its nearest rivals in average benchmark score are the NVIDIA GeForce GTX 1660 (11680, delta -0.5%), the AMD Radeon RX 6500 XT (11842, delta -1.8%), the AMD Radeon Pro 5500M (11528, delta 0.9%), and the NVIDIA Tesla K20c (11479, delta 1.3%). The RX 7800 XT's average benchmark score across all recorded tests is 11627. This puts it in a cluster with older mainstream cards, which is curious given its peak 3DMark result. The high variance between tests suggests the card's performance is highly workload-dependent.

The Intel Arc B390, by contrast, sits at the 9th percentile overall. Its average benchmark score is 1482, which is the same as its single 3DMark result. Its nearest rivals include the NVIDIA GeForce GT 520MX (1463, delta 1.3%), the NVIDIA GeForce 800M (1460, delta 1.5%), the NVIDIA GeForce GT 625 OEM (1446, delta 2.5%), and the NVIDIA GeForce GT 710 (1443, delta 2.7%). These are all entry-level or legacy discrete parts, and the B390 edges past them by narrow margins. The data indicates the B390 is not competing in the same performance class as the RX 7800 XT, despite its modern architecture.

The only direct comparison, 3DMark Steel Nomad, shows the AMD card ahead by 180.5%. No other shared tests exist in the database, so the head-to-head picture is limited to this single DX12 workload. Even so, the scale of the difference is consistent with the hardware specifications: the RX 7800 XT has 3840 shading units, 240 TMUs, and 96 ROPs, while the B390 has 1536 shading units, 48 TMUs, and 24 ROPs. Pixel rate for the AMD card is 233.3 GPixel/s versus 60.00 GPixel/s for Intel. Texture rate is 583.2 GTexel/s versus 120.0 GTexel/s. FP32 compute is 37.32 TFLOPS versus 7.680 TFLOPS.

Where Each One Wins

The AMD Radeon RX 7800 XT wins the only recorded head-to-head benchmark, and it also holds a decisive advantage in every raw throughput metric available in the database. Its FP32 output of 37.32 TFLOPS is roughly 4.9 times the Intel part's 7.680 TFLOPS. The AMD card's FP16 performance is identical at 37.32 TFLOPS (1:1 ratio), while the Intel part reaches 15.36 TFLOPS with a 2:1 ratio. That means in workloads that can use FP16, the B390 narrows the gap somewhat, but it still trails by a factor of about 2.4.

Memory bandwidth heavily favors the AMD card. The RX 7800 XT has 624.1 GB/s of bandwidth over a 256-bit GDDR6 bus, while the B390 uses system shared memory with bandwidth labeled as system dependent. The Intel part has no dedicated VRAM, no fixed bus width, and no independent memory clock. This makes the B390 more sensitive to system memory configuration, and the database records its memory bandwidth as dependent on the host platform. The AMD card's 16 GB of GDDR6 is a fixed resource, which removes a major variable from performance analysis.

For compute-heavy tasks, the PassMark GPU Compute score for the RX 7800 XT is 13473, which is far above the Intel part's single 3DMark result of 1482. The RX 7800 XT also has a PassMark G3D score of 24176 and a G2D score of 1177. The B390 has no recorded PassMark results, so direct comparison in those tests is impossible. However, the absence of data does not change the pattern: the AMD card dominates in every metric where both are measured.

The Intel Arc B390 does hold some structural advantages. It is an integrated GPU with no power connectors, no slot width, and no suggested PSU requirement. Its TDP is 80 W, which is 183 W lower than the RX 7800 XT's 263 W. It uses a 3 nm process node from Intel, while the AMD card uses a 5 nm node from TSMC. The B390 also has a higher boost clock at 2500 MHz versus 2430 MHz for the AMD part, though the AMD card's base clock is much higher at 1295 MHz versus 300 MHz. These advantages matter for power-constrained or thermally constrained environments, but they do not translate into a performance win in the recorded benchmark data.

Architecture Differences

The two GPUs come from different architectural generations and design philosophies. The AMD Radeon RX 7800 XT uses the RDNA 3.0 architecture, built on the Navi 32 chip, with the codename "Wheat Nas." It belongs to the Navi III (RX 7000) generation and succeeds the Navi II line. The Intel Arc B390 uses the Xe3-LPG architecture, built on the Panther Lake chip, and belongs to the Arc Graphics-M (Panther Lake) generation. The AMD part is manufactured by TSMC on a 5 nm process, while the Intel part is manufactured by Intel on a 3 nm process.

The AMD chip contains 28,100 million transistors on a 346 mm² die, giving a transistor density of 81.2 million per square millimeter. The Intel chip's transistor count and die size are listed as unknown in the database, so no direct density comparison is possible. The AMD card is a discrete, dual-slot PCIe 4.0 x16 add-in board with two 8-pin power connectors and a suggested PSU of 600 W. The Intel part is an IGP with no slot width, no power connectors, no dimensions, and no PSU recommendation. Its bus interface is listed simply as IGP.

Memory architecture is fundamentally different. The RX 7800 XT has 16 GB of GDDR6 on a 256-bit bus, with memory clocked at 2438 MHz (19.5 Gbps effective) and bandwidth of 624.1 GB/s. The B390 uses system shared memory, with the type, bus width, and bandwidth all listed as system dependent or shared. This means the B390's memory performance is not fixed and depends entirely on the host laptop or portable device. The AMD card's memory subsystem is self-contained and predictable.

Compute resources differ substantially. The RX 7800 XT has 3840 shading units, 240 TMUs, 96 ROPs, and 60 ray tracing cores. The B390 has 1536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. The AMD card has 2.5 times the shading units, 5 times the TMUs, 4 times the ROPs, and 5 times the ray tracing cores. Neither card lists tensor cores in the database. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity is present.

Pixel rate for the RX 7800 XT is 233.3 GPixel/s, and texture rate is 583.2 GTexel/s. The B390 manages 60.00 GPixel/s and 120.0 GTexel/s. The AMD card's FP32 rate is 37.32 TFLOPS with 1:1 FP16, while the B390's FP32 is 7.680 TFLOPS with 2:1 FP16. The B390's FP16 advantage comes from its 2:1 ratio, which allows it to double output when using half-precision workloads. The RX 7800 XT does not gain this benefit, but its raw FP16 output still exceeds the Intel part.

The Verdict

The recorded data is unambiguous: the AMD Radeon RX 7800 XT outperforms the Intel Arc B390 by 180.5% in the only shared benchmark, 3DMark Steel Nomad (DX12). The AMD card also holds commanding leads in shading units, TMUs, ROPs, ray tracing cores, pixel rate, texture rate, FP32 compute, and memory bandwidth. Its 16 GB GDDR6 buffer with 624.1 GB/s of bandwidth is a fixed, high-bandwidth resource, whereas the B390 depends on system shared memory.

The Intel Arc B390 is not a competitor in the same performance class. Its 1482 score in 3DMark places it near legacy parts like the NVIDIA GeForce GT 710 and GT 625 OEM. Its nearest rivals all have average scores between 1443 and 1463, and the B390 beats them by only 1.3% to 2.7%. The RX 7800 XT, meanwhile, sits near the NVIDIA GeForce GTX 1660 and AMD Radeon RX 6500 XT in average score, with deltas under 2%. The AMD card's percentile rank of 51 versus the Intel part's 9 reflects this gap.

For a user seeking maximum performance in DX12 workloads, the RX 7800 XT is the clear choice. For a system requiring minimal power draw, no external power connectors, and integrated graphics, the B390 is the only option that fits those constraints. The B390's 80 W TDP and 3 nm process node indicate efficiency is its design priority, not absolute performance. The RX 7800 XT's 263 W TDP and 2x 8-pin connectors show it is built for throughput, not portability.

The data does not support any scenario where the B390 wins a performance comparison. Its only advantages are structural: lower power, no power connectors, integrated form factor, and a higher boost clock. None of these translate into a benchmark win. The verdict from the database is straightforward: the AMD Radeon RX 7800 XT is the superior performer, and the Intel Arc B390 serves a different, power-limited purpose.

FAQ

Q: How much faster is the AMD Radeon RX 7800 XT than the Intel Arc B390 in 3DMark Steel Nomad?

A: The RX 7800 XT scores 4157, while the B390 scores 1482. The delta is 180.5% in favor of AMD.

Q: What is the average benchmark score for each GPU?

A: The RX 7800 XT has an average benchmark score of 11627 across all recorded tests. The B390 has an average score of 1482, which matches its single 3DMark result.

Q: Which GPU has more shading units?

A: The RX 7800 XT has 3840 shading units, while the B390 has 1536. The AMD card also has 240 TMUs versus 48, and 96 ROPs versus 24.

Q: What memory configuration does each GPU use?

A: The RX 7800 XT uses 16 GB of GDDR6 on a 256-bit bus with 624.1 GB/s bandwidth. The B390 uses system shared memory, with type, bus width, and bandwidth all listed as system dependent.

Q: What is the TDP difference between the two cards?

A: The RX 7800 XT has a TDP of 263 W and requires two 8-pin power connectors. The B390 has a TDP of 80 W and uses no power connectors.

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.

Specification Differences

| Specification | AMD Radeon RX 7800 XT | Intel Arc B390 |

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

| Architecture | RDNA 3.0 | Xe3-LPG |

| Chip | Navi 32 | Panther Lake |

| Process Node | 5 nm | 3 nm |

| Foundry | TSMC | Intel |

| Transistors | 28,100 million | Unknown |

| Die Size | 346 mm² | Unknown |

| Base Clock | 1295 MHz | 300 MHz |

| Boost Clock | 2430 MHz | 2500 MHz |

| Memory Size | 16 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 256 bit | System Shared |

| Memory Bandwidth | 624.1 GB/s | System Dependent |

| Shading Units | 3840 | 1536 |

| TMUs | 240 | 48 |

| ROPs | 96 | 24 |

| Ray Tracing Cores | 60 | 12 |

| Pixel Rate | 233.3 GPixel/s | 60.00 GPixel/s |

| Texture Rate | 583.2 GTexel/s | 120.0 GTexel/s |

| FP32 | 37.32 TFLOPS | 7.680 TFLOPS |

| FP16 | 37.32 TFLOPS (1:1) | 15.36 TFLOPS (2:1) |

| TDP | 263 W | 80 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 2x 8-pin | None |

| Suggested PSU | 600 W | None |

| Bus Interface | PCIe 4.0 x16 | IGP |

| Release Date | 2023-09-05 | 2026-01-26 |

| Launch MSRP | 499 USD | None |

| Percentile | 51 | 9 |