AMD Radeon AI PRO R9700 vs Intel Arc B390 Comparison

Where Each One Wins

The recorded benchmark data contains only one shared test between the AMD Radeon AI PRO R9700 and the Intel Arc B390: 3DMark Steel Nomad DX12. In that single comparison, the AMD part wins decisively. The AMD Radeon AI PRO R9700 scores 7006, while the Intel Arc B390 scores 1482, a delta of 372.7% in favor of the AMD product. Every available head-to-head benchmark result therefore belongs to the AMD card.

The AMD Radeon AI PRO R9700 also carries a much broader benchmark profile. The database includes results for Geekbench OpenCL, Geekbench Vulkan, Passmark DirectX 10, DirectX 11, DirectX 12, DirectX 9, G2D, G3D, and GPU Compute. Across those tests, its average benchmark score is 23315, and its percentile versus all GPUs is 69. The Intel Arc B390, by contrast, has only the single Steel Nomad result, an average score of 1482, and a percentile of 9. The data shows no test in which the Intel part outperforms the AMD part.

The AMD card's nearest rivals in the database are not the Intel part. The Radeon AI PRO R9700 sits within 0.3% of the AMD Radeon R9 M290X, the AMD Radeon RX 6600M, the AMD Radeon Pro Vega 16, and the NVIDIA P106-100. The Intel Arc B390, by contrast, sits within 2.7% of the NVIDIA GeForce GT 520MX, the NVIDIA GeForce 800M, the NVIDIA GeForce GT 625 OEM, and the NVIDIA GeForce GT 710. These rival groups indicate the two products occupy entirely different performance strata.

Architecture Differences

The AMD Radeon AI PRO R9700 uses the Navi 48 chip with RDNA 4.0 architecture, fabricated on a 4 nm process at TSMC. The Intel Arc B390 uses the Panther Lake chip with Xe3-LPG architecture, fabricated on a 3 nm process at Intel. The AMD part integrates 53,900 million transistors on a 357 mm² die, with a transistor density of 151.0 million per mm². The Intel part's transistor count and die size are recorded as unknown.

The AMD card has 4096 shading units, 256 TMUs, and 128 ROPs. It includes 64 ray tracing cores. The Intel part has 1536 shading units, 48 TMUs, and 24 ROPs, with 12 ray tracing cores. The AMD card's FP32 throughput is 47.84 TFLOPS, while the Intel part delivers 7.680 TFLOPS. The FP16 figures differ in ratio: the AMD card runs FP16 at 47.84 TFLOPS (1:1), while the Intel part reaches 15.36 TFLOPS (2:1).

Memory architecture separates the two sharply. The AMD Radeon AI PRO R9700 uses 32 GB of GDDR6 on a 256-bit bus, with 644.6 GB/s of bandwidth and a memory clock of 2518 MHz (20.1 Gbps effective). The Intel Arc B390 uses system shared memory, with system dependent bandwidth and no dedicated memory clock. Its bus width is also recorded as system shared.

Clock behavior also differs. The AMD card has a base clock of 1660 MHz, a boost clock of 2920 MHz, and a game clock of 2350 MHz. The Intel part has a base clock of 300 MHz and a boost clock of 2500 MHz, with no recorded game clock. Pixel rate and texture rate follow the same pattern: the AMD card posts 373.8 GPixel/s and 747.5 GTexel/s, while the Intel part posts 60.00 GPixel/s and 120.0 GTexel/s.

Power and physical design diverge as well. The AMD Radeon AI PRO R9700 has a TDP of 300 W, uses a dual-slot cooler, requires a single 16-pin power connector, and suggests a 700 W power supply. The Intel Arc B390 has a TDP of 80 W, is an IGP (integrated graphics processor), requires no power connectors, and has no suggested PSU. The AMD card is 267 mm long, 111 mm tall, and 40 mm wide. The Intel part has no recorded dimensions. The AMD card uses PCIe 5.0 x16, while the Intel part uses an IGP bus interface.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs differ: the AMD card provides 1x HDMI 2.1b and 3x DisplayPort 2.1a, while the Intel part's outputs are recorded as portable device dependent.

The Verdict

The data indicates a clear performance hierarchy. The AMD Radeon AI PRO R9700 outscores the Intel Arc B390 by 372.7% in the only shared benchmark test. The AMD part's average benchmark score of 23315 places it in the 69th percentile of all GPUs, while the Intel part's average of 1482 places it in the 9th percentile. The AMD card's nearest rivals are all substantially faster than the Intel card's nearest rivals.

The AMD Radeon AI PRO R9700 is a discrete, dual-slot, 300 W card with 32 GB of dedicated GDDR6 memory, 4096 shading units, and 64 ray tracing cores. It targets workloads that require high FP32 throughput, high memory bandwidth, and a full PCIe 5.0 x16 interface. The Intel Arc B390 is an integrated graphics solution within the Panther Lake chip, using system shared memory, 1536 shading units, and 12 ray tracing cores, with an 80 W TDP and no power connectors.

The release dates differ: the AMD card was released on 2025-07-22, while the Intel part is dated 2026-01-26. The AMD card's launch MSRP is 1,299 USD. The Intel part has no recorded launch MSRP.

The recorded data does not show any workload where the Intel part wins. The AMD card wins the single available head-to-head test, and its architecture provides more shading units, more TMUs, more ROPs, more ray tracing cores, dedicated memory, and higher clock rates. The Intel part's advantages are limited to power efficiency (80 W vs 300 W) and a smaller process node (3 nm vs 4 nm), but no benchmark data confirms a performance benefit from those characteristics.

FAQ

Q: Which GPU wins the only directly comparable benchmark?

A: The AMD Radeon AI PRO R9700 scores 7006 in 3DMark Steel Nomad DX12, while the Intel Arc B390 scores 1482. The AMD part leads by 372.7%.

Q: How do the average benchmark scores compare?

A: The AMD Radeon AI PRO R9700 has an average benchmark score of 23315, while the Intel Arc B390 has an average of 1482.

Q: What memory configurations do the two GPUs use?

A: The AMD Radeon AI PRO R9700 uses 32 GB of GDDR6 on a 256-bit bus with 644.6 GB/s bandwidth. The Intel Arc B390 uses system shared memory with system dependent bandwidth.

Q: What are the shading unit counts?

A: The AMD Radeon AI PRO R9700 has 4096 shading units, while the Intel Arc B390 has 1536 shading units.

Q: What is the power draw difference?

A: The AMD Radeon AI PRO R9700 has a TDP of 300 W, while the Intel Arc B390 has a TDP of 80 W.

Q: Do both GPUs support the same APIs?

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

Head-to-Head Benchmarks

The only direct comparison in the database is 3DMark Steel Nomad DX12. The AMD Radeon AI PRO R9700 scores 7006. The Intel Arc B390 scores 1482. The delta is 372.7% in favor of the AMD part. This is not a marginal lead; it is a difference of approximately 4.7 times the Intel part's score.

The AMD card's broader benchmark profile reinforces its position. In Geekbench OpenCL, it scores 118202. In Geekbench Vulkan, it scores 63550. Passmark results range from 83 in DirectX 12 to 363 in DirectX 9, with 155 in DirectX 10, 272 in DirectX 11, 1236 in G2D, 26979 in G3D, and 15300 in GPU Compute. These numbers cannot be compared directly to the Intel part because the Intel part has no corresponding entries in the database.

The Intel Arc B390's nearest rivals in the database are all low-end parts. The NVIDIA GeForce GT 520MX sits 1.3% behind, the NVIDIA GeForce 800M sits 1.5% behind, the NVIDIA GeForce GT 625 OEM sits 2.5% behind, and the NVIDIA GeForce GT 710 sits 2.7% behind. The AMD card's nearest rivals, by contrast, are the AMD Radeon R9 M290X (0.2% behind), the AMD Radeon RX 6600M (0.2% behind), the AMD Radeon Pro Vega 16 (0.3% behind), and the NVIDIA P106-100 (0.3% behind). The performance gap between the two products is therefore not subtle; it spans entirely different segments of the GPU landscape.

The AMD card's FP32 throughput of 47.84 TFLOPS is roughly 6.2 times the Intel part's 7.680 TFLOPS. The AMD card's texture rate of 747.5 GTexel/s is roughly 6.2 times the Intel part's 120.0 GTexel/s. The AMD card's pixel rate of 373.8 GPixel/s is roughly 6.2 times the Intel part's 60.00 GPixel/s. These ratios align with the shading unit count difference (4096 vs 1536, a 2.67x ratio) but are amplified by the clock advantage.

Specification Differences

The AMD Radeon AI PRO R9700 and Intel Arc B390 differ in nearly every recorded specification. The chip, architecture, process node, and foundry are all different: Navi 48 with RDNA 4.0 on 4 nm at TSMC versus Panther Lake with Xe3-LPG on 3 nm at Intel. The AMD card has 53,900 million transistors on a 357 mm² die; the Intel part's transistor count and die size are unknown.

Shading units: 4096 vs 1536. TMUs: 256 vs 48. ROPs: 128 vs 24. Ray tracing cores: 64 vs 12. The AMD card's FP32 is 47.84 TFLOPS; the Intel part's is 7.680 TFLOPS. FP16 is 47.84 TFLOPS (1:1) for AMD and 15.36 TFLOPS (2:1) for Intel. Pixel rate is 373.8 GPixel/s vs 60.00 GPixel/s. Texture rate is 747.5 GTexel/s vs 120.0 GTexel/s.

Memory: 32 GB GDDR6 on a 256-bit bus with 644.6 GB/s bandwidth vs system shared memory with system dependent bandwidth. Clocks: base 1660 MHz vs 300 MHz, boost 2920 MHz vs 2500 MHz, game 2350 MHz vs no recorded game clock. Memory clock: 2518 MHz (20.1 Gbps effective) vs no dedicated memory clock.

TDP: 300 W vs 80 W. Slot width: dual-slot vs IGP. Power connectors: 1x 16-pin vs none. Suggested PSU: 700 W vs none recorded. Bus interface: PCIe 5.0 x16 vs IGP. Display outputs: 1x HDMI 2.1b and 3x DisplayPort 2.1a vs portable device dependent.

Dimensions: 267 mm length, 111 mm height, 40 mm width vs no recorded dimensions. Release date: 2025-07-22 vs 2026-01-26. Launch MSRP: 1,299 USD vs none recorded. Predecessor: Radeon Pro Vega vs none recorded. Production status: both active.

Shared specifications include DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both have no recorded tensor cores. Both have no recorded codename. The AMD card's generation is listed as Radeon Pro Navi (Navi IV Series), while the Intel part's generation is listed as Arc Graphics-M (Panther Lake).