Intel Arc B390 vs NVIDIA GeForce RTX 5090 D V2 Comparison

The Verdict

The recorded benchmark data presents a decisive outcome: the NVIDIA GeForce RTX 5090 D V2 is overwhelmingly faster than the Intel Arc B390 in the single tested workload. In the 3DMark Steel Nomad DX12 test, the RTX 5090 D V2 scores 16,504 points against the Arc B390's 1,482 points, a delta of 91 percent in favor of NVIDIA. The Arc B390 sits at the 9th percentile among all GPUs in the database, while the RTX 5090 D V2 reaches the 59th percentile. These are not comparable products for any performance-oriented use case.

The Intel Arc B390 is an integrated graphics processor (IGP) with a 3 nm process node, 80 W power draw, and no dedicated memory. Its benchmark placement places it alongside legacy entry-level discrete GPUs such as the NVIDIA GeForce GT 520MX (1,463 average score, 1.3 percent behind) and the GeForce 800M (1,460, 1.5 percent behind). The RTX 5090 D V2, by contrast, belongs to the GeForce 50-series with a 5 nm process node from TSMC, 92,200 million transistors, 24 GB of GDDR7 memory, and a 575 W power draw. Its nearest rivals in the database include the NVIDIA T400 (16,508, 0 percent delta), AMD Radeon PRO W7500 (16,415, 0.5 percent), and NVIDIA RTX PRO 6000 Blackwell (16,408, 0.6 percent).

The verdict is straightforward: the RTX 5090 D V2 is the only choice for demanding graphics workloads. The Arc B390 should be considered solely for systems where integrated graphics are acceptable, such as basic display output or low-intensity tasks. No benchmark result suggests otherwise.

Where Each One Wins

The head-to-head data records one benchmark: 3DMark Steel Nomad DX12. The RTX 5090 D V2 wins that test outright with a score of 16,504 versus 1,482, a 91 percent advantage. The Arc B390 has zero wins in the comparison set.

The Arc B390's strength is not in raw performance but in integration. It uses system-shared memory, has no power connectors, occupies an IGP slot width, and draws only 80 W. These characteristics make it suitable for compact or portable devices where a discrete GPU cannot fit. Its display outputs are portable-device dependent, reinforcing its role in mobile or embedded systems.

The RTX 5090 D V2 wins every measurable performance category. It delivers 104.8 TFLOPS FP32 against 7.680 TFLOPS for the Arc B390. Its pixel rate is 423.6 GPixel/s versus 60.00 GPixel/s. Its texture rate is 1,636.8 GTexel/s versus 120.0 GTexel/s. Memory bandwidth is 1.34 TB/s from a 384-bit GDDR7 interface, while the Arc B390's bandwidth is system dependent. The RTX 5090 D V2 also carries 170 RT cores and 680 tensor cores; the Arc B390 has 12 RT cores and no listed tensor core count.

For users who need maximum frame rates, high-resolution rendering, or ray-traced workloads, the RTX 5090 D V2 is the clear winner. For users who need a graphics solution embedded in a processor with minimal power draw, the Arc B390 exists, but its benchmark position does not support any performance-oriented claim.

Architecture Differences

The Intel Arc B390 uses the Panther Lake chip with Xe3-LPG architecture, part of the Arc Graphics-M (Panther Lake) generation. It is fabricated on a 3 nm process by Intel. The transistor count and die size are unknown in the database. The GPU contains 1,536 shading units, 48 texture mapping units, 24 ROPs, and 12 RT cores. Its base clock is 300 MHz with a boost clock of 2,500 MHz. Memory is system shared, meaning capacity, type, bus width, and bandwidth all depend on the host system.

The NVIDIA GeForce RTX 5090 D V2 uses the GB202 chip with Blackwell 2.0 architecture, part of the GeForce 50 generation. It is fabricated on a 5 nm process by TSMC. The die measures 750 mm² with 92,200 million transistors, yielding a transistor density of 122.9M per mm². The GPU contains 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. Its base clock is 2,017 MHz with a boost clock of 2,407 MHz. Memory is 24 GB of GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth.

The architecture gap is massive. The RTX 5090 D V2 has roughly 14 times the shading units, 14 times the TMUs, and 7 times the ROPs of the Arc B390. Its RT core count is over 14 times higher. The FP32 throughput difference is about 14 times. FP16 performance is 104.8 TFLOPS for NVIDIA, while the Arc B390 delivers 15.36 TFLOPS at a 2:1 ratio. The RTX 5090 D V2 runs FP16 at a 1:1 ratio, meaning no halving of throughput.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 5090 D V2 uses a PCIe 5.0 x16 interface; the Arc B390 is an IGP. The NVIDIA card uses a dual-slot design, measures 304 mm by 137 mm by 48 mm, and requires a 16-pin power connector plus a 950 W suggested PSU. The Arc B390 has no power connectors and no suggested PSU. The NVIDIA card's display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b; the Arc B390's outputs depend on the portable device.

FAQ

Q: Which GPU has a higher benchmark score in 3DMark Steel Nomad DX12?

A: The NVIDIA GeForce RTX 5090 D V2 scores 16,504, while the Intel Arc B390 scores 1,482. The delta is 91 percent in favor of NVIDIA.

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

A: The Arc B390's average score of 1,482 puts it 1.3 percent ahead of the NVIDIA GeForce GT 520MX (1,463), 1.5 percent ahead of the GeForce 800M (1,460), 2.5 percent ahead of the GT 625 OEM (1,446), and 2.7 percent ahead of the GT 710 (1,443).

Q: What are the memory specifications of each GPU?

A: The RTX 5090 D V2 has 24 GB of GDDR7 memory on a 384-bit bus with 1.34 TB/s bandwidth. The Arc B390 uses system-shared memory, so its size, type, bus width, and bandwidth are all system dependent.

Q: What is the power draw difference?

A: The Arc B390 has a TDP of 80 W and uses no power connectors. The RTX 5090 D V2 has a TDP of 575 W, uses a single 16-pin connector, and has a suggested PSU of 950 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.

Q: What is the RT core count for each GPU?

A: The RTX 5090 D V2 has 170 RT cores. The Arc B390 has 12 RT cores.

Head-to-Head Benchmarks

The only recorded head-to-head benchmark is 3DMark Steel Nomad DX12. The RTX 5090 D V2 scores 16,504 points. The Arc B390 scores 1,482 points. The delta is 91 percent, meaning the Arc B390 delivers only about 9 percent of the NVIDIA card's score.

This single result dominates the comparison. A 91 percent gap in a DX12 workload indicates that the Arc B390 cannot compete in any modern gaming or compute scenario that relies on GPU throughput. The RTX 5090 D V2's nearest rivals in the database, such as the NVIDIA T400 (16,508, 0 percent delta) and the AMD Radeon PRO W7500 (16,415, 0.5 percent), all cluster near its score. The Arc B390's nearest rivals are decades-old or low-end parts: the GT 520MX (1,463, 1.3 percent behind) and the GeForce 800M (1,460, 1.5 percent behind).

The performance per watt also favors NVIDIA decisively. The RTX 5090 D V2 draws 575 W and scores 16,504, which is about 28.7 points per watt. The Arc B390 draws 80 W and scores 1,482, which is about 18.5 points per watt. Even on an efficiency basis, the NVIDIA card is far ahead.

In terms of raw compute, the FP32 throughput difference is 104.8 TFLOPS versus 7.680 TFLOPS, roughly a 13.6 times gap. The texture rate is 1,636.8 GTexel/s versus 120.0 GTexel/s, a 13.6 times gap. The pixel rate is 423.6 GPixel/s versus 60.00 GPixel/s, a 7 times gap. These figures align with the benchmark delta and confirm that the Arc B390 is not a competitor in any GPU-bound task.

Specification Differences

The two GPUs differ in nearly every measurable specification.

• Process node: Intel Arc B390 uses 3 nm; NVIDIA RTX 5090 D V2 uses 5 nm.
• Foundry: Intel for the Arc B390; TSMC for the RTX 5090 D V2.
• Transistors: Unknown for the Arc B390; 92,200 million for the RTX 5090 D V2.
• Die size: Unknown for the Arc B390; 750 mm² for the RTX 5090 D V2.
• Transistor density: Not listed for the Arc B390; 122.9M per mm² for the RTX 5090 D V2.
• Base clock: 300 MHz for the Arc B390; 2,017 MHz for the RTX 5090 D V2.
• Boost clock: 2,500 MHz for the Arc B390; 2,407 MHz for the RTX 5090 D V2.
• Memory size: System shared for the Arc B390; 24 GB for the RTX 5090 D V2.
• Memory type: System shared for the Arc B390; GDDR7 for the RTX 5090 D V2.
• Memory bus width: System shared for the Arc B390; 384 bit for the RTX 5090 D V2.
• Memory bandwidth: System dependent for the Arc B390; 1.34 TB/s for the RTX 5090 D V2.
• Shading units: 1,536 for the Arc B390; 21,760 for the RTX 5090 D V2.
• TMUs: 48 for the Arc B390; 680 for the RTX 5090 D V2.
• ROPs: 24 for the Arc B390; 176 for the RTX 5090 D V2.
• RT cores: 12 for the Arc B390; 170 for the RTX 5090 D V2.
• Tensor cores: Not listed for the Arc B390; 680 for the RTX 5090 D V2.
• Pixel rate: 60.00 GPixel/s for the Arc B390; 423.6 GPixel/s for the RTX 5090 D V2.
• Texture rate: 120.0 GTexel/s for the Arc B390; 1,636.8 GTexel/s for the RTX 5090 D V2.
• FP32: 7.680 TFLOPS for the Arc B390; 104.8 TFLOPS for the RTX 5090 D V2.
• FP16: 15.36 TFLOPS (2:1) for the Arc B390; 104.8 TFLOPS (1:1) for the RTX 5090 D V2.
• TDP: 80 W for the Arc B390; 575 W for the RTX 5090 D V2.
• Slot width: IGP for the Arc B390; Dual-slot for the RTX 5090 D V2.
• Power connectors: None for the Arc B390; 1x 16-pin for the RTX 5090 D V2.
• Suggested PSU: Not listed for the Arc B390; 950 W for the RTX 5090 D V2.
• Bus interface: IGP for the Arc B390; PCIe 5.0 x16 for the RTX 5090 D V2.
• Display outputs: Portable device dependent for the Arc B390; 1x HDMI 2.1b and 3x DisplayPort 2.1b for the RTX 5090 D V2.
• Dimensions: Not listed for the Arc B390; 304 mm by 137 mm by 48 mm for the RTX 5090 D V2.
• Release date: January 26, 2026 for the Arc B390; August 14, 2025 for the RTX 5090 D V2.
• Launch MSRP: Not available for the Arc B390; 2,299 USD for the RTX 5090 D V2.

The RTX 5090 D V2 is also listed with a predecessor (GeForce 40) and successor (GeForce 60) in the database, while the Arc B390 has neither listed. The production status for both is active.