Intel Arc Pro B390 vs NVIDIA RTX PRO 6000 Blackwell Comparison

Head-to-Head Benchmarks

The recorded database contains no direct head-to-head benchmark comparisons between the Intel Arc Pro B390 and the NVIDIA RTX PRO 6000 Blackwell. The headToHeadBenchmarks array is empty, and the Intel part has no benchmark entries at all. Its avgBenchmarkScore is listed as 0, and its percentileVsAllGpus sits at 50. The NVIDIA card, by contrast, has a single recorded benchmark result: 3DMark Steel Nomad DX12, where it scored 16,408 points.

That score places the RTX PRO 6000 Blackwell at the 59th percentile among all GPUs in the database. Its nearest rivals in that test show how tightly clustered the results are. The AMD Radeon PRO W7500 scores 16,415, a delta of 0 percent. The AMD Radeon RX 5700 XT scores 16,361, a delta of 0.3 percent. The AMD Radeon Pro 5600M scores 16,351, a delta of 0.4 percent. The NVIDIA GeForce RTX 5090 D V2 scores 16,504, which is 0.6 percent higher, giving it a delta of -0.6 percent from the NVIDIA card's perspective. The RTX PRO 6000 Blackwell is effectively within a rounding error of its closest competitors in this single test, trailing the RTX 5090 D V2 by less than one percent.

Because the Intel Arc Pro B390 has no recorded benchmark scores, no wins can be credited to either side. The winsA and winsB fields both read 0. The data simply does not permit a meaningful performance comparison between these two products. The NVIDIA part is benchmarked and ranked; the Intel part is not. Any attempt to extrapolate relative performance from the available numbers would exceed what the database supports.

Architecture Differences

The two GPUs represent fundamentally different design philosophies and manufacturing approaches. The Intel Arc Pro B390 uses the Panther Lake chip with the Xe3-LPG architecture, built on a 3 nm process at Intel's own foundry. The NVIDIA RTX PRO 6000 Blackwell uses the GB202 chip with the Blackwell 2.0 architecture, built on a 5 nm process at TSMC. Intel's transistor count and die size are listed as unknown; NVIDIA's are documented at 92,200 million transistors across a 750 mm² die, producing a transistor density of 122.9M per mm².

The memory subsystems could hardly be more different. The Intel part uses system shared memory, with its size, type, and bus width all listed as System Shared. Its bandwidth is System Dependent. The NVIDIA card carries 96 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s of bandwidth. The Intel part's memory clock is also System Shared, while the NVIDIA card runs at 1750 MHz with 28 Gbps effective.

Compute resources show a similar gulf. The Intel Arc Pro B390 has 1,536 shading units, 48 texture mapping units, 24 ROPs, and 12 ray tracing cores. It has no tensor cores listed. The NVIDIA RTX PRO 6000 Blackwell has 24,064 shading units, 752 TMUs, 192 ROPs, 188 RT cores, and 752 tensor cores. The raw throughput figures follow: Intel delivers 60.00 GPixel/s pixel rate, 120.0 GTexel/s texture rate, 7.680 TFLOPS FP32, and 15.36 TFLOPS FP16 at a 2:1 ratio. NVIDIA delivers 502.5 GPixel/s, 1,968.0 GTexel/s, 126.0 TFLOPS FP32, and 126.0 TFLOPS FP16 at a 1:1 ratio.

Clock behavior also differs. The Intel GPU has a base clock of 300 MHz and a boost clock of 2500 MHz. The NVIDIA GPU has a base clock of 1590 MHz and a boost clock of 2617 MHz. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API coverage is identical.

Power and physical design are in different leagues. The Intel part is an integrated GPU with an 80 W TDP, no power connectors, and an IGP slot width. The NVIDIA part is a dual-slot add-in card with a 600 W TDP, a single 16-pin power connector, a suggested 1000 W power supply, and a PCIe 5.0 x16 bus interface. Its dimensions are 304 mm in length, 137 mm in height, and 40 mm in width. Display outputs differ as well: the Intel part uses Portable Device Dependent outputs, while the NVIDIA card provides 4x DisplayPort 2.1b.

Release timing shows the NVIDIA card arrived first, with a release date of March 17, 2025, succeeding Workstation Ada. The Intel part followed with a release date of January 26, 2026, succeeding HD Graphics-WM. Both are listed as Active in production status. Neither has a listed successor.

FAQ

Q: Which GPU has more shading units?

A: The NVIDIA RTX PRO 6000 Blackwell has 24,064 shading units, compared to 1,536 on the Intel Arc Pro B390.

Q: What memory configuration does each card use?

A: The Intel Arc Pro B390 uses System Shared memory, so its size, type, bus width, and bandwidth are all dependent on the host platform. The NVIDIA RTX PRO 6000 Blackwell has 96 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth.

Q: How do their power requirements compare?

A: The Intel Arc Pro B390 has an 80 W TDP, needs no power connectors, and is an integrated GPU. The NVIDIA RTX PRO 6000 Blackwell has a 600 W TDP, requires one 16-pin power connector, and lists a suggested power supply of 1000 W.

Q: Does either GPU support DirectX 12 Ultimate?

A: Both support DirectX 12 Ultimate (12_2). They also both support OpenGL 4.6 and Vulkan 1.4.

Q: What is the only recorded benchmark score between the two?

A: The NVIDIA RTX PRO 6000 Blackwell has a 3DMark Steel Nomad DX12 score of 16,408. The Intel Arc Pro B390 has no benchmark scores recorded in the database.

Q: What are the process nodes for each chip?

A: The Intel Arc Pro B390 uses a 3 nm process at Intel's foundry. The NVIDIA RTX PRO 6000 Blackwell uses a 5 nm process at TSMC.

Specification Differences

The two products differ in nearly every measurable specification. Process node: Intel is 3 nm; NVIDIA is 5 nm. Foundry: Intel versus TSMC. Transistor count: unknown versus 92,200 million. Die size: unknown versus 750 mm². Transistor density: null versus 122.9M per mm².

Clock speeds diverge sharply. Base clock: 300 MHz versus 1590 MHz. Boost clock: 2500 MHz versus 2617 MHz. Memory clock: System Shared versus 1750 MHz with 28 Gbps effective.

Memory configuration is entirely different. Size: System Shared versus 96 GB. Type: System Shared versus GDDR7. Bus width: System Shared versus 512 bit. Bandwidth: System Dependent versus 1.79 TB/s.

Compute unit counts differ by an order of magnitude or more. Shading units: 1,536 versus 24,064. TMUs: 48 versus 752. ROPs: 24 versus 192. RT cores: 12 versus 188. Tensor cores: null versus 752.

Throughput rates follow the same pattern. Pixel rate: 60.00 GPixel/s versus 502.5 GPixel/s. Texture rate: 120.0 GTexel/s versus 1,968.0 GTexel/s. FP32: 7.680 TFLOPS versus 126.0 TFLOPS. FP16: 15.36 TFLOPS (2:1) versus 126.0 TFLOPS (1:1).

Power and physical specifications show the starkest contrast. TDP: 80 W versus 600 W. Slot width: IGP versus dual-slot. Power connectors: none versus 1x 16-pin. Suggested PSU: null versus 1000 W. Bus interface: IGP versus PCIe 5.0 x16. Display outputs: Portable Device Dependent versus 4x DisplayPort 2.1b.

Dimensions are only documented for the NVIDIA card: 304 mm length, 137 mm height, 40 mm width. The Intel part has null dimensions. Launch MSRP is only present for the NVIDIA card: 8,565 USD. The Intel part has no listed launch MSRP.

Release dates differ: March 17, 2025 for NVIDIA, January 26, 2026 for Intel. Predecessors also differ: Workstation Ada for NVIDIA, HD Graphics-WM for Intel. The benchmark and percentile fields are equally divergent, with NVIDIA holding a 16,408 average score and a 59th percentile ranking, while Intel has a 0 average score and a 50th percentile ranking.

The Verdict

The data describes two products that occupy entirely different segments. The Intel Arc Pro B390 is an integrated GPU with system-shared memory, an 80 W TDP, and no discrete board presence. The NVIDIA RTX PRO 6000 Blackwell is a 600 W, dual-slot, PCIe 5.0 x16 add-in card with 96 GB of dedicated GDDR7 memory and a documented launch MSRP of 8,565 USD.

The only benchmark evidence available belongs to the NVIDIA card. Its 3DMark Steel Nomad DX12 score of 16,408 places it at the 59th percentile, and its nearest rivals in that test are all within 0.6 percent of its score. The Intel part has no benchmark results, so the database records no basis for claiming either GPU outperforms the other.

The architecture differences are extreme. NVIDIA's 24,064 shading units, 752 TMUs, 192 ROPs, and 188 RT cores dwarf Intel's 1,536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores. NVIDIA's 126.0 TFLOPS FP32 is roughly 16 times the Intel part's 7.680 TFLOPS. NVIDIA's 1.79 TB/s memory bandwidth stands against Intel's system-dependent shared memory. NVIDIA's 502.5 GPixel/s pixel rate and 1,968.0 GTexel/s texture rate exceed Intel's 60.00 GPixel/s and 120.0 GTexel/s by a wide margin.

The specifications make the intended use cases obvious. The Intel Arc Pro B390 targets portable or integrated systems where power draw is minimal and memory is shared with the host. The NVIDIA RTX PRO 6000 Blackwell targets professional workstations with its 1000 W suggested power supply, 16-pin connector, and 96 GB frame buffer. The NVIDIA part's release date of March 2025 and its lineage from Workstation Ada reinforce its workstation positioning, while Intel's January 2026 release and HD Graphics-WM predecessor point to a continuation of integrated graphics.

Users choosing between these two should decide based on the form factor and workload requirements visible in the data. A system that cannot accommodate a dual-slot, 304 mm, 600 W card with a 16-pin connector has no option but the Intel integrated part. A workload that needs 96 GB of dedicated GDDR7 memory, 188 RT cores, or 752 tensor cores requires the NVIDIA card. The benchmark gap, as far as the database records it, is one-sided: only NVIDIA has a score, and that score is competitive with its nearest rivals. The Intel part's absence of recorded benchmarks means no performance claim can be made for it from this data.