Intel Arc Pro B390 vs NVIDIA RTX 6000D Comparison

The Intel Arc Pro B390 and NVIDIA RTX 6000D occupy opposite ends of the GPU spectrum. The recorded data shows a stark contrast in performance class, with the RTX 6000D delivering roughly 12.6 times the average benchmark score of the Arc Pro B390. This analysis relies solely on the specifications and benchmark results in the database.

Head-to-Head Benchmarks

The benchmark data available for these two products is asymmetric. The NVIDIA RTX 6000D has two recorded benchmark scores, while the Intel Arc Pro B390 has none in the database. This absence of direct comparative data means any head-to-head analysis must infer relative performance from the RTX 6000D's results and the Arc Pro B390's raw specification limits.

The RTX 6000D achieves an average benchmark score of 195,964 across its two tests. Its performance percentile sits at 98, meaning it outperforms 98% of all GPUs tracked in the database. In the 3DMark Steel Nomad DX12 test, it scores 3,522. In the Geekbench OpenCL test, it scores 388,405. These results place it in the top tier of professional workstation graphics cards.

The database provides nearest rival comparisons for the RTX 6000D, which helps contextualize its performance. It leads the NVIDIA Tesla V100S PCIe 32 GB by 0.8%, with that rival scoring 194,415. It leads the NVIDIA A100 SXM4 40 GB by 4.7%, with that rival scoring 187,147. It trails the NVIDIA A100 PCIe 80 GB by 5.4%, with that rival scoring 207,124. It leads the NVIDIA RTX 5000 Ada Generation by 6.1%, with that rival scoring 184,664. These deltas show the RTX 6000D sits in a competitive band among high-end accelerators, slightly ahead of some data center workhorses and slightly behind one.

The Arc Pro B390 has no recorded benchmark scores and no nearest rivals in the database. Its percentile versus all GPUs is 50, which in this data context indicates a mid-pack position, but without actual scores, this figure represents a projected or derived placement rather than a measured result. The Arc Pro B390's FP32 compute of 7.680 TFLOPS is only 7.9% of the RTX 6000D's 97.04 TFLOPS, and its pixel rate of 60.00 GPixel/s is only 12.9% of the RTX 6000D's 466.6 GPixel/s. These specification gaps strongly suggest the Arc Pro B390 would trail in every compute-heavy benchmark.

Architecture Differences

The architectural divide between these two GPUs is fundamental. The Intel Arc Pro B390 uses the Xe3-LPG architecture built on the Panther Lake chip, fabricated on a 3 nm process at Intel. The NVIDIA RTX 6000D uses the Blackwell 2.0 architecture built on the GB202 chip, fabricated on a 5 nm process at TSMC.

The RTX 6000D integrates 92,200 million transistors on a 750 mm² die, yielding a transistor density of 122.9M per mm². The Arc Pro B390's transistor count and die size are both listed as unknown in the database, so no direct density comparison is possible. The process node difference is notable: the Arc Pro B390 uses a more advanced 3 nm node versus the RTX 6000D's 5 nm node, but the RTX 6000D compensates with a vastly larger die and far more functional units.

The RTX 6000D is a discrete dual-slot card with a PCIe 5.0 x16 bus interface. The Arc Pro B390 is an integrated graphics processor (IGP) with no slot width and no power connectors. The Arc Pro B390 uses system shared memory, while the RTX 6000D has 84 GB of dedicated GDDR7 memory on a 448-bit bus with 1.40 TB/s bandwidth. The Arc Pro B390's memory bandwidth is listed as system dependent, which means it shares system memory bandwidth with the CPU.

The RTX 6000D carries 19,968 shading units, 624 texture mapping units, 192 raster output units, 156 ray tracing cores, and 624 tensor cores. The Arc Pro B390 has 1,536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores, with no tensor core count listed. This means the RTX 6000D has 13 times the shading units, 13 times the TMUs, 8 times the ROPs, and 13 times the ray tracing cores. The RTX 6000D also supports FP16 at a 1:1 ratio with FP32, both at 97.04 TFLOPS, while the Arc Pro B390's FP16 is 15.36 TFLOPS at a 2:1 ratio.

Where Each One Wins

The use-case split between these two products is clear from the data. The RTX 6000D wins in every performance-oriented workload category. Its 97.04 TFLOPS of FP32 compute targets professional simulation, AI training, and high-end rendering. Its 1.40 TB/s memory bandwidth and 84 GB capacity support massive datasets that exceed the system shared memory available to the Arc Pro B390. Its 156 ray tracing cores and 624 tensor cores provide dedicated acceleration for ray-traced visualization and neural network inference.

The RTX 6000D's pixel rate of 466.6 GPixel/s and texture rate of 1,516.3 GTexel/s position it for high-resolution, multi-display workstation use. Its four DisplayPort 2.1b outputs support professional multi-monitor setups. Its 600 W TDP and 1000 W suggested PSU requirement indicate a system designed for sustained heavy compute loads.

The Arc Pro B390 wins in power efficiency and system integration. Its 80 W TDP is just 13.3% of the RTX 6000D's 600 W TDP. As an integrated GPU with no power connectors and no slot width, it requires no additional cooling or power delivery infrastructure. Its 3 nm process node gives it a manufacturing advantage in transistor efficiency per watt. The Arc Pro B390 is suited for thin-and-light portable devices where the display outputs are portable device dependent, meaning it serves integrated graphics duties in laptops or compact systems.

The Arc Pro B390 also supports the same API feature set as the RTX 6000D in terms of DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. This means software compatibility for graphics APIs is equivalent, though performance will differ dramatically. The Arc Pro B390's 60.00 GPixel/s pixel rate is sufficient for standard desktop and light creative workloads, but not for heavy 3D rendering or video processing at scale.

Specification Differences

The following specifications differ between the two products, based solely on the recorded data:

| Specification | Intel Arc Pro B390 | NVIDIA RTX 6000D |

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

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Chip | Panther Lake | GB202 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Unknown | 92,200 million |

| Die Size | Unknown | 750 mm² |

| Base Clock | 300 MHz | 1992 MHz |

| Boost Clock | 2500 MHz | 2430 MHz |

| Memory Size | System Shared | 84 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 448 bit |

| Memory Bandwidth | System Dependent | 1.40 TB/s |

| Shading Units | 1536 | 19968 |

| TMUs | 48 | 624 |

| ROPs | 24 | 192 |

| Ray Tracing Cores | 12 | 156 |

| Tensor Cores | Not listed | 624 |

| Pixel Rate | 60.00 GPixel/s | 466.6 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 1,516.3 GTexel/s |

| FP32 Performance | 7.680 TFLOPS | 97.04 TFLOPS |

| FP16 Performance | 15.36 TFLOPS (2:1) | 97.04 TFLOPS (1:1) |

| TDP | 80 W | 600 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 16-pin |

| Suggested PSU | Not listed | 1000 W |

| Bus Interface | IGP | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 4x DisplayPort 2.1b |

| Release Date | 2026-01-26 | 2025-07-13 |

| Predecessor | HD Graphics-WM | Workstation Ada |

The launch MSRP for the RTX 6000D is 8,565 USD. The Arc Pro B390 has no launch MSRP recorded. The RTX 6000D's dimensions are 304 mm in length, 137 mm in height, and 40 mm in width, while the Arc Pro B390 has no dimensions recorded due to its integrated nature.

FAQ

Q: Which GPU has more shading units?

A: The NVIDIA RTX 6000D has 19,968 shading units, which is 13 times the 1,536 shading units of the Intel Arc Pro B390.

Q: What is the memory capacity difference?

A: The RTX 6000D has 84 GB of dedicated GDDR7 memory on a 448-bit bus, while the Arc Pro B390 uses system shared memory with bandwidth that is system dependent.

Q: How do the power requirements compare?

A: The RTX 6000D has a 600 W TDP and requires a 1000 W suggested PSU with a 1x 16-pin power connector. The Arc Pro B390 has an 80 W TDP, no power connectors, and is an integrated GPU (IGP).

Q: Which GPU has a higher boost clock?

A: The Intel Arc Pro B390 has a boost clock of 2500 MHz, which is 70 MHz higher than the RTX 6000D's boost clock of 2430 MHz. However, the RTX 6000D has a much higher base clock of 1992 MHz versus 300 MHz for the Arc Pro B390.

Q: What API support do both GPUs share?

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

Q: What is the performance percentile of each GPU?

A: The RTX 6000D sits at the 98th percentile versus all GPUs in the database, while the Arc Pro B390 sits at the 50th percentile.

The Verdict

The data directs a clear verdict. The NVIDIA RTX 6000D is a high-end professional workstation GPU designed for maximum compute throughput. Its 97.04 TFLOPS FP32 performance, 84 GB GDDR7 memory, and 98th percentile ranking confirm it as a top-tier accelerator. The Arc Pro B390 is an integrated graphics solution with an 80 W TDP, designed for power-constrained portable systems. Its 7.680 TFLOPS FP32 performance and system shared memory place it in a fundamentally different performance class.

Users requiring heavy 3D rendering, large-scale simulation, AI inference, or multi-display professional visualization should select the RTX 6000D. Its 156 ray tracing cores, 624 tensor cores, and 1.40 TB/s memory bandwidth deliver the necessary resources for these workloads. The RTX 6000D leads its nearest rivals by 0.8% over the Tesla V100S, 4.7% over the A100 SXM4, and 6.1% over the RTX 5000 Ada Generation, though it trails the A100 PCIe 80 GB by 5.4%.

Users building compact or battery-powered systems with no discrete graphics slot should select the Arc Pro B390. Its 3 nm process node and 80 W TDP make it a capable integrated option for standard productivity and light creative tasks. The Arc Pro B390's 2500 MHz boost clock and 12 ray tracing cores provide modern API support without dedicated power delivery. The choice hinges entirely on the target system form factor and workload intensity, as the performance gap between these two products spans more than an order of magnitude in compute throughput.