Intel Arc Pro B390 vs NVIDIA RTX PRO 2000 Blackwell Comparison

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark results between the Intel Arc Pro B390 and the NVIDIA RTX PRO 2000 Blackwell. Instead, the comparison relies on the recorded benchmark scores for the NVIDIA part and its position relative to other GPUs in the database.

The NVIDIA RTX PRO 2000 Blackwell delivers a 3DMark Steel Nomad DX12 score of 2374.5 points. Its Geekbench OpenCL result reaches 106087 points, while the Vulkan score sits at 113865 points. PassMark results show a G3D score of 20049, a GPU compute score of 8396, and legacy DirectX tests ranging from 80 points in DirectX 12 to 241 points in DirectX 9. The G2D score is 1303 points.

The Intel Arc Pro B390 has no recorded benchmarks in the database, and its average benchmark score is listed as zero. Its percentile ranking against all GPUs is 50, while the RTX PRO 2000 Blackwell sits at the 70th percentile. The NVIDIA card's average benchmark score of 25269 points places it in a competitive range: it trails the AMD Radeon RX 6700M by 1.4 percent, the AMD Radeon Pro W5700 by 1.8 percent, and the NVIDIA GeForce RTX 3080 Ti Mobile by 1.8 percent. Against the NVIDIA RTX A5000 Mobile, it leads by 2 percent.

The Intel part's 50th percentile ranking indicates it lands near the middle of the database's GPU population, but without any measured scores, there is no benchmark data to quantify its performance against the RTX PRO 2000 Blackwell. The NVIDIA card's 70th percentile ranking places it clearly above the median GPU.

FAQ

Q: How does the RTX PRO 2000 Blackwell compare to its nearest rivals in the database?

A: Its average benchmark score of 25269 points is 1.4 percent behind the AMD Radeon RX 6700M, 1.8 percent behind the AMD Radeon Pro W5700, and 1.8 percent behind the NVIDIA GeForce RTX 3080 Ti Mobile. It runs 2 percent ahead of the NVIDIA RTX A5000 Mobile.

Q: Does the Intel Arc Pro B390 have any recorded benchmark results?

A: No. The database lists no benchmarks for the Intel part, and its average benchmark score is zero. The only performance indicator is its 50th percentile ranking against all GPUs.

Q: What is the memory configuration of each card?

A: The RTX PRO 2000 Blackwell uses 16 GB of GDDR7 memory on a 128-bit bus with 288.0 GB/s of bandwidth. The Arc Pro B390 uses system shared memory, with its bandwidth listed as system dependent.

Q: Which card has higher raw compute throughput?

A: The RTX PRO 2000 Blackwell delivers 17.03 TFLOPS of FP32 performance and 17.03 TFLOPS of FP16 performance at a 1:1 ratio. The Arc Pro B390 delivers 7.680 TFLOPS of FP32 and 15.36 TFLOPS of FP16 at a 2:1 ratio.

Q: What are the power requirements for each card?

A: The RTX PRO 2000 Blackwell has a 70 W TDP and a suggested PSU rating of 250 W. The Arc Pro B390 has an 80 W TDP and no suggested PSU listed.

Q: How do the physical specifications differ?

A: The RTX PRO 2000 Blackwell is a dual-slot card measuring 167 mm in length, 69 mm in height, and 20 mm in width. The Arc Pro B390 is an integrated graphics processor with no physical dimensions listed.

Where Each One Wins

The RTX PRO 2000 Blackwell wins on every measurable performance axis in the database. Its 17.03 TFLOPS FP32 throughput is more than double the Arc Pro B390's 7.680 TFLOPS. In FP16 compute, the NVIDIA card's 17.03 TFLOPS at a 1:1 ratio edges out the Intel part's 15.36 TFLOPS at a 2:1 ratio. Pixel fill rate favors NVIDIA at 93.94 GPixel/s versus 60.00 GPixel/s, and texture fill rate favors NVIDIA at 266.2 GTexel/s versus 120.0 GTexel/s.

The NVIDIA card also brings dedicated resources that the Intel part lacks or implements differently. It has 34 ray tracing cores and 136 tensor cores, while the Arc Pro B390 lists 12 ray tracing cores and no tensor core count. The RTX PRO 2000 Blackwell's 4352 shading units, 136 texture mapping units, and 48 render output units all exceed the Intel part's 1536 shading units, 48 TMUs, and 24 ROPs.

Memory bandwidth is another clear differentiator. The NVIDIA card delivers 288.0 GB/s from its dedicated 16 GB GDDR7 frame buffer. The Intel part relies on system shared memory with bandwidth that varies depending on the host platform. For workloads that depend on consistent memory performance, the dedicated frame buffer provides a structural advantage.

The Arc Pro B390 does hold advantages in a few areas. Its base clock of 300 MHz and boost clock of 2500 MHz are not directly comparable to the NVIDIA card's 982 MHz base and 1957 MHz boost, since clock behavior depends on architecture and thermal design. The Intel part's 80 W TDP is 10 W higher than the NVIDIA card's 70 W TDP, but the Intel part is an integrated GPU with no power connectors, which simplifies installation in systems with an IGP-compatible socket. The NVIDIA card also runs without power connectors, but its dual-slot footprint and 250 W suggested PSU indicate it draws power through the PCIe slot and expects a more substantial system power supply.

Specification Differences

The RTX PRO 2000 Blackwell uses 16 GB of GDDR7 memory on a 128-bit bus, providing 288.0 GB/s of bandwidth. The Arc Pro B390 uses system shared memory, with no dedicated VRAM, no fixed bus width, and bandwidth rated as system dependent. The NVIDIA card's memory clock is 1125 MHz with 18 Gbps effective data rate; the Intel part's memory clock is listed as system shared.

The NVIDIA card has 4352 shading units, 136 TMUs, and 48 ROPs. The Intel part has 1536 shading units, 48 TMUs, and 24 ROPs. Ray tracing core counts are 34 for NVIDIA and 12 for Intel. Tensor cores are present only on the NVIDIA card, at 136. The Intel part lists no tensor core count.

The RTX PRO 2000 Blackwell's process node is 5 nm at TSMC, with 21,900 million transistors on a 181 mm² die and a transistor density of 121.0M per mm². The Arc Pro B390 uses a 3 nm node at Intel, with transistor count and die size listed as unknown. The NVIDIA card uses a PCIe 5.0 x8 bus interface; the Intel part uses an IGP bus interface. The NVIDIA card is a dual-slot add-in card; the Intel part is an integrated GPU.

Display outputs differ significantly. The RTX PRO 2000 Blackwell provides 4x mini-DisplayPort 2.1b connectors. The Arc Pro B390's display outputs are listed as portable device dependent, meaning they vary by the host system.

Architecture Differences

The Intel Arc Pro B390 is built on the Panther Lake chip using the Xe3-LPG architecture, part of the Arc Graphics-WM generation. It uses a 3 nm process from Intel's own foundry. The architecture supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

The NVIDIA RTX PRO 2000 Blackwell uses the GB206 chip with Blackwell 2.0 architecture, part of the Blackwell PRO W generation. It is fabricated on a 5 nm process at TSMC. Like the Intel part, it supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

The Xe3-LPG architecture in the Intel part is an integrated GPU design with 12 ray tracing cores and no tensor cores listed. Its FP16 throughput of 15.36 TFLOPS at a 2:1 ratio suggests a shader-based approach to reduced-precision compute. The Blackwell 2.0 architecture in the NVIDIA part includes 34 ray tracing cores and 136 tensor cores, with FP16 performance of 17.03 TFLOPS at a 1:1 ratio, indicating dedicated tensor hardware for AI workloads.

The NVIDIA card's predecessor is listed as Workstation Ada, while the Intel part's predecessor is HD Graphics-WM. The NVIDIA card was released on 2025-08-10, and the Intel part was released on 2026-01-26. The NVIDIA card's die size of 181 mm² with 21,900 million transistors gives it a density of 121.0M transistors per mm². The Intel part's transistor count and die size are unknown, so its density cannot be calculated. The Intel part's 3 nm node is a smaller process geometry than the NVIDIA part's 5 nm node, though the lack of transistor data for the Intel part prevents direct density comparisons.

The Verdict

The recorded data points to the NVIDIA RTX PRO 2000 Blackwell as the stronger performer. It carries a 70th percentile ranking against all GPUs, an average benchmark score of 25269, and a full suite of measured results across DirectX, OpenCL, Vulkan, and PassMark tests. The Intel Arc Pro B390 has no measured benchmarks in the database, only a 50th percentile ranking and a zero average score.

Workloads that demand raw throughput favor the NVIDIA card. Its FP32 compute of 17.03 TFLOPS more than doubles the Intel part's 7.680 TFLOPS. Texture and pixel fill rates are significantly higher on the NVIDIA side. The 288.0 GB/s of dedicated GDDR7 bandwidth provides predictable memory performance that the Intel part's system dependent shared memory cannot guarantee. The presence of 136 tensor cores also gives the NVIDIA card a clear path for AI and compute workloads that can use tensor operations.

The Intel Arc Pro B390 is an integrated GPU with no power connectors, no add-in card footprint, and display output dependent on the host device. It fits systems where a discrete card is not an option. Its 80 W TDP is modest, though it is 10 W higher than the NVIDIA card's 70 W TDP. The Intel part's 3 nm process node is smaller than the NVIDIA card's 5 nm node, but without transistor or die size data, the practical impact of that process advantage cannot be assessed from the database.

For users selecting a GPU based on measured performance, the RTX PRO 2000 Blackwell is the only one of the two with benchmark data, and that data places it in the upper tier of the database. The Arc Pro B390's lack of recorded scores means any performance claims about it would be speculative. The NVIDIA card also offers a dedicated 16 GB frame buffer, a dual-slot form factor with four mini-DisplayPort 2.1b outputs, and a PCIe 5.0 x8 interface, all of which are concrete specifications that support professional workstation use. The Intel part's integrated design and system dependent memory make it suitable primarily for compact or low-power systems where its IGP form factor is a requirement rather than a choice.