Intel Arc Pro B60 vs NVIDIA GeForce RTX 5090 SE Comparison

The Verdict

The recorded data presents a stark contrast between these two 24 GB GPUs. The Intel Arc Pro B60, with an average benchmark score of 3182, sits at the 20th percentile of all GPUs. Its nearest rivals in the database include the NVIDIA Quadro P1000 (3163, 0.6% ahead), the GeForce GT 640 (3210, 0.9% behind), and the GeForce 920M (3287, 3.2% behind). The Arc Pro B60 also trails the GeForce RTX 5080 SUPER (3075) by 3.5%. This places it firmly in entry-level or professional-visualization territory based on aggregate scores.

The NVIDIA GeForce RTX 5090 SE, by contrast, holds the 50th percentile ranking. Its average benchmark score is recorded as 0, and it has no nearest rivals listed, indicating insufficient benchmark data in the database for direct numerical comparison. However, the architectural specifications reveal a GPU of an entirely different scale: 14,080 shading units versus 2,560, 110 RT cores versus 20, and 440 tensor cores where the Intel part has none listed. The RTX 5090 SE delivers 66.94 TFLOPS FP32 performance against 12.29 TFLOPS for the Arc Pro B60.

From the data available, the RTX 5090 SE is the compute and rendering powerhouse, while the Arc Pro B60 is a compact, low-power professional card. The RTX 5090 SE should be selected for tasks requiring maximum throughput, such as high-resolution rendering, AI inference, or large-scale compute workloads. The Arc Pro B60, with its 200 W TDP and 167 mm length, is suited for space-constrained or power-sensitive professional systems where the 24 GB frame buffer is the primary requirement. The data does not support any scenario where the RTX 5090 SE is not the faster card; it only shows that the Arc Pro B60 is significantly smaller and more power-efficient.

Where Each One Wins

The Arc Pro B60 wins in physical footprint and power draw. Its dimensions are 167 mm (6.6 inches) in length, 69 mm (2.7 inches) in height, and 40 mm (1.6 inches) in width, versus 267 mm (10.5 inches), 111 mm (4.4 inches), and 40 mm (1.6 inches) for the RTX 5090 SE. The Intel card requires a single 8-pin power connector and a 550 W suggested PSU, whereas the NVIDIA card uses one 16-pin connector and recommends a 900 W PSU. The TDP difference is 200 W for the Arc Pro B60 versus 500 W for the RTX 5090 SE.

The RTX 5090 SE wins in every performance metric recorded. Its pixel rate is 380.3 GPixel/s versus 192.0 GPixel/s. Its texture rate is 1,045.9 GTexel/s versus 384.0 GTexel/s. Its FP32 throughput is 66.94 TFLOPS versus 12.29 TFLOPS. Memory bandwidth is 1.34 TB/s versus 456.0 GB/s. The RTX 5090 SE also has a wider 384-bit memory bus compared to 192-bit, and uses GDDR7 memory versus GDDR6. Both cards have 24 GB of memory, but the NVIDIA part moves data nearly three times faster.

The Arc Pro B60 uses a PCIe 5.0 x8 interface, while the RTX 5090 SE uses PCIe 5.0 x16. Display outputs differ: the Intel card provides four mini-DisplayPort 2.1 connectors, while the NVIDIA card provides one HDMI 2.1b and three DisplayPort 2.1b outputs. The Intel card's boost clock is 2400 MHz, slightly higher than the RTX 5090 SE's 2377 MHz boost, but this does not compensate for the massive difference in shading units.

Architecture Differences

The Intel Arc Pro B60 is built on the Xe2-HPG architecture, using the BMG-G21 chip, and belongs to the Battlemage (Pro Series) generation. The NVIDIA GeForce RTX 5090 SE uses the Blackwell 2.0 architecture with the GB202 chip, part of the GeForce 50 series. Both are fabricated on a 5 nm process at TSMC, but the transistor counts differ enormously: 19,600 million for Intel versus 92,200 million for NVIDIA. The die sizes are 272 mm² and 750 mm² respectively, giving transistor densities of 72.1 million per mm² for the Intel chip and 122.9 million per mm² for the NVIDIA chip.

The RTX 5090 SE has 14,080 shading units, 440 TMUs, and 160 ROPs. The Arc Pro B60 has 2,560 shading units, 160 TMUs, and 80 ROPs. RT core counts are 110 versus 20. The NVIDIA card includes 440 tensor cores; the Intel card does not list any. FP16 performance on the RTX 5090 SE is 66.94 TFLOPS at a 1:1 ratio with FP32, while the Arc Pro B60 achieves 24.58 TFLOPS at a 2:1 ratio, indicating the NVIDIA card has full-rate FP16 throughput.

Memory subsystems diverge significantly. The RTX 5090 SE uses 24 GB of GDDR7 on a 384-bit bus at 28 Gbps effective, yielding 1.34 TB/s bandwidth. The Arc Pro B60 uses 24 GB of GDDR6 on a 192-bit bus at 19 Gbps effective, yielding 456.0 GB/s. Base clocks are 1740 MHz for NVIDIA versus 2000 MHz for Intel, with boost clocks of 2377 MHz and 2400 MHz respectively. The Intel card has a higher base clock, but the NVIDIA card has a much larger execution resource pool.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 5090 SE is a dual-slot card with a 1x 16-pin power connector, while the Arc Pro B60 is also dual-slot but uses a 1x 8-pin connector. The Intel card was released on September 4, 2025, and the NVIDIA card on December 31, 2025. The RTX 5090 SE lists a predecessor (GeForce 40) and successor (GeForce 60), while the Arc Pro B60 lists neither.

FAQ

Q: Which GPU has more raw compute power according to the data?

A: The NVIDIA GeForce RTX 5090 SE. Its FP32 throughput is 66.94 TFLOPS versus 12.29 TFLOPS for the Intel Arc Pro B60. The NVIDIA card also has 14,080 shading units compared to 2,560.

Q: Do both cards have the same amount of memory?

A: Yes, both have 24 GB. However, the RTX 5090 SE uses GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth, while the Arc Pro B60 uses GDDR6 on a 192-bit bus with 456.0 GB/s bandwidth.

Q: What is the power consumption difference?

A: The Intel Arc Pro B60 has a TDP of 200 W with a suggested 550 W PSU and a single 8-pin connector. The NVIDIA GeForce RTX 5090 SE has a TDP of 500 W, a suggested 900 W PSU, and a single 16-pin connector.

Q: Which card is physically smaller?

A: The Intel Arc Pro B60. It measures 167 mm in length and 69 mm in height, versus 267 mm and 111 mm for the RTX 5090 SE. Both are 40 mm wide and dual-slot.

Q: How do the benchmark scores compare?

A: The Arc Pro B60 has an average benchmark score of 3182 and sits at the 20th percentile. The RTX 5090 SE has an average benchmark score of 0 and sits at the 50th percentile, but no nearest rivals are listed for it, so direct score comparison is not possible from the recorded data.

Q: What are the display output options?

A: The Intel card provides four mini-DisplayPort 2.1 outputs. The NVIDIA card provides one HDMI 2.1b and three DisplayPort 2.1b outputs.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries between the Intel Arc Pro B60 and the NVIDIA GeForce RTX 5090 SE. The wins counters for both cards are 0. Therefore, the comparison must rely on the individual benchmark results for the Arc Pro B60 and the architectural specifications for both cards.

The Arc Pro B60's recorded benchmarks show a 3DMark Steel Nomad DX12 score of 2646. Its PassMark scores are 61 for DirectX 10, 122 for DirectX 11, 76 for DirectX 12, 179 for DirectX 9, 763 for G2D, 14580 for G3D, and 7029 for GPU compute. These figures establish a baseline for its performance class. The average benchmark score of 3182 places it near the Quadro P1000 (3163, 0.6% difference) and the GeForce GT 640 (3210, 0.9% difference).

For the RTX 5090 SE, no benchmark scores are recorded. The only quantitative performance indicators are its theoretical rates: 380.3 GPixel/s pixel fill rate, 1,045.9 GTexel/s texture fill rate, and 66.94 TFLOPS FP32. These are approximately 2x, 2.7x, and 5.4x the corresponding figures for the Arc Pro B60 (192.0 GPixel/s, 384.0 GTexel/s, 12.29 TFLOPS).

The memory bandwidth gap is similarly wide. The RTX 5090 SE's 1.34 TB/s is roughly 2.9x the Arc Pro B60's 456.0 GB/s. The RT core count difference (110 versus 20) suggests a 5.5x advantage in ray tracing hardware resources, and the tensor core presence (440 on NVIDIA) versus none on Intel indicates a fundamental feature gap for AI workloads.

The percentile rankings are informative. The Arc Pro B60 at the 20th percentile means 80% of GPUs in the database outperform it. The RTX 5090 SE at the 50th percentile sits at the median, but this ranking is based on a database entry with zero benchmark scores, so it likely reflects the specifications rather than measured performance. The nearest rivals for the Arc Pro B60 all have average scores within 3.5% of its own, confirming its position in the lower-middle tier of GPU performance.

The clock speeds tell a partial story. The Arc Pro B60 runs at 2000 MHz base and 2400 MHz boost, while the RTX 5090 SE runs at 1740 MHz base and 2377 MHz boost. The Intel card has higher clocks on paper, but with 5.5x fewer shading units, its aggregate throughput is far lower. The RTX 5090 SE compensates for a lower boost clock with sheer execution resource count.

The bus interface also differs. The RTX 5090 SE uses PCIe 5.0 x16, providing double the lane count of the Arc Pro B60's PCIe 5.0 x8. For bandwidth-sensitive workloads that exceed the local memory capacity, this could amplify the NVIDIA card's advantage further, though both cards have 24 GB of local memory.

In summary, the recorded data shows no scenario where the Intel Arc Pro B60 outperforms the NVIDIA GeForce RTX 5090 SE. The Intel card wins only on power draw, physical size, and clock frequency. The NVIDIA card dominates in every computational throughput metric, memory bandwidth, and feature set, including tensor cores and a larger RT core array. The lack of head-to-head benchmark scores means the exact performance delta cannot be quantified, but the architectural disparity is decisive.