Intel Arc Pro B60 vs NVIDIA N1X 48SM Comparison

Intel Arc Pro B60 vs NVIDIA N1X 48SM

The database records two active graphics processors with fundamentally different design goals. The Intel Arc Pro B60 is a discrete add-in board built for professional workstations, while the NVIDIA N1X 48SM is an integrated graphics processor (IGP) designed for a system-on-chip platform. Benchmark coverage is uneven: the Intel part has eight recorded test scores, the NVIDIA part has none, so the comparison relies on architectural data and the Intel card's position among its nearest rivals. The Intel Arc Pro B60 sits at the 20th percentile among all GPUs in the database, with an average benchmark score of 3182. Its nearest rivals, all NVIDIA products, cluster closely: the Quadro P1000 scores 3163 (0.6% behind), the GeForce GT 640 scores 3210 (0.9% ahead), the GeForce 920M scores 3287 (3.2% ahead), and the GeForce RTX 5080 SUPER scores 3075 (3.5% behind). This narrow spread indicates that the Arc Pro B60 lands in a crowded mid-range performance band, where small percentage swings separate adjacent products.

Head-to-Head Benchmarks

The head-to-head benchmark table between the Intel Arc Pro B60 and the NVIDIA N1X 48SM is empty in the database. No direct tests were run pairing these two parts, and the NVIDIA N1X 48SM has no individual benchmark scores recorded at all. Consequently, there are zero wins recorded for either side in direct competition. The Intel Arc Pro B60, however, has its own benchmark suite that provides context for its capabilities. In 3DMark Steel Nomad DX12, it scores 2646. PassMark results show a DirectX 10 score of 61, DirectX 11 score of 122, DirectX 12 score of 76, and DirectX 9 score of 179. The 2D graphics score is 763, while the 3D graphics score reaches 14580. GPU compute, measured separately, returns 7029. These numbers place the Arc Pro B60 at 20th percentile overall, with an average score of 3182 across its eight tests.

Because the NVIDIA part lacks recorded scores, the data cannot show a direct percentage lead for either side. What the data does show is that the Intel card's nearest rival, the Quadro P1000, is essentially tied with it at 0.6% difference in average score. The GeForce GT 640 and GeForce 920M sit slightly ahead by 0.9% and 3.2% respectively, while the RTX 5080 SUPER sits 3.5% behind. This indicates that the Arc Pro B60's performance profile is not dramatically separated from its neighbors; it occupies a specific point in the distribution where rivals are within a few percentage points. The absence of data for the NVIDIA N1X 48SM means no head-to-head percentages can be calculated, and the comparison must rely on specifications and architectural analysis rather than measured performance deltas.

Architecture Differences

The Intel Arc Pro B60 uses the BMG-G21 chip based on the Xe2-HPG architecture, belonging to the Battlemage Pro Series generation. It is fabricated on a 5 nm process at TSMC, with 19,600 million transistors on a die size of 272 mm², yielding a transistor density of 72.1 million per square millimeter. The NVIDIA N1X 48SM uses the GB20B chip based on Blackwell 2.0 architecture, part of the Blackwell IGP N1x generation. It also uses a 5 nm TSMC process, but its die size is 382 mm², larger than the Intel chip. The transistor count for the NVIDIA part is listed as unknown, and transistor density is not recorded.

Clock speeds differ substantially. The Intel card runs at a base clock of 2000 MHz and boosts to 2400 MHz, with memory clocked at 2375 MHz (19 Gbps effective). The NVIDIA IGP has a much lower base clock of 741 MHz but boosts to 2346 MHz, nearly matching the Intel boost. Its memory clock is 1067 MHz (8.5 Gbps effective). Shading unit counts diverge sharply: the Intel part has 2560 shading units, while the NVIDIA part has 6144, more than double. Texture mapping units are 160 for Intel versus 384 for NVIDIA. Raster output units are 80 for Intel versus 48 for NVIDIA, meaning Intel has higher pixel throughput potential per clock. Ray tracing cores number 20 on Intel and 48 on NVIDIA. Tensor cores are not listed for Intel, while NVIDIA has 192.

Memory configurations are also different. The Intel Arc Pro B60 carries 24 GB of GDDR6 on a 192 bit bus, delivering 456.0 GB/s bandwidth. The NVIDIA N1X 48SM carries 128 GB of LPDDR5X on a 256 bit bus, but bandwidth is only 273.2 GB/s, lower than the Intel card despite the wider bus, due to the slower memory type and clock. Pixel rate for Intel is 192.0 GPixel/s, while NVIDIA is 112.6 GPixel/s. Texture rate is 384.0 GTexel/s for Intel and 900.9 GTexel/s for NVIDIA. Floating point performance shows Intel at 12.29 TFLOPS FP32 and 24.58 TFLOPS FP16 (2:1 ratio), while NVIDIA reaches 28.83 TFLOPS FP32 and 28.83 TFLOPS FP16 (1:1 ratio). The NVIDIA part does not sacrifice FP16 throughput, indicating a different compute balance.

Power and physical design diverge completely. The Intel card has a TDP of 200 W, is dual-slot, uses one 8-pin power connector, and suggests a 550 W power supply. It connects via PCIe 5.0 x8. The NVIDIA IGP has an unknown TDP, is an integrated graphics processor with no power connectors, and connects via PCIe 5.0 x16. Display outputs are four mini-DisplayPort 2.1 on Intel versus one HDMI on NVIDIA. API support also differs: Intel lists DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while NVIDIA lists N/A for DirectX, OpenGL, and Vulkan, indicating the IGP's API exposure is not recorded in the database. Release dates are separated by months: Intel launched on September 4, 2025, with a launch MSRP of 499 USD, while NVIDIA's release date is May 31, 2026, with no MSRP recorded.

Where Each One Wins

Based on the recorded data, the Intel Arc Pro B60 wins in memory bandwidth, delivering 456.0 GB/s versus 273.2 GB/s for the NVIDIA IGP. This is a 1.67x advantage in raw bandwidth, which benefits texture-heavy workloads and large data transfers. Intel also wins in pixel rate, 192.0 GPixel/s versus 112.6 GPixel/s, a significant margin for fill-rate-bound tasks. The Intel card has a higher boost clock at 2400 MHz versus 2346 MHz, though the base clock difference is large in Intel's favor at 2000 MHz versus 741 MHz. The Intel card's 24 GB GDDR6 memory, while smaller in capacity, uses a faster memory type with higher effective speed, and its 80 ROPs exceed the NVIDIA part's 48 ROPs.

The NVIDIA N1X 48SM wins in shading unit count, with 6144 versus 2560, a 2.4x advantage. This translates to higher FP32 compute: 28.83 TFLOPS versus 12.29 TFLOPS, a 2.35x lead. Texture rate also favors NVIDIA at 900.9 GTexel/s versus 384.0 GTexel/s, a 2.35x advantage. The NVIDIA part has 384 TMUs versus 160, and 48 ray tracing cores versus 20. Tensor cores, present only on NVIDIA at 192, give it a dedicated AI compute path that Intel lacks entirely. Memory capacity is also a clear NVIDIA win: 128 GB versus 24 GB, a 5.33x difference, which matters for large datasets or in-memory workloads. The NVIDIA part's 256 bit bus is wider than Intel's 192 bit, though bandwidth is lower. The PCIe 5.0 x16 interface on NVIDIA provides double the lane width of Intel's x8, potentially improving host communication for compute tasks.

The data suggests a use-case split. For rasterization-heavy graphics with high pixel fill rates and memory bandwidth, the Intel Arc Pro B60 shows advantages in the recorded specifications. For compute-heavy workloads that leverage massive shading unit counts, FP32 throughput, and tensor cores, the NVIDIA N1X 48SM is clearly positioned ahead. The Intel card's 20th percentile ranking among all GPUs, with an average score of 3182, indicates it is not a top-tier performer globally. The NVIDIA IGP's percentile is listed at 50, but with no benchmark scores, this percentile cannot be validated against recorded tests.

The Verdict

The data indicates that these two processors serve different roles, and the choice depends on the workload. For a discrete graphics card used in professional workstations, the Intel Arc Pro B60 delivers 24 GB of GDDR6 memory with 456.0 GB/s bandwidth, 12.29 TFLOPS FP32, and 20 ray tracing cores. It has a launch MSRP of 499 USD, a 200 W TDP, and requires a 550 W power supply. Its nearest rivals, all NVIDIA discrete cards, are within 3.5% in average score, so it performs in a specific mid-range band. The 20th percentile ranking places it below the median GPU in the database.

For the NVIDIA N1X 48SM, the absence of benchmark scores means performance cannot be quantified from measurements. The specifications show a 128 GB LPDDR5X memory pool, 28.83 TFLOPS FP32, 192 tensor cores, and 48 ray tracing cores. Its base clock is low at 741 MHz, but the boost reaches 2346 MHz. As an IGP with no power connectors and one HDMI output, it is not designed for standalone graphics tasks but for integration into a larger system. The 50th percentile ranking, without supporting benchmark data, should be treated as provisional.

The recorded data does not show a head-to-head winner, because no direct benchmarks exist. The Intel card's wins are in bandwidth, pixel rate, and boost clock. The NVIDIA part's wins are in compute throughput, memory capacity, and tensor core availability. Users who need a discrete, field-replaceable graphics card with established benchmark scores and a known performance tier would select the Intel Arc Pro B60. Users who need an integrated processor with massive memory and compute resources for AI-adjacent workloads, and who do not require discrete graphics outputs beyond one HDMI port, would select the NVIDIA N1X 48SM. The data cannot support a recommendation beyond these specification-driven conclusions.

FAQ

Q: Which GPU has higher memory bandwidth?

A: The Intel Arc Pro B60 has 456.0 GB/s bandwidth, while the NVIDIA N1X 48SM has 273.2 GB/s, so Intel leads by 182.8 GB/s.

Q: What is the FP32 performance difference?

A: The NVIDIA N1X 48SM delivers 28.83 TFLOPS FP32, while the Intel Arc Pro B60 delivers 12.29 TFLOPS, giving NVIDIA a 16.54 TFLOPS advantage.

Q: How much memory does each GPU have?

A: The Intel Arc Pro B60 has 24 GB of GDDR6, and the NVIDIA N1X 48SM has 128 GB of LPDDR5X, a 104 GB difference.

Q: What are the ray tracing core counts?

A: The Intel Arc Pro B60 has 20 ray tracing cores, and the NVIDIA N1X 48SM has 48, so NVIDIA has 28 more.

Q: Which GPU has more shading units?

A: The NVIDIA N1X 48SM has 6144 shading units, while the Intel Arc Pro B60 has 2560, a difference of 3584 units.

Q: What is the launch date for each?

A: The Intel Arc Pro B60 released on September 4, 2025, and the NVIDIA N1X 48SM has a release date of May 31, 2026.

Specification Differences

| Specification | Intel Arc Pro B60 | NVIDIA N1X 48SM |

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

| Chip | BMG-G21 | GB20B |

| Architecture | Xe2-HPG | Blackwell 2.0 |

| Generation | Battlemage (Pro Series) | Blackwell IGP (N1x) |

| Process Node | 5 nm | 5 nm |

| Die Size | 272 mm² | 382 mm² |

| Transistors | 19,600 million | unknown |

| Transistor Density | 72.1M / mm² | null |

| Base Clock | 2000 MHz | 741 MHz |

| Boost Clock | 2400 MHz | 2346 MHz |

| Memory Clock | 2375 MHz 19 Gbps effective | 1067 MHz 8.5 Gbps effective |

| Memory Size | 24 GB | 128 GB |

| Memory Type | GDDR6 | LPDDR5X |

| Memory Bus Width | 192 bit | 256 bit |

| Memory Bandwidth | 456.0 GB/s | 273.2 GB/s |

| Shading Units | 2560 | 6144 |

| TMUs | 160 | 384 |

| ROPs | 80 | 48 |

| RT Cores | 20 | 48 |

| Tensor Cores | null | 192 |

| Pixel Rate | 192.0 GPixel/s | 112.6 GPixel/s |

| Texture Rate | 384.0 GTexel/s | 900.9 GTexel/s |

| FP32 | 12.29 TFLOPS | 28.83 TFLOPS |

| FP16 | 24.58 TFLOPS (2:1) | 28.83 TFLOPS (1:1) |

| TDP | 200 W | unknown |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 8-pin | None |

| Suggested PSU | 550 W | null |

| Bus Interface | PCIe 5.0 x8 | PCIe 5.0 x16 |

| Display Outputs | 4x mini-DisplayPort 2.1 | 1x HDMI |

| DirectX | 12 Ultimate (12_2) | N/A |

| OpenGL | 4.6 | N/A |

| Vulkan | 1.4 | N/A |

| Dimensions (L x H x W) | 167 mm x 69 mm x 40 mm | null |

| Release Date | 2025-09-04 | 2026-05-31 |

| Launch MSRP | 499 USD | null |

| Production Status | Active | Active |

| Percentile vs All GPUs | 20 | 50 |

| Average Benchmark Score | 3182 | 0 |

| Benchmark Scores Recorded | 8 | 0 |