Intel Arc Pro B60 Dual vs NVIDIA RTX 2000 Mobile Ada Generation Comparison

FAQ

Q: What are the core architecture and process differences between these two GPUs?

A: The Intel Arc Pro B60 Dual uses the BMG-G21 chip with Xe2-HPG architecture, built on TSMC's 5 nm process with 19,600 million transistors on a 272 mm² die. The NVIDIA RTX 2000 Mobile Ada Generation uses the AD107 chip with Ada Lovelace architecture, also on TSMC's 5 nm process, with 18,900 million transistors on a 159 mm² die. The transistor density differs significantly, with the NVIDIA chip at 118.9M / mm² versus Intel's 72.1M / mm².

Q: How do the memory subsystems compare?

A: The Intel Arc Pro B60 Dual has 24 GB of GDDR6 memory on a 192-bit bus, delivering 456.0 GB/s bandwidth. The NVIDIA RTX 2000 Mobile Ada Generation has 8 GB of GDDR6 on a 128-bit bus, providing 256.0 GB/s. The Intel card offers three times the capacity and nearly double the bandwidth.

Q: What are the power requirements for each?

A: The Intel Arc Pro B60 Dual has a 400 W TDP, uses a single 16-pin power connector, and requires an 800 W suggested PSU. The NVIDIA RTX 2000 Mobile Ada Generation has a 50 W TDP, uses no external power connectors, and has no suggested PSU listed, as it is an IGP (integrated graphics processor) for mobile platforms.

Q: Which GPU has higher raw compute throughput?

A: The NVIDIA RTX 2000 Mobile Ada Generation has a slight edge in FP32 performance at 12.99 TFLOPS versus Intel's 12.29 TFLOPS. However, the Intel Arc Pro B60 Dual achieves 24.58 TFLOPS in FP16 (2:1 ratio), while the NVIDIA card delivers 12.99 TFLOPS in FP16 (1:1 ratio).

Q: What are the physical form factor differences?

A: The Intel Arc Pro B60 Dual is a dual-slot, 300 mm long, 110 mm tall, and 40 mm wide card with 4x mini-DisplayPort 2.1 outputs. The NVIDIA RTX 2000 Mobile Ada Generation is an IGP with no defined dimensions, no slot width, and displays dependent on the portable device.

Q: How do the shading and ray tracing resources compare?

A: The NVIDIA RTX 2000 Mobile Ada Generation has 3072 shading units, 96 TMUs, 48 ROPs, 24 RT cores, and 96 tensor cores. The Intel Arc Pro B60 Dual has 2560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores, with no tensor cores listed.

Where Each One Wins

The Intel Arc Pro B60 Dual wins decisively in memory-centric workloads. With 24 GB of GDDR6 versus 8 GB, it supports substantially larger datasets in GPU-accelerated rendering, AI inference, and professional visualization tasks. Its 456.0 GB/s bandwidth is 78% higher than the NVIDIA's 256.0 GB/s, which translates directly to faster texture streaming, larger frame buffers, and reduced swapping in memory-bound scenarios. The Intel card also leads in texture and pixel throughput: 384.0 GTexel/s versus 203.0 GTexel/s, and 192.0 GPixel/s versus 101.5 GPixel/s. For fill-rate-heavy workloads like high-resolution compositing or multi-viewport rendering, the Intel card holds a clear advantage.

The NVIDIA RTX 2000 Mobile Ada Generation wins in efficiency and portability. Its 50 W TDP is one-eighth of Intel's 400 W, making it suitable for mobile workstations where power budgets are tight and cooling is limited. The NVIDIA chip also has a slight FP32 compute lead, 12.99 TFLOPS versus 12.29 TFLOPS, which matters for general compute tasks that do not leverage FP16. Additionally, the NVIDIA card includes 96 tensor cores, which the Intel card lacks entirely, giving it a structural advantage in AI acceleration workloads that use Tensor Core instructions. The NVIDIA card also has more shading units (3072 versus 2560) and more RT cores (24 versus 20), providing a foundation for higher shader complexity and ray tracing throughput per clock.

Architecture Differences

The Intel Arc Pro B60 Dual uses the Xe2-HPG architecture, which is Intel's second-generation high-performance graphics design. It is built on the BMG-G21 chip with 19,600 million transistors spread across a 272 mm² die. The architecture uses a tile-based design with 2560 shading units organized into 20 RT cores. Notably, the Intel card has no tensor cores, relying instead on general-purpose compute for AI tasks. The FP16 throughput is double the FP32 rate, indicating a 2:1 ratio that accelerates mixed-precision workloads.

The NVIDIA RTX 2000 Mobile Ada Generation uses the Ada Lovelace architecture, NVIDIA's latest professional mobile design. It is built on the AD107 chip with 18,900 million transistors on a 159 mm² die, achieving a much higher transistor density of 118.9M / mm² versus Intel's 72.1M / mm². The Ada architecture includes dedicated tensor cores (96) and RT cores (24), giving it specialized hardware for AI and ray tracing. The FP16 throughput equals the FP32 rate at 12.99 TFLOPS (1:1 ratio), meaning it does not gain a speed advantage from reduced precision.

Both architectures support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical. The Intel card uses PCIe 5.0 x8 for its host interface, while the NVIDIA card uses PCIe 4.0 x16. The Intel card has a larger die but lower transistor density, indicating a more spread-out design with fewer transistors per area. The NVIDIA chip packs more transistors into a smaller area, reflecting a denser layout.

Specification Differences

| Specification | Intel Arc Pro B60 Dual | NVIDIA RTX 2000 Mobile Ada Generation |

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

| Chip | BMG-G21 | AD107 |

| Architecture | Xe2-HPG | Ada Lovelace |

| Transistors | 19,600 million | 18,900 million |

| Die Size | 272 mm² | 159 mm² |

| Transistor Density | 72.1M / mm² | 118.9M / mm² |

| Base Clock | 2000 MHz | 1635 MHz |

| Boost Clock | 2400 MHz | 2115 MHz |

| Memory Clock | 2375 MHz (19 Gbps effective) | 2000 MHz (16 Gbps effective) |

| Memory Size | 24 GB | 8 GB |

| Memory Bus Width | 192 bit | 128 bit |

| Memory Bandwidth | 456.0 GB/s | 256.0 GB/s |

| Shading Units | 2560 | 3072 |

| TMUs | 160 | 96 |

| ROPs | 80 | 48 |

| RT Cores | 20 | 24 |

| Tensor Cores | None | 96 |

| Pixel Rate | 192.0 GPixel/s | 101.5 GPixel/s |

| Texture Rate | 384.0 GTexel/s | 203.0 GTexel/s |

| FP32 | 12.29 TFLOPS | 12.99 TFLOPS |

| FP16 | 24.58 TFLOPS (2:1) | 12.99 TFLOPS (1:1) |

| TDP | 400 W | 50 W |

| Slot Width | Dual-slot | IGP |

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

| Suggested PSU | 800 W | None |

| Bus Interface | PCIe 5.0 x8 | PCIe 4.0 x16 |

| Display Outputs | 4x mini-DisplayPort 2.1 | Portable Device Dependent |

| Dimensions | 300 mm x 110 mm x 40 mm | Not specified |

| Release Date | 2025-09-04 | 2023-03-20 |

Head-to-Head Benchmarks

The recorded data shows no direct head-to-head benchmark scores, but the specification-derived metrics provide clear performance deltas. The most significant gap is in memory bandwidth: the Intel Arc Pro B60 Dual delivers 456.0 GB/s, which is 78% higher than the NVIDIA's 256.0 GB/s. This advantage directly impacts any workload that streams large textures, geometry, or simulation data.

In texture throughput, the Intel card reaches 384.0 GTexel/s, which is 89% higher than the NVIDIA's 203.0 GTexel/s. Similarly, pixel throughput favors Intel at 192.0 GPixel/s versus 101.5 GPixel/s, an 89% lead. These fill-rate advantages make the Intel card markedly faster for rasterization-heavy scenes with high resolution and multiple render targets.

The NVIDIA card counters with a small FP32 compute lead of 12.99 TFLOPS versus 12.29 TFLOPS, a 5.7% advantage. In FP16, however, the Intel card more than doubles NVIDIA's throughput: 24.58 TFLOPS versus 12.99 TFLOPS, a 89% lead. This makes the Intel card substantially faster for FP16-optimized AI and compute workloads, provided the software does not require Tensor Core instructions.

Clock speeds also differ: the Intel card has a 2000 MHz base and 2400 MHz boost, while the NVIDIA card has a 1635 MHz base and 2115 MHz boost. The Intel card's higher clocks contribute to its fill-rate advantages, but the NVIDIA card's higher shading unit count (3072 versus 2560) partially compensates in shader-bound scenarios. The NVIDIA card also has more RT cores (24 versus 20), which could benefit ray-traced workloads, though the Intel card's higher pixel and texture rates may offset this in hybrid rendering.

The memory clock difference is notable: Intel runs at 2375 MHz with 19 Gbps effective, while NVIDIA runs at 2000 MHz with 16 Gbps effective. Combined with the wider 192-bit bus versus 128-bit, the Intel card's memory system is clearly designed for higher data throughput.

The Verdict

The data indicates a clear split by use case. The Intel Arc Pro B60 Dual is the stronger choice for desktop professional workstations where power draw is not a constraint and memory capacity is critical. Its 24 GB frame buffer and 456.0 GB/s bandwidth support large-scale rendering, simulation, and data-intensive visualization tasks. The 89% lead in texture and pixel rates, plus the 89% lead in FP16 throughput, makes it the faster card for fill-rate-bound and mixed-precision workloads. The 400 W TDP and 800 W suggested PSU are acceptable for a dual-slot desktop card with 4x mini-DisplayPort 2.1 outputs.

The NVIDIA RTX 2000 Mobile Ada Generation is the appropriate choice for mobile workstations and power-constrained environments. Its 50 W TDP and IGP form factor enable deployment in laptops and compact systems where the Intel card cannot physically fit. The 96 tensor cores provide dedicated AI acceleration hardware that the Intel card lacks, and the slightly higher FP32 compute (12.99 TFLOPS versus 12.29 TFLOPS) gives it a small edge in general compute. The 8 GB memory and 256.0 GB/s bandwidth are sufficient for portable workloads but will limit large datasets.

Both cards occupy the 50th percentile in the database's all-GPU ranking, indicating comparable overall positioning. The release dates differ substantially: the Intel card launched on 2025-09-04, while the NVIDIA card launched on 2023-03-20. The NVIDIA card has defined predecessors and successors (Ampere-MW and Blackwell-MW), while the Intel card has none listed. The Intel card has a launch MSRP of 1,199 USD. The choice ultimately depends on whether the workload demands the Intel card's memory and throughput advantages or the NVIDIA card's efficiency and tensor core capabilities.