Intel Arc Pro B60 Dual vs NVIDIA RTX 4000 Mobile Ada Generation Comparison

FAQ

Q: What are the core architectural identities of the Intel Arc Pro B60 Dual and the NVIDIA RTX 4000 Mobile Ada Generation?

A: The Intel Arc Pro B60 Dual uses the Xe2-HPG architecture with the BMG-G21 chip, part of the Battlemage (Pro Series) generation. The NVIDIA RTX 4000 Mobile Ada Generation uses the Ada Lovelace architecture with the AD104 chip, part of the GeForce 40-series.

Q: How do the two cards compare in memory capacity and bandwidth?

A: The Intel card has 24 GB of GDDR6 memory on a 192-bit bus, delivering 456.0 GB/s bandwidth. The NVIDIA card has 12 GB of GDDR6 memory on a 192-bit bus, delivering 432.0 GB/s bandwidth. The Intel card offers double the capacity and a 5.6% bandwidth advantage.

Q: What are the power consumption figures for each card?

A: The Intel Arc Pro B60 Dual has a TDP of 400 W and requires a 1x 16-pin power connector with an 800 W suggested PSU. The NVIDIA RTX 4000 Mobile Ada Generation has a TDP of 110 W, uses no external power connectors, and has no suggested PSU listed.

Q: Which card has higher raw FP32 compute throughput?

A: The NVIDIA card delivers 24.72 TFLOPS of FP32 performance, which is approximately double the 12.29 TFLOPS of the Intel card. Both cards support FP16, but the NVIDIA card achieves this at a 1:1 ratio while the Intel card uses a 2:1 ratio for 24.58 TFLOPS.

Q: What are the physical form factor differences?

A: The Intel card is a dual-slot, 300 mm long, 110 mm high, 40 mm wide PCIe 5.0 x8 card with four mini-DisplayPort 2.1 outputs. The NVIDIA card is an IGP (integrated graphics processor) form factor with portable-device-dependent display outputs and PCIe 4.0 x16 interface.

Q: When were these products released?

A: The Intel Arc Pro B60 Dual has a release date of September 4, 2025, and an active production status. The NVIDIA RTX 4000 Mobile Ada Generation has a release date of March 20, 2023, with an active production status and a predecessor of Ampere-MW and successor of Blackwell-MW.

The Verdict

The data presents two distinctly different products serving different segments. The NVIDIA RTX 4000 Mobile Ada Generation is the compute-oriented choice for portable workstations, delivering roughly double the FP32 performance of the Intel card (24.72 TFLOPS vs 12.29 TFLOPS) while consuming only 110 W TDP. Its 7424 shading units, 232 tensor cores, and 58 RT cores provide substantial parallel processing resources in a compact IGP form factor. This card suits workloads where raw computational throughput per watt matters, particularly in mobile environments where the 400 W TDP of the Intel card would be impractical.

The Intel Arc Pro B60 Dual targets desktop professional workstations with its 24 GB memory capacity, which is double the NVIDIA card's 12 GB. It also has a higher boost clock (2400 MHz vs 1665 MHz) and a newer PCIe 5.0 x8 interface compared to the NVIDIA card's PCIe 4.0 x16. The Intel card's 456.0 GB/s memory bandwidth slightly exceeds the NVIDIA card's 432.0 GB/s. For workloads that are memory-capacity bound, such as large dataset processing or high-resolution rendering, the Intel card's 24 GB pool provides a clear advantage.

The architecture difference is significant. The Intel Xe2-HPG design uses 2560 shading units, 160 TMUs, and 80 ROPs, while the NVIDIA Ada Lovelace design uses 7424 shading units, 232 TMUs, and 80 ROPs. The NVIDIA card has nearly three times the shading units, enabling higher throughput in shader-bound workloads. The RT core counts also differ substantially: 20 on Intel vs 58 on NVIDIA, indicating the NVIDIA card has more dedicated ray tracing hardware.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical. The Intel card has a launch MSRP of 1,199 USD.

The verdict from the recorded data: choose the NVIDIA RTX 4000 Mobile Ada Generation for maximum compute performance in a low-power, portable form factor, and choose the Intel Arc Pro B60 Dual for double the memory capacity, higher clock speeds, and desktop workstation integration with PCIe 5.0 support.

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark results between these two products, and both currently hold a 50th percentile position among all GPUs with an average benchmark score of zero. However, the specification data provides clear quantitative comparisons.

The most decisive difference is FP32 compute throughput. The NVIDIA card delivers 24.72 TFLOPS, which is 12.43 TFLOPS higher than the Intel card's 12.29 TFLOPS, representing a 101% advantage. This doubling of compute throughput stems from the NVIDIA card's 7424 shading units operating at a lower 1665 MHz boost clock versus the Intel card's 2560 shading units at 2400 MHz boost. The NVIDIA card's higher shader count more than compensates for its lower clock speed.

Memory bandwidth shows a narrower margin. The Intel card achieves 456.0 GB/s versus the NVIDIA card's 432.0 GB/s, a 24 GB/s or 5.6% advantage. Both use GDDR6 memory on a 192-bit bus, but the Intel card runs at 2375 MHz (19 Gbps effective) compared to the NVIDIA card's 2250 MHz (18 Gbps effective).

Texture processing rates are nearly identical. The Intel card achieves 384.0 GTexel/s, while the NVIDIA card achieves 386.3 GTexel/s, a difference of only 2.3 GTexel/s in favor of NVIDIA. The Intel card's 160 TMUs at higher clocks nearly match the NVIDIA card's 232 TMUs at lower clocks.

Pixel fill rates differ more substantially. The Intel card produces 192.0 GPixel/s versus the NVIDIA card's 133.2 GPixel/s, giving Intel a 44% advantage. This results from the Intel card's higher clock speed combined with the same 80 ROP count.

Clock speeds show a significant gap. The Intel card has a 2000 MHz base clock and 2400 MHz boost, while the NVIDIA card has a 1290 MHz base and 1665 MHz boost. The Intel card's boost clock is 44% higher than NVIDIA's.

Transistor counts reveal different design approaches. The NVIDIA AD104 chip contains 35,800 million transistors on a 294 mm² die, while the Intel BMG-G21 chip contains 19,600 million transistors on a 272 mm² die. The NVIDIA chip achieves a transistor density of 121.8M per mm² versus Intel's 72.1M per mm².

Specification Differences

The two cards differ across nearly every major specification category. Memory capacity: Intel has 24 GB, NVIDIA has 12 GB. Memory clock: Intel runs at 2375 MHz (19 Gbps effective), NVIDIA at 2250 MHz (18 Gbps effective). Memory bandwidth: Intel achieves 456.0 GB/s, NVIDIA achieves 432.0 GB/s.

Shading units: Intel has 2560, NVIDIA has 7424. TMUs: Intel has 160, NVIDIA has 232. ROPs: both have 80. RT cores: Intel has 20, NVIDIA has 58. Tensor cores: Intel has none listed, NVIDIA has 232.

Clock specifications: Intel base clock is 2000 MHz, boost is 2400 MHz. NVIDIA base clock is 1290 MHz, boost is 1665 MHz. The Intel card's clocks are substantially higher across the board.

Power and physical specifications: Intel TDP is 400 W with a dual-slot form factor, 300 mm length, 110 mm height, 40 mm width, and a 1x 16-pin power connector. NVIDIA TDP is 110 W with an IGP form factor and no listed dimensions or power connectors.

Bus interface: Intel uses PCIe 5.0 x8, NVIDIA uses PCIe 4.0 x16. Display outputs: Intel provides 4x mini-DisplayPort 2.1, NVIDIA provides portable-device-dependent outputs.

Process technology: both use 5 nm at TSMC. Transistor counts differ: Intel has 19,600 million on a 272 mm² die, NVIDIA has 35,800 million on a 294 mm² die. Transistor density: Intel is 72.1M per mm², NVIDIA is 121.8M per mm².

Release dates: Intel released September 4, 2025, NVIDIA released March 20, 2023. The NVIDIA card has a predecessor (Ampere-MW) and successor (Blackwell-MW), while the Intel card lists neither.

Architecture Differences

The architectural designs diverge fundamentally. Intel's Xe2-HPG architecture with the BMG-G21 chip is built for the Battlemage Pro Series, targeting professional desktop workloads. It uses 2560 shading units arranged with 160 TMUs and 80 ROPs, plus 20 dedicated RT cores. The design relies on high clock speeds (2400 MHz boost) to achieve performance, with a relatively modest transistor count of 19,600 million on a 272 mm² die.

NVIDIA's Ada Lovelace architecture with the AD104 chip represents a different approach. It packs 7424 shading units, 232 TMUs, 80 ROPs, 58 RT cores, and 232 tensor cores into a 294 mm² die with 35,800 million transistors. The tensor cores provide dedicated AI acceleration hardware that the Intel card lacks entirely. The NVIDIA card operates at lower clocks (1665 MHz boost) but compensates with nearly three times the shading units and significantly more specialized hardware.

The FP16 computation paths differ: Intel achieves 24.58 TFLOPS at a 2:1 ratio, meaning it uses paired FP32 units for FP16 work. NVIDIA achieves 24.72 TFLOPS at a 1:1 ratio, indicating native FP16 throughput equal to its FP32 rate. This makes the NVIDIA implementation more efficient for FP16 workloads despite similar peak numbers.

Memory architecture is similar in bus width (192 bit for both) but different in capacity and clock. The Intel card's 24 GB GDDR6 at 2375 MHz provides 456.0 GB/s, while the NVIDIA card's 12 GB GDDR6 at 2250 MHz provides 432.0 GB/s. The Intel card's larger capacity suggests a design focus on memory-intensive professional applications.

Power delivery reflects the form factor difference. The Intel card's 400 W TDP with a 16-pin connector indicates a desktop-oriented design with robust cooling requirements. The NVIDIA card's 110 W TDP with no external power connectors confirms its mobile IGP design, optimized for battery-powered portable workstations.

Both cards support identical API levels: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. This ensures software compatibility parity across the two architectures.

Where Each One Wins

The Intel Arc Pro B60 Dual wins in memory capacity. With 24 GB versus 12 GB, it provides double the memory for workloads that require large datasets, high-resolution textures, or multiple simultaneous rendering tasks. This capacity advantage is critical for professional 3D modeling, scientific visualization, and AI inference with large models that exceed 12 GB.

The Intel card also wins in memory bandwidth at 456.0 GB/s versus 432.0 GB/s, a 5.6% advantage that compounds with the larger capacity. Its higher pixel fill rate of 192.0 GPixel/s versus 133.2 GPixel/s gives it a 44% advantage in rasterization-bound scenarios, making it stronger for traditional 2D and 3D rendering at high resolutions.

The Intel card's higher clock speeds (2400 MHz boost vs 1665 MHz) and PCIe 5.0 x8 interface provide advantages for latency-sensitive workloads and faster data transfer with modern platforms. Its four mini-DisplayPort 2.1 outputs support multi-display professional setups.

The NVIDIA RTX 4000 Mobile Ada Generation wins decisively in compute throughput. Its 24.72 TFLOPS FP32 is 101% higher than the Intel card's 12.29 TFLOPS. The 7424 shading units and 232 tensor cores provide substantial parallel processing for GPU compute tasks, scientific simulations, and machine learning workloads.

The NVIDIA card's 58 RT cores versus 20 provide nearly three times the dedicated ray tracing hardware, making it stronger for ray-traced rendering and real-time ray tracing workloads. The 232 tensor cores enable dedicated AI acceleration, a feature entirely absent from the Intel card's specification list.

The NVIDIA card wins in power efficiency. Its 110 W TDP is 27.5% of the Intel card's 400 W TDP, while delivering double the FP32 throughput. This makes it the only viable choice for mobile workstations, compact systems, or any deployment with power constraints.

The NVIDIA card's IGP form factor enables integration into thin-and-light portable workstations, while the Intel card requires a full desktop slot with dual-slot spacing and an 800 W suggested PSU. The NVIDIA card's PCIe 4.0 x16 interface provides more lanes than the Intel card's PCIe 5.0 x8, though the newer PCIe generation on Intel may offer higher per-lane bandwidth.

For users prioritizing raw compute, ray tracing, AI acceleration, and power efficiency, the NVIDIA card is the clear choice. For users needing maximum memory capacity, higher pixel fill rates, and desktop workstation integration with DisplayPort 2.1 outputs, the Intel card offers specific advantages.