Intel Arc Pro B60 Dual vs NVIDIA N1 20SM Comparison

# Head-to-Head Benchmarks

The recorded data for the Intel Arc Pro B60 Dual and the NVIDIA N1 20SM shows both parts landing at the 50th percentile among all GPUs in the database, with an average benchmark score of zero for each. This parity in percentile rank indicates that neither card separates itself from the broader field in aggregate performance, but the underlying specifications reveal distinct strengths.

The Intel Arc Pro B60 Dual delivers 12.29 TFLOPS of FP32 compute, a modest 2.3% advantage over the NVIDIA N1 20SM's 12.01 TFLOPS. In FP16 workloads, the separation widens considerably. The Intel part reaches 24.58 TFLOPS with a 2:1 rate, while the NVIDIA N1 20SM is capped at 12.01 TFLOPS with a 1:1 ratio. This gives Intel a 104.7% lead in FP16 throughput, a decisive margin for applications that leverage mixed-precision arithmetic.

Texture processing favors Intel as well. The Arc Pro B60 Dual produces 384.0 GTexel/s against the N1 20SM's 375.4 GTexel/s, a 2.3% edge that aligns with the FP32 difference since both cards carry 160 TMUs. Pixel throughput tells a different story with a much larger gap: Intel's 192.0 GPixel/s more than triples the NVIDIA part's 56.30 GPixel/s, a 241% advantage stemming from the N1 20SM's 24 ROPs versus Intel's 80 ROPs.

Memory capacity and bandwidth pull in opposite directions. The NVIDIA N1 20SM offers 128 GB of LPDDR5X on a 256-bit bus, delivering 273.2 GB/s of bandwidth. The Intel Arc Pro B60 Dual provides 24 GB of GDDR6 on a 192-bit bus but achieves 456.0 GB/s, a 66.9% bandwidth advantage. This means Intel can feed its shaders faster, while NVIDIA's larger pool suits capacity-hungry workloads that tolerate lower throughput.

Clock behavior also differentiates the pair. Intel's base clock of 2000 MHz boosts to 2400 MHz, whereas NVIDIA starts at 741 MHz and boosts to 2346 MHz. The N1 20SM's boost clock sits only 2.3% below Intel's, but its base clock is 63% lower, suggesting a wider dynamic range and potentially higher idle efficiency for the integrated part.

The NVIDIA N1 20SM counters with 80 tensor cores, a feature the Intel Arc Pro B60 Dual lacks entirely. This gives NVIDIA a clear functional advantage in AI inference and training workloads that rely on tensor operations. Both cards include 20 ray tracing cores, so hardware-accelerated ray tracing capability is present on each side.

Rasterization output favors Intel decisively. The 80 ROPs on the Arc Pro B60 Dual produce 192.0 GPixel/s, while the N1 20SM's 24 ROPs yield only 56.30 GPixel/s. For traditional fill-rate-bound rendering, Intel holds a 3.4x advantage. The N1 20SM's lower ROP count appears consistent with its integrated design, where memory bandwidth and capacity take priority over pixel throughput.

# The Verdict

The data indicates two different design philosophies. The Intel Arc Pro B60 Dual is a discrete add-in card built for throughput: higher FP32, dramatically higher FP16, faster pixel fill, and greater memory bandwidth. The NVIDIA N1 20SM is an integrated graphics processor (IGP) with a 382 mm² die, 128 GB of LPDDR5X, and 80 tensor cores, aimed at systems where unified memory capacity and AI acceleration matter more than raw rasterization speed.

For FP32 compute, Intel leads by 2.3%. For FP16, Intel leads by 104.7%. For pixel rate, Intel leads by 241%. For memory bandwidth, Intel leads by 66.9%. For texture rate, Intel leads by 2.3%. These are the numbers as recorded, and they consistently point to Intel's advantage in conventional graphics and compute workloads.

The NVIDIA N1 20SM's advantages are structural rather than performance-based. Its 128 GB memory capacity is 5.3x larger than Intel's 24 GB. Its 80 tensor cores provide dedicated AI acceleration that Intel cannot match. Its 256-bit memory bus is 33% wider than Intel's 192-bit bus. Its PCIe 5.0 x16 interface doubles the lane count of Intel's PCIe 5.0 x8 connection. Its IGP form factor requires no power connector, whereas Intel uses a 1x 16-pin connector and an 800 W suggested PSU.

The launch MSRP for the Intel Arc Pro B60 Dual is 1,199 USD. The NVIDIA N1 20SM has no recorded launch MSRP.

Benchmark results indicate that users prioritizing raw graphics throughput, FP16 compute, or memory bandwidth should select the Intel Arc Pro B60 Dual. Users requiring massive unified memory capacity, tensor-based AI acceleration, or an integrated form factor should select the NVIDIA N1 20SM. The choice hinges on workload priorities, not on overall performance parity.

# Architecture Differences

The Intel Arc Pro B60 Dual uses the BMG-G21 chip built on the Xe2-HPG architecture, part of the Battlemage (Pro Series) generation. It is fabricated by TSMC on a 5 nm process with 19,600 million transistors across a 272 mm² die, yielding a transistor density of 72.1M per mm².

The NVIDIA N1 20SM uses the GB20B chip built on the Blackwell 2.0 architecture, part of the Blackwell IGP (N1x) generation. It is also fabricated by TSMC on a 5 nm process, but its transistor count is unknown. Its die size measures 382 mm², which is 40.4% larger than Intel's die, though no transistor density figure is recorded for comparison.

Both GPUs share identical counts of 2,560 shading units, 160 TMUs, and 20 RT cores. The NVIDIA part adds 80 tensor cores, which Intel does not include. ROP counts differ substantially: Intel has 80, NVIDIA has 24. This 56-ROP gap explains Intel's 3.4x pixel rate advantage.

The memory subsystems diverge in type and configuration. Intel uses 24 GB of GDDR6 on a 192-bit bus, while NVIDIA uses 128 GB of LPDDR5X on a 256-bit bus. Despite the narrower bus, Intel's GDDR6 runs at 2375 MHz (19 Gbps effective) versus NVIDIA's 1067 MHz (8.5 Gbps effective), producing higher bandwidth on Intel's side.

FP16 processing differs by design. Intel achieves 24.58 TFLOPS with a 2:1 ratio relative to FP32, effectively doubling throughput for half-precision work. NVIDIA's FP16 rate matches its FP32 rate at 12.01 TFLOPS with a 1:1 ratio, meaning no throughput gain for half-precision operations.

API support separates the two as well. Intel exposes DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. NVIDIA lists DirectX as N/A, OpenGL as N/A, and Vulkan as N/A, reflecting its IGP status and possibly different driver or software stack priorities.

# Specification Differences

Process and Die: Both use TSMC 5 nm. Intel's die is 272 mm² with 19,600 million transistors. NVIDIA's die is 382 mm² with unknown transistor count. Intel records a transistor density of 72.1M per mm²; NVIDIA records none.

Clocks: Intel base clock is 2000 MHz, boost is 2400 MHz. NVIDIA base clock is 741 MHz, boost is 2346 MHz. Memory clocks differ: Intel at 2375 MHz (19 Gbps effective), NVIDIA at 1067 MHz (8.5 Gbps effective).

Memory: Intel has 24 GB GDDR6, 192-bit bus, 456.0 GB/s bandwidth. NVIDIA has 128 GB LPDDR5X, 256-bit bus, 273.2 GB/s bandwidth.

Compute: FP32 is 12.29 TFLOPS for Intel versus 12.01 TFLOPS for NVIDIA. FP16 is 24.58 TFLOPS for Intel versus 12.01 TFLOPS for NVIDIA. Intel has no tensor cores; NVIDIA has 80.

Rasterization: Intel has 80 ROPs and 192.0 GPixel/s pixel rate. NVIDIA has 24 ROPs and 56.30 GPixel/s pixel rate. Texture rate is 384.0 GTexel/s for Intel and 375.4 GTexel/s for NVIDIA.

Power and Form Factor: Intel is a dual-slot card, 300 mm long, 110 mm tall, 40 mm wide, with a 1x 16-pin power connector and 800 W suggested PSU. NVIDIA is an IGP with no power connectors and no recorded dimensions or suggested PSU.

Interface and Outputs: Intel uses PCIe 5.0 x8 with 4x mini-DisplayPort 2.1 outputs. NVIDIA uses PCIe 5.0 x16 with 1x HDMI output.

Release and Status: Intel released on 2025-09-04 with a launch MSRP of 1,199 USD. NVIDIA releases on 2026-05-31 with no launch MSRP recorded. Both are listed as Active production.

# FAQ

Q: Which GPU has higher FP32 compute performance?

A: The Intel Arc Pro B60 Dual leads with 12.29 TFLOPS versus the NVIDIA N1 20SM's 12.01 TFLOPS, a 2.3% advantage.

Q: How does FP16 performance compare between the two?

A: Intel delivers 24.58 TFLOPS with a 2:1 FP16 ratio, while NVIDIA provides 12.01 TFLOPS with a 1:1 ratio, giving Intel a 104.7% lead in half-precision throughput.

Q: Which GPU offers more memory bandwidth?

A: The Intel Arc Pro B60 Dual achieves 456.0 GB/s from 24 GB of GDDR6 on a 192-bit bus. The NVIDIA N1 20SM provides 273.2 GB/s from 128 GB of LPDDR5X on a 256-bit bus.

Q: What AI acceleration does each GPU provide?

A: The NVIDIA N1 20SM includes 80 tensor cores. The Intel Arc Pro B60 Dual has no tensor cores recorded.

Q: How do the form factors differ?

A: Intel is a dual-slot discrete card measuring 300 mm by 110 mm by 40 mm, requiring a 1x 16-pin power connector and an 800 W suggested PSU. NVIDIA is an IGP with no power connectors and no recorded dimensions.

Q: What display outputs are available?

A: Intel provides 4x mini-DisplayPort 2.1 outputs. NVIDIA provides 1x HDMI output.