Intel Arc 130V Mobile vs NVIDIA N1 16SM Comparison

The two GPUs examined here, the Intel Arc 130V Mobile and the NVIDIA N1 16SM, occupy the same integrated graphics segment but approach it from entirely different design philosophies. The Intel part is built for efficiency on a 3 nm process, while the NVIDIA chip uses a 5 nm node and pairs its GPU with a massive 128 GB memory pool. Neither part currently has recorded benchmark scores in the database, so the analysis below relies entirely on the architectural specifications and raw computed throughput values provided.

Where Each One Wins

The data indicates a clear split in intended workloads. The Intel Arc 130V Mobile delivers 3.315 TFLOPS of FP32 performance and 6.630 TFLOPS of FP16 performance using a 2:1 ratio, which points toward consumer graphics and media tasks that benefit from packed math. Its 7 ray tracing cores support hardware-accelerated DirectX 12 Ultimate (12_2) features, making it suitable for modern game APIs and visual effects. The NVIDIA N1 16SM, by contrast, reaches 9.609 TFLOPS in both FP32 and FP16 at a 1:1 ratio, meaning it does not rely on packed arithmetic to reach its half-precision throughput. This configuration, combined with 64 tensor cores, suggests the NVIDIA part is oriented toward compute-heavy workloads where consistent precision is more important than peak theoretical output.

The Intel GPU wins in pixel processing efficiency relative to its shading resources. It has 28 ROPs and produces 51.80 GPixel/s, while the NVIDIA part has fewer ROPs at 24 but still reaches 56.30 GPixel/s due to its higher clock speed. The texture throughput gap is far larger: Intel manages 103.6 GTexel/s with 56 TMUs, while NVIDIA reaches 300.3 GTexel/s with 128 TMUs. That 2.9x advantage in texture rate indicates the NVIDIA chip is built for scenes with heavy texture sampling, common in professional visualization or complex 3D rendering.

Memory access patterns separate the two further. The Intel part uses system shared memory with bandwidth described as system dependent, meaning its performance scales with the host platform's memory configuration. The NVIDIA part has dedicated 128 GB of LPDDR5X memory on a 256 bit bus, delivering 273.2 GB/s of fixed bandwidth. That dedicated pool removes contention with the CPU and provides predictable performance for large datasets.

Architecture Differences

The process nodes differ significantly. Intel uses a 3 nm TSMC process for its Lunar Lake chip, which is part of the Arc Graphics-M generation and uses the Xe2-LPG architecture. NVIDIA uses a 5 nm TSMC process for its GB20B chip, built on the Blackwell 2.0 architecture, part of the Blackwell IGP (N1x) generation. The die sizes reflect their different scopes: Intel's die measures 172 mm², while NVIDIA's measures 382 mm², more than double the area.

The NVIDIA GPU integrates 2048 shading units, 128 TMUs, 24 ROPs, 16 ray tracing cores, and 64 tensor cores. The Intel GPU has 896 shading units, 56 TMUs, 28 ROPs, and 7 ray tracing cores, with no tensor core count listed. This means NVIDIA has 2.3x the shading units, 2.3x the TMUs, 2.3x the ray tracing cores, and a substantial tensor core presence that Intel lacks entirely in this configuration.

Clock behavior also differs. Intel runs at a base of 300 MHz with a boost of 1850 MHz, a wide range suggesting aggressive power management for mobile efficiency. NVIDIA runs at a base of 741 MHz with a boost of 2346 MHz, a higher floor and a higher ceiling. The NVIDIA memory clock is listed as 1067 MHz with 8.5 Gbps effective, while Intel's memory clock is system shared with no fixed value.

The API support diverges completely. Intel exposes DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it compatible with standard consumer graphics stacks. NVIDIA lists DirectX as N/A, OpenGL as N/A, and Vulkan as N/A, which indicates the N1 16SM is not designed for conventional gaming or desktop graphics APIs at all. Its output is limited to a single HDMI port, while Intel's display outputs are portable device dependent.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark results for these two GPUs, and neither has any individual benchmark entries. The wins count stands at zero for both parts. However, the specification data allows for computed throughput comparisons that reveal the relative strengths.

The FP32 compute gap is substantial. NVIDIA's 9.609 TFLOPS is 2.9x higher than Intel's 3.315 TFLOPS. That same ratio appears in texture rate, where NVIDIA's 300.3 GTexel/s compares to Intel's 103.6 GTexel/s, again a 2.9x difference. The consistency of these ratios suggests the NVIDIA design scales its compute and texture units proportionally.

The FP16 comparison is more nuanced. Intel reaches 6.630 TFLOPS using a 2:1 ratio, meaning it achieves double rate by processing two FP16 operations per FP32 operation. NVIDIA reaches 9.609 TFLOPS at 1:1, meaning its FP16 rate equals its FP32 rate without any packed arithmetic trick. In absolute terms, NVIDIA is 1.5x faster in FP16, but the architectural approach differs: Intel uses shader packing, while NVIDIA uses dedicated high-throughput units.

Pixel rate is the closest metric. NVIDIA produces 56.30 GPixel/s versus Intel's 51.80 GPixel/s, a modest 8.7% advantage. This narrow gap, despite NVIDIA's higher boost clock of 2346 MHz versus 1850 MHz, reflects Intel's higher ROP count of 28 versus NVIDIA's 24. The pixel rate gap is smaller than the compute or texture gaps, indicating that rasterization output is not where the NVIDIA part gains its largest advantage.

The memory bandwidth difference is extreme. NVIDIA's 273.2 GB/s on a 256 bit bus dwarfs Intel's system dependent shared memory. For workloads that stream large textures or datasets, NVIDIA's fixed bandwidth provides a consistent floor that Intel cannot guarantee without knowing the host system's memory speed.

Specification Differences

The two parts differ across nearly every major specification field. Intel uses a 3 nm process, NVIDIA uses 5 nm. Intel's die is 172 mm², NVIDIA's is 382 mm². Intel's base clock is 300 MHz with a boost of 1850 MHz, NVIDIA's is 741 MHz with a boost of 2346 MHz. Intel's memory is system shared with system dependent bandwidth, NVIDIA has 128 GB of LPDDR5X on a 256 bit bus with 273.2 GB/s.

The shading units number 896 for Intel and 2048 for NVIDIA. TMUs are 56 versus 128, ROPs are 28 versus 24, ray tracing cores are 7 versus 16, and tensor cores are absent for Intel while NVIDIA has 64. Intel's FP32 is 3.315 TFLOPS, NVIDIA's is 9.609 TFLOPS. Intel's FP16 is 6.630 TFLOPS at 2:1, NVIDIA's is 9.609 TFLOPS at 1:1.

Intel's pixel rate is 51.80 GPixel/s, NVIDIA's is 56.30 GPixel/s. Intel's texture rate is 103.6 GTexel/s, NVIDIA's is 300.3 GTexel/s. Intel's TDP is listed as 37 W, NVIDIA's TDP is unknown. Intel's bus interface is IGP, NVIDIA's is PCIe 5.0 x16. Intel supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while NVIDIA lists N/A for all three. Intel's display outputs are portable device dependent, NVIDIA has 1x HDMI. Intel's release date is 2024-09-23, NVIDIA's is 2026-05-31.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The NVIDIA N1 16SM delivers 9.609 TFLOPS, which is 2.9x higher than the Intel Arc 130V Mobile's 3.315 TFLOPS.

Q: Does either GPU support DirectX 12 Ultimate?

A: Yes, the Intel Arc 130V Mobile supports DirectX 12 Ultimate (12_2), along with OpenGL 4.6 and Vulkan 1.4. The NVIDIA N1 16SM lists DirectX, OpenGL, and Vulkan as N/A.

Q: What is the memory configuration of each GPU?

A: The Intel part uses system shared memory with system dependent bandwidth. The NVIDIA part has 128 GB of LPDDR5X memory on a 256 bit bus with 273.2 GB/s of bandwidth.

Q: How do the process nodes compare?

A: Intel uses a 3 nm TSMC process, while NVIDIA uses a 5 nm TSMC process. Intel's die measures 172 mm², and NVIDIA's measures 382 mm².

Q: Which GPU has more ray tracing cores?

A: The NVIDIA N1 16SM has 16 ray tracing cores, while the Intel Arc 130V Mobile has 7.

Q: Do both GPUs use the same FP16 approach?

A: No. Intel reaches 6.630 TFLOPS FP16 using a 2:1 ratio, meaning it doubles throughput for packed half-precision operations. NVIDIA reaches 9.609 TFLOPS FP16 at a 1:1 ratio, equal to its FP32 rate.

The Verdict

The data points to two different markets. The Intel Arc 130V Mobile, with its DirectX 12 Ultimate support, Vulkan 1.4, and 7 ray tracing cores, is positioned for consumer graphics workloads on portable devices. Its 37 W TDP and system shared memory reinforce this mobile-focused design. The NVIDIA N1 16SM, with no conventional graphics API support, a single HDMI output, and 64 tensor cores, appears tailored for compute and AI inference tasks where its 9.609 TFLOPS FP32 and FP16 throughput at 1:1 ratio can be fully utilized.

For traditional gaming or desktop graphics applications, the Intel part has the necessary API support and a lower TDP of 37 W. The NVIDIA part's lack of DirectX, OpenGL, and Vulkan support makes it incompatible with those workloads regardless of its higher raw throughput. For compute tasks that rely on FP32 or FP16 precision, the NVIDIA part's 2.9x higher FP32 throughput and 1.5x higher FP16 throughput, combined with 273.2 GB/s of dedicated memory bandwidth, give it a clear advantage. The 128 GB memory capacity also allows larger working sets than any system shared configuration.

The pixel rate difference is small at 56.30 GPixel/s versus 51.80 GPixel/s, suggesting neither part is constrained by ROP throughput. The texture rate difference is large at 300.3 GTexel/s versus 103.6 GTexel/s, which would matter for heavily textured rendering. The NVIDIA part's PCIe 5.0 x16 interface provides a wider host connection than Intel's IGP bus, though Intel's system shared memory architecture relies on the same memory pool as the CPU.

The release dates place Intel's part in late 2024 and NVIDIA's in mid-2026, so the NVIDIA chip is a later design with access to newer memory and interface standards. The 5 nm process is older than Intel's 3 nm node, but the larger die size of 382 mm² allows NVIDIA to pack more than double the shading units and 64 tensor cores. The Intel part achieves its results on a 172 mm² die with 896 shading units, indicating a focus on density and efficiency rather than absolute throughput.

Users with existing software that relies on DirectX 12 Ultimate or Vulkan will find support only in the Intel part. Users with compute pipelines that can ignore graphics APIs and take advantage of raw FP32 or FP16 throughput will find the NVIDIA part faster by a wide margin. The absence of benchmark scores in the database leaves these specification-derived conclusions as the primary basis for comparison, but the architectural differences are pronounced enough to indicate distinct target applications.