Intel Iris Xe Graphics 80EU Mobile vs NVIDIA N1X 48SM Comparison

The Verdict

The database comparison shows two fundamentally different integrated graphics solutions. The Intel Iris Xe Graphics 80EU Mobile is a mainstream IGP designed for thin-and-light portables, positioned at the 50th percentile among all GPUs. The NVIDIA N1X 48SM is a far larger Blackwell-based IGP, also at the 50th percentile, but with vastly different silicon resources. Benchmark results indicate the NVIDIA part is the clear performance leader on paper, with roughly 15.5 times the FP32 throughput of the Intel solution. The Intel part is the practical choice for systems requiring broad API support and established driver maturity, while the NVIDIA part targets high-bandwidth, high-throughput applications within a PCIe 5.0 platform.

Architecture Differences

The Intel Iris Xe Graphics 80EU Mobile is built on Raptor Lake silicon using Intel's Generation 12.2 architecture, fabricated on Intel's 10 nm process. It uses a Ring Bus interface and integrates 640 shading units, 40 texture mapping units, and 20 raster output units. The GPU operates at a base clock of 300 MHz with a boost clock of 1450 MHz.

The NVIDIA N1X 48SM is built on the GB20B chip using the Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC. The die measures 382 mm², a substantial physical footprint for an integrated part. It contains 6144 shading units, 384 TMUs, 48 ROPs, 48 ray tracing cores, and 192 tensor cores. Clocks range from 741 MHz base to 2346 MHz boost. The bus interface is PCIe 5.0 x16, a significant difference from Intel's Ring Bus.

The process node difference is notable: 10 nm Intel versus 5 nm TSMC. The NVIDIA part packs more than nine times the shading units into a dedicated 382 mm² die. The Intel part has no dedicated ray tracing or tensor cores, while the NVIDIA part includes both. API support also diverges sharply: Intel lists DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, while the NVIDIA part lists N/A for DirectX, OpenGL, and Vulkan, indicating a specialized compute-oriented feature set rather than a general-purpose graphics API stack.

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark entries, so the comparison rests on the raw specification deltas. The most striking gap is in FP32 performance. The Intel part delivers 1.856 TFLOPS, while the NVIDIA part delivers 28.83 TFLOPS, a 15.5-fold difference. This is the largest single numeric gap in the comparison.

Texture throughput shows a similar scale of separation. The Intel GPU achieves 58.00 GTexel/s, while the NVIDIA part achieves 900.9 GTexel/s, roughly 15.5 times higher. Pixel rate is less lopsided but still heavily favors NVIDIA: 29.00 GPixel/s versus 112.6 GPixel/s, a 3.9-fold difference.

Memory bandwidth is another decisive factor. The Intel part uses system shared memory with bandwidth listed as system dependent, meaning it cannot be quantified without a specific host platform. The NVIDIA part uses 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s of bandwidth. That fixed 273.2 GB/s figure stands in contrast to the variable, platform-dependent memory performance of the Intel IGP.

FP16 compute also differs in ratio. Intel lists 3.712 TFLOPS with a 2:1 ratio, meaning FP16 is exactly double FP32. NVIDIA lists 28.83 TFLOPS with a 1:1 ratio, meaning FP16 and FP32 throughput are identical. This indicates the NVIDIA part does not gain a compute advantage from reduced precision, while the Intel part doubles its throughput in FP16 workloads.

Clock behavior also favors NVIDIA. The Intel boost of 1450 MHz is lower than the NVIDIA boost of 2346 MHz. The NVIDIA base clock of 741 MHz is higher than the Intel base of 300 MHz, though both are integrated parts where sustained clocks depend on thermals and power delivery.

Specification Differences

The two GPUs differ across nearly every measurable field. The Intel part is built on a 10 nm process at Intel's foundry; the NVIDIA part uses a 5 nm process at TSMC with a 382 mm² die. Transistor counts are not listed for Intel and listed as unknown for NVIDIA.

Memory configurations are entirely different. Intel uses system shared memory with a system shared bus width and system dependent bandwidth. NVIDIA uses 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth. The memory clock for NVIDIA is listed as 1067 MHz with 8.5 Gbps effective data rate; Intel's memory clock is system shared.

Shading units: 640 versus 6144. TMUs: 40 versus 384. ROPs: 20 versus 48. Ray tracing cores: none versus 48. Tensor cores: none versus 192. Pixel rate: 29.00 GPixel/s versus 112.6 GPixel/s. Texture rate: 58.00 GTexel/s versus 900.9 GTexel/s. FP32: 1.856 TFLOPS versus 28.83 TFLOPS. FP16: 3.712 TFLOPS (2:1) versus 28.83 TFLOPS (1:1).

Power specifications: Intel lists 15 W TDP; NVIDIA lists unknown TDP. Both are IGP slot width with no power connectors listed for Intel and None for NVIDIA. The Intel bus interface is Ring Bus; NVIDIA uses PCIe 5.0 x16. Display outputs: Intel is portable device dependent; NVIDIA lists 1x HDMI.

API support differs completely. Intel supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. NVIDIA lists N/A for all three APIs, which is a decisive limitation for conventional graphics workloads.

Release dates: Intel launched 2023-01-03, NVIDIA launches 2026-05-31. Intel's successor is Arc Graphics-M; NVIDIA has no listed successor. Production status for both is Active.

FAQ

Q: Which GPU has higher raw compute throughput?

A: The NVIDIA N1X 48SM delivers 28.83 TFLOPS FP32, compared to 1.856 TFLOPS for the Intel Iris Xe Graphics 80EU Mobile, a 15.5-fold advantage.

Q: Do both GPUs support standard graphics APIs?

A: No. The Intel part supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The NVIDIA part lists N/A for DirectX, OpenGL, and Vulkan, indicating it is not positioned for conventional graphics API workloads.

Q: How does memory bandwidth compare?

A: The NVIDIA part has 273.2 GB/s of bandwidth from 128 GB of LPDDR5X on a 256-bit bus. The Intel part uses system shared memory with bandwidth listed as system dependent, so a fixed figure cannot be assigned.

Q: What is the difference in pixel throughput?

A: The NVIDIA part reaches 112.6 GPixel/s, while the Intel part reaches 29.00 GPixel/s, a 3.9-fold difference in favor of NVIDIA.

Q: Which GPU has ray tracing and tensor cores?

A: The NVIDIA N1X 48SM includes 48 ray tracing cores and 192 tensor cores. The Intel Iris Xe Graphics 80EU Mobile has no ray tracing cores and no tensor cores.

Q: What are the process nodes for each GPU?

A: The Intel part is fabricated on a 10 nm process at Intel, while the NVIDIA part uses a 5 nm process at TSMC with a die size of 382 mm².

Where Each One Wins

The NVIDIA N1X 48SM wins decisively in every raw throughput category recorded. Its 28.83 TFLOPS FP32 and FP16 (1:1) performance positions it for compute-heavy workloads that can utilize its 192 tensor cores and 48 ray tracing cores. The 273.2 GB/s memory bandwidth and 128 GB LPDDR5X capacity provide a fixed, high-bandwidth memory pool that does not depend on host system configuration. The 900.9 GTexel/s texture rate and 112.6 GPixel/s pixel rate give it substantial fill-rate advantages. The PCIe 5.0 x16 interface allows high-speed host communication, and the 2346 MHz boost clock indicates a high-frequency design. The 382 mm² die size reflects a large silicon investment, and the 5 nm TSMC process provides a density advantage over Intel's 10 nm node.

The Intel Iris Xe Graphics 80EU Mobile wins in API compatibility and ecosystem integration. It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, making it usable for standard graphics applications, games, and compute frameworks built on those APIs. The NVIDIA part lists N/A for all three, which excludes it from those workloads entirely. The Intel part also has a defined 15 W TDP, whereas the NVIDIA TDP is unknown, so the Intel solution offers a known power envelope for system designers. The Intel part was released in 2023, giving it a longer period of driver maturity and platform validation compared to the NVIDIA part, which launches in 2026. Intel's Ring Bus interface is suited to its integrated role in Raptor Lake systems, and its system shared memory model simplifies memory allocation in unified memory architectures. The Intel part's 3.712 TFLOPS FP16 (2:1) throughput, while far below NVIDIA's absolute numbers, still offers a 2:1 compute ratio that can accelerate certain FP16 workloads.

For conventional graphics rendering through established APIs, the Intel part is the only viable option in this comparison. For raw compute throughput, high bandwidth, and specialized cores, the NVIDIA part dominates every measured specification. The data shows two GPUs designed for different purposes: Intel targets broad compatibility and standard API support, while NVIDIA targets maximum throughput within a Blackwell IGP framework.