Intel Arc Graphics 64EU Mobile vs NVIDIA GeForce RTX 5090 SE Comparison

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark comparisons between the Intel Arc Graphics 64EU Mobile and the NVIDIA GeForce RTX 5090 SE. Both entries list empty benchmark arrays, and the wins counter shows zero for each side. Without measured frame rates, render times, or synthetic scores, the database cannot produce a direct performance delta between these two parts.

What can be established from the available figures is the theoretical throughput gap. The Intel part delivers 1.792 TFLOPS of FP32 compute, while the NVIDIA part delivers 66.94 TFLOPS. That places the RTX 5090 SE at roughly 37 times the raw floating-point throughput of the Arc Graphics 64EU Mobile. The pixel throughput difference is similarly stark: 28.00 GPixel/s versus 380.3 GPixel/s, a factor of about 13.6. Texture rate shows 56.00 GTexel/s on the Intel side versus 1,045.9 GTexel/s on the NVIDIA side, a factor of about 18.7.

Both products sit at the 50th percentile in the database's "percentileVsAllGpus" ranking, which means neither part has enough submitted benchmark data to push it above or below the median position. The average benchmark score for both is 0, confirming the absence of recorded performance samples. The head-to-head section therefore relies on specification-derived deltas rather than empirical test results.

Architecture Differences

The two GPUs come from different manufacturers, different foundries, and different design philosophies. Intel builds the Arc Graphics 64EU Mobile on a 10 nm process at Intel's own foundry, using the Meteor Lake chip and the Xe-LPG architecture. NVIDIA builds the RTX 5090 SE on a 5 nm process at TSMC, using the GB202 chip and the Blackwell 2.0 architecture. The node difference alone explains a substantial portion of the transistor density and power behavior gap.

The Intel part is an integrated graphics processor (IGP) with a Ring Bus interface, designed to live inside a mobile processor package. The NVIDIA part is a discrete dual-slot expansion card with a PCIe 5.0 x16 bus interface. The RTX 5090 SE carries 92,200 million transistors on a 750 mm² die, yielding a transistor density of 122.9M per mm². The Intel part lists no transistor count, die size, or density figure in the database, but its 10 nm Intel process and 512 shading units indicate a far smaller implementation.

Memory architecture separates the two completely. The Intel Arc Graphics 64EU Mobile uses System Shared memory for size, type, bus width, and bandwidth. Its memory clock is also System Shared, and its bandwidth is System Dependent. The RTX 5090 SE uses 24 GB of GDDR7 on a 384 bit bus, with 1.34 TB/s of bandwidth and a memory clock of 1750 MHz, 28 Gbps effective. Shared memory means the Intel GPU borrows bandwidth from the host system's RAM, while the NVIDIA card has dedicated VRAM with its own high-speed interface.

Ray tracing and tensor hardware also differ. The Intel part lists no RT cores and no tensor cores in the database. The RTX 5090 SE lists 110 RT cores and 440 tensor cores. This is a structural difference: the NVIDIA architecture has dedicated hardware paths for ray traversal and AI-accelerated workloads, while the Intel Xe-LPG implementation in this configuration does not expose such units in the recorded data.

The DirectX support level reflects the architectural split. Intel lists DirectX 12 (12_1), while NVIDIA lists DirectX 12 Ultimate (12_2). Both support OpenGL 4.6 and Vulkan 1.4. The 12_2 feature level on the NVIDIA side includes additional DXR and shader model support that the Intel part does not match in the recorded API profile.

Where Each One Wins

The Intel Arc Graphics 64EU Mobile wins on power efficiency targets and integration flexibility. Its 65 W TDP makes it suitable for thin-and-light mobile systems where the GPU shares thermal and power budgets with the CPU. The NVIDIA RTX 5090 SE carries a 500 W TDP and requires a 900 W suggested power supply, placing it in a completely different system class. The Intel part needs no power connectors because it is an IGP; the NVIDIA card requires a single 16-pin connector.

The Intel part also wins on physical footprint. As an IGP, it has no slot width, no length, no height, and no width in the database. It occupies no expansion slot and produces no add-in card dimensions. The NVIDIA card is a dual-slot unit measuring 267 mm in length, 111 mm in height, and 40 mm in width, or 10.5 by 4.4 by 1.6 inches. Systems without a full-length PCIe slot cannot accommodate the RTX 5090 SE at all.

The NVIDIA part wins on every raw performance metric in the database. Its shading unit count of 14,080 versus 512 gives it a 27.5-to-1 advantage in parallel execution lanes. Its 440 texture mapping units versus 32 gives a 13.75-to-1 advantage in texture work. Its 160 ROPs versus 16 gives a 10-to-1 advantage in pixel output. The 1.34 TB/s memory bandwidth against a system-dependent shared memory figure means the NVIDIA card will not contend with CPU traffic for memory access.

The RTX 5090 SE also wins on API feature completeness. DirectX 12 Ultimate (12_2) support exceeds the Intel part's DirectX 12 (12_1) level. The presence of 110 RT cores and 440 tensor cores gives the NVIDIA card dedicated acceleration for ray-traced rendering and tensor operations, neither of which the Intel part exposes in its specifications.

Specification Differences

| Specification | Intel Arc Graphics 64EU Mobile | NVIDIA GeForce RTX 5090 SE |

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

| Manufacturer | Intel | NVIDIA |

| Chip | Meteor Lake | GB202 |

| Architecture | Xe-LPG | Blackwell 2.0 |

| Process node | 10 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Not listed | 92,200 million |

| Die size | Not listed | 750 mm² |

| Base clock | 300 MHz | 1740 MHz |

| Boost clock | 1750 MHz | 2377 MHz |

| Memory size | System Shared | 24 GB |

| Memory type | System Shared | GDDR7 |

| Memory bus width | System Shared | 384 bit |

| Memory bandwidth | System Dependent | 1.34 TB/s |

| Shading units | 512 | 14080 |

| TMUs | 32 | 440 |

| ROPs | 16 | 160 |

| RT cores | Not listed | 110 |

| Tensor cores | Not listed | 440 |

| Pixel rate | 28.00 GPixel/s | 380.3 GPixel/s |

| Texture rate | 56.00 GTexel/s | 1,045.9 GTexel/s |

| FP32 | 1.792 TFLOPS | 66.94 TFLOPS |

| FP16 | 3.584 TFLOPS (2:1) | 66.94 TFLOPS (1:1) |

| TDP | 65 W | 500 W |

| Slot width | IGP | Dual-slot |

| Power connectors | Not listed | 1x 16-pin |

| Suggested PSU | Not listed | 900 W |

| Bus interface | Ring Bus | PCIe 5.0 x16 |

| Display outputs | Portable Device Dependent | 1x HDMI 2.1b, 3x DisplayPort 2.1b |

| DirectX | 12 (12_1) | 12 Ultimate (12_2) |

| Release date | 2023-12-13 | 2025-12-31 |

| Predecessor | HD Graphics-M | GeForce 40 |

| Successor | Not listed | GeForce 60 |

The FP16 ratio deserves attention. The Intel part achieves 3.584 TFLOPS FP16 at a 2:1 ratio relative to FP32, meaning it uses a packed math path that halves throughput when moving to full FP32 precision. The NVIDIA part delivers 66.94 TFLOPS FP16 at a 1:1 ratio, meaning FP16 and FP32 run at identical rates. This indicates the Blackwell architecture does not rely on packed half-precision tricks for its FP16 number.

The release dates place these products in different eras. The Intel Arc Graphics 64EU Mobile launched on 2023-12-13, while the RTX 5090 SE launched on 2025-12-31. The Intel part's predecessor is HD Graphics-M, and it has no successor listed. The NVIDIA part's predecessor is GeForce 40, and its successor is GeForce 60.

The display output situation also differs. The Intel IGP has Portable Device Dependent outputs, meaning the laptop or handheld manufacturer decides what physical ports exist. The NVIDIA card lists a fixed set: one HDMI 2.1b and three DisplayPort 2.1b connectors.

FAQ

Q: Which GPU has more shading units?

A: The NVIDIA GeForce RTX 5090 SE has 14,080 shading units, compared to 512 on the Intel Arc Graphics 64EU Mobile.

Q: What memory configuration does each GPU use?

A: The Intel Arc Graphics 64EU Mobile uses System Shared memory with System Dependent bandwidth. The NVIDIA GeForce RTX 5090 SE uses 24 GB of GDDR7 on a 384 bit bus with 1.34 TB/s bandwidth.

Q: Do both GPUs support the same DirectX version?

A: No. The Intel part supports DirectX 12 (12_1), while the NVIDIA part supports DirectX 12 Ultimate (12_2). Both support OpenGL 4.6 and Vulkan 1.4.

Q: What is the FP32 compute difference?

A: The Intel Arc Graphics 64EU Mobile delivers 1.792 TFLOPS, while the NVIDIA GeForce RTX 5090 SE delivers 66.94 TFLOPS, a difference of roughly 37 times in favor of NVIDIA.

Q: Does the Intel GPU have ray tracing cores?

A: The database lists no RT cores for the Intel Arc Graphics 64EU Mobile. The NVIDIA GeForce RTX 5090 SE lists 110 RT cores.

Q: What are the power requirements for each?

A: The Intel Arc Graphics 64EU Mobile has a 65 W TDP and lists no power connectors or suggested PSU. The NVIDIA GeForce RTX 5090 SE has a 500 W TDP, uses one 16-pin power connector, and lists a 900 W suggested power supply.