Intel Arc Graphics 112EU Mobile vs NVIDIA GeForce RTX 5090 SE Comparison

Head-to-Head Benchmarks

The recorded data shows no direct head-to-head benchmark results between the Intel Arc Graphics 112EU Mobile and the NVIDIA GeForce RTX 5090 SE. Both entries in the database carry empty benchmark arrays, zero average scores, and identical 50th percentile positions among all GPUs. This makes a traditional numeric comparison impossible from the available measurements.

What the database does provide is a clear specification-level contrast. The Intel part operates as an integrated graphics processor within the Meteor Lake chip, while the RTX 5090 SE is a discrete add-in board. The raw compute figures illustrate the gap: Intel's FP32 throughput sits at 3.942 TFLOPS, whereas NVIDIA's part reaches 66.94 TFLOPS. That is a 16.98x difference in single-precision floating-point work per clock cycle, assuming both run at their respective boost clocks.

Pixel throughput tells a similar story. The Intel Arc Graphics 112EU Mobile delivers 52.80 GPixel/s, while the RTX 5090 SE hits 380.3 GPixel/s, a 7.2x advantage. Texture rate favors NVIDIA even more strongly: 123.2 GTexel/s versus 1,045.9 GTexel/s, an 8.49x margin. These are the largest single-metric wins for the discrete card in the database.

Clock speeds narrow the gap somewhat. Intel's base clock is 300 MHz with a boost of 2200 MHz. NVIDIA starts at 1740 MHz and boosts to 2377 MHz. The boost clocks are relatively close, but the Intel part has far fewer execution resources. The RTX 5090 SE packs 14,080 shading units, 440 TMUs, and 160 ROPs. The Intel chip has 896 shading units, 56 TMUs, and 24 ROPs. That means NVIDIA runs 15.7x more shaders, 7.86x more texture units, and 6.67x more raster units.

Memory configuration differs fundamentally. The Intel iGPU uses system shared memory with system-dependent bandwidth. The RTX 5090 SE carries 24 GB of GDDR7 on a 384-bit bus, delivering 1.34 TB/s of dedicated bandwidth. For memory-intensive workloads, the discrete card's advantage is structural rather than merely numerical.

The Verdict

From the recorded data, these two products occupy entirely different segments. The Intel Arc Graphics 112EU Mobile is an integrated solution rated at 65 W TDP, designed for portable devices. The NVIDIA GeForce RTX 5090 SE is a dual-slot discrete card with a 500 W TDP, requiring a 900 W suggested power supply and a 16-pin power connector.

Benchmark results indicate no overlap in target use cases. The Intel part sits in the 50th percentile among all GPUs in the database, and so does the RTX 5090 SE, but that percentile is based on the same empty benchmark set. Without actual scores, the percentile values carry no comparative weight.

The data confirms that the RTX 5090 SE is the stronger GPU by every measurable compute metric. The FP32 rate alone is 16.98x higher. The texture rate is 8.49x higher. The pixel rate is 7.2x higher. The Intel part has no ray tracing cores listed, while NVIDIA provides 110 RT cores and 440 tensor cores.

For a builder choosing between these two, the decision rests on system form factor and power budget. The Intel iGPU fits into a laptop or compact desktop with no additional power connectors. The RTX 5090 SE requires a full PCIe 5.0 x16 slot, a dual-slot cooler, a 900 W PSU, and a case that can handle a 267 mm long, 111 mm tall, 40 mm wide card.

Where Each One Wins

The Intel Arc Graphics 112EU Mobile wins in integration and power efficiency. It draws 65 W, uses system shared memory, connects via Ring Bus, and requires no dedicated power connector. It suits portable devices where space and cooling are constrained. Its predecessor is listed as HD Graphics-M, and it remains in active production.

The NVIDIA GeForce RTX 5090 SE wins in every raw performance category recorded. It has 24 GB of GDDR7 memory, a 384-bit bus, and 1.34 TB/s bandwidth. Its 66.94 TFLOPS FP32 rate is 16.98x the Intel part. The 440 TMUs and 160 ROPs support high-resolution texture work and heavy rasterization. The 110 RT cores and 440 tensor cores enable hardware ray tracing and AI acceleration, features the Intel chip does not list.

The RTX 5090 SE also leads in API support. It lists DirectX 12 Ultimate (12_2) versus Intel's DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The NVIDIA card provides 1x HDMI 2.1b and 3x DisplayPort 2.1b outputs, while the Intel part's display outputs are portable device dependent.

For workloads that fit within 65 W and shared memory, the Intel iGPU is the only option that fits the power envelope. For anything requiring dedicated memory bandwidth, high shader counts, or ray tracing, the RTX 5090 SE is the clear choice from the data.

FAQ

Q: Which GPU has higher FP32 compute?

A: The NVIDIA GeForce RTX 5090 SE delivers 66.94 TFLOPS, which is 16.98x the Intel Arc Graphics 112EU Mobile's 3.942 TFLOPS.

Q: How much memory does each GPU have?

A: The Intel Arc Graphics 112EU Mobile uses system shared memory with system-dependent bandwidth. The RTX 5090 SE has 24 GB of GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth.

Q: What are the power requirements?

A: The Intel iGPU is rated at 65 W TDP and uses no power connector. The RTX 5090 SE is rated at 500 W TDP, uses a 1x 16-pin connector, and requires a 900 W suggested PSU.

Q: Do both support ray tracing?

A: The Intel part lists no RT cores. The RTX 5090 SE has 110 RT cores and 440 tensor cores.

Q: What is the DirectX support difference?

A: The Intel Arc Graphics 112EU Mobile supports DirectX 12 (12_1). The RTX 5090 SE supports DirectX 12 Ultimate (12_2).

Q: What is the physical size of the RTX 5090 SE?

A: The card measures 267 mm in length, 111 mm in height, and 40 mm in width, and it is dual-slot.

Architecture Differences

The Intel Arc Graphics 112EU Mobile uses the Xe-LPG architecture on a 10 nm process node, built by Intel as part of the Meteor Lake chip. It belongs to the Arc Graphics-M (Meteor Lake) generation. Its predecessor is HD Graphics-M. The chip integrates 896 shading units, 56 TMUs, and 24 ROPs. No RT cores or tensor cores are listed. The memory interface is system shared, meaning the GPU relies on the host system's RAM.

The NVIDIA GeForce RTX 5090 SE uses the Blackwell 2.0 architecture on a 5 nm process node, fabricated by TSMC. The chip is GB202, containing 92,200 million transistors on a 750 mm² die, giving a transistor density of 122.9M per mm². It belongs to the GeForce 50 generation, with predecessor GeForce 40 and successor GeForce 60. The GPU has 14,080 shading units, 440 TMUs, 160 ROPs, 110 RT cores, and 440 tensor cores.

The process node difference is significant: 10 nm for Intel versus 5 nm for TSMC. The NVIDIA chip uses GDDR7 memory with a dedicated 384-bit interface, while Intel's design shares system memory. The bus interface differs as well: Intel uses Ring Bus, while NVIDIA uses PCIe 5.0 x16.

The Tensor cores on the RTX 5090 SE enable AI workloads, and the RT cores handle hardware ray tracing. The Intel part lists neither, indicating it relies on software or general-purpose shader execution for those tasks.

Specification Differences

The two GPUs differ in nearly every recorded specification field.

Process and foundry: Intel uses 10 nm at Intel. NVIDIA uses 5 nm at TSMC.

Transistors and die size: Intel lists no transistor count or die size. NVIDIA has 92,200 million transistors on a 750 mm² die with 122.9M transistors per mm².

Clocks: Intel base clock is 300 MHz, boost is 2200 MHz. NVIDIA base clock is 1740 MHz, boost is 2377 MHz. NVIDIA memory clock is 1750 MHz with 28 Gbps effective.

Memory: Intel uses system shared memory with system-dependent bandwidth. NVIDIA has 24 GB GDDR7, 384-bit bus, 1.34 TB/s bandwidth.

Shading units: Intel has 896. NVIDIA has 14,080.

TMUs: Intel has 56. NVIDIA has 440.

ROPs: Intel has 24. NVIDIA has 160.

RT cores: Intel lists none. NVIDIA has 110.

Tensor cores: Intel lists none. NVIDIA has 440.

Pixel rate: Intel is 52.80 GPixel/s. NVIDIA is 380.3 GPixel/s.

Texture rate: Intel is 123.2 GTexel/s. NVIDIA is 1,045.9 GTexel/s.

FP32: Intel is 3.942 TFLOPS. NVIDIA is 66.94 TFLOPS.

FP16: Intel is 7.885 TFLOPS (2:1). NVIDIA is 66.94 TFLOPS (1:1).

TDP: Intel is 65 W. NVIDIA is 500 W.

Slot width: Intel is IGP. NVIDIA is dual-slot.

Power connectors: Intel has none. NVIDIA uses 1x 16-pin.

Suggested PSU: Intel lists none. NVIDIA lists 900 W.

Bus interface: Intel uses Ring Bus. NVIDIA uses PCIe 5.0 x16.

Display outputs: Intel is portable device dependent. NVIDIA has 1x HDMI 2.1b and 3x DisplayPort 2.1b.

Dimensions: Intel lists none. NVIDIA is 267 mm x 111 mm x 40 mm.

DirectX: Intel supports 12 (12_1). NVIDIA supports 12 Ultimate (12_2).

Release date: Intel launched 2023-12-13. NVIDIA launches 2025-12-31.

Launch MSRP: Intel lists none. NVIDIA lists 1,499 USD.

Predecessor: Intel's is HD Graphics-M. NVIDIA's is GeForce 40.

Successor: Intel lists none. NVIDIA lists GeForce 60.

Both are in active production. Both have a 50th percentile among all GPUs in the database. Both have zero average benchmark scores and no nearest rivals recorded.