Intel Iris Xe MAX Graphics vs NVIDIA GeForce GTX 1080 Comparison

Head-to-Head Benchmarks

The database contains a single directly comparable benchmark between the Intel Iris Xe MAX Graphics and the NVIDIA GeForce GTX 1080: the Geekbench OpenCL test. In this test, the Intel part scores 14,315, while the NVIDIA part scores 51,204. The recorded delta is -72%, meaning the GTX 1080 outperforms the Iris Xe MAX by a margin of 72 percentage points. That is a decisive victory for the NVIDIA card. The Intel GPU’s score lands within a tight cluster of similarly performing parts: the AMD Radeon Vega 11 averages 14,352 (just 0.3% ahead), the NVIDIA GeForce GTX 1070 Ti averages 14,277 (0.3% behind), the NVIDIA GeForce GTX TITAN averages 14,373 (0.4% ahead), and the AMD Radeon RX Vega 11 averages 14,385 (0.5% ahead). This places the Iris Xe MAX at the lower end of that peer group, though the spread among these rivals is less than one percent in either direction.

The GTX 1080, by contrast, sits in a different performance neighborhood. Its average benchmark score across all recorded tests is 11,960, which reflects a mix of DirectX, OpenCL, Vulkan, Metal, and compute workloads, not just the single OpenCL result. Its nearest rivals in the database are the NVIDIA GeForce GTX 960A (11,998 average, 0.3% lower), the AMD Radeon RX 6500 XT (11,842, 1% lower), the NVIDIA GeForce GTX 1660 (11,680, 2.4% lower), and the AMD Radeon RX 7800 XT (11,627, 2.9% lower). The GTX 1080 leads that group, with the closest competitor trailing by only a fraction of a percent. In the OpenCL head-to-head, the GTX 1080’s score is approximately 3.6 times higher than the Iris Xe MAX’s score, a gap that no other benchmark in the database contradicts.

The Iris Xe MAX’s single recorded benchmark, the OpenCL result, gives it a percentile ranking of 56 among all GPUs in the database. The GTX 1080, despite its much higher OpenCL score, has a percentile ranking of 51, because its average is dragged down by several lower-scoring tests such as the Passmark DirectX 12 score of 55 and the DirectX 10 score of 93. This illustrates a key point: the GTX 1080 is a strong OpenCL performer but has uneven results across other APIs. The Iris Xe MAX, having only the one OpenCL measurement, appears more consistent but that consistency is an artifact of sparse data.

FAQ

Q: Which GPU wins the direct OpenCL benchmark?

A: The NVIDIA GeForce GTX 1080 wins decisively, scoring 51,204 versus the Intel Iris Xe MAX’s 14,315, a difference of -72% from Intel’s perspective.

Q: How does the Intel Iris Xe MAX compare to its closest rivals?

A: The Iris Xe MAX’s 14,315 OpenCL score is within 0.5% of the AMD Radeon Vega 11 (14,352), the NVIDIA GeForce GTX 1070 Ti (14,277), the NVIDIA GeForce GTX TITAN (14,373), and the AMD Radeon RX Vega 11 (14,385). It is essentially tied with all four.

Q: What is the GTX 1080’s overall performance percentile?

A: The GTX 1080 sits at the 51st percentile among all GPUs in the database, despite its high OpenCL score, because its average benchmark score of 11,960 includes several low DirectX results.

Q: Does the Iris Xe MAX have any benchmark where it beats the GTX 1080?

A: No. In the only head-to-head benchmark recorded, the OpenCL test, the GTX 1080 wins. The database shows zero wins for the Intel part and one win for the NVIDIA part.

Q: How do the two GPUs compare in terms of memory bandwidth?

A: The GTX 1080 has a 256-bit bus and 320.3 GB/s bandwidth, while the Iris Xe MAX has a 128-bit bus and 68.26 GB/s bandwidth. The NVIDIA card offers roughly 4.7 times the bandwidth.

Q: Which GPU has a higher pixel rate?

A: The GTX 1080 has a pixel rate of 110.9 GPixel/s, while the Iris Xe MAX has 39.60 GPixel/s. The NVIDIA part is about 2.8 times faster in this metric.

Architecture Differences

The two GPUs come from different architectural generations and fabrication processes. The Intel Iris Xe MAX uses the DG1 chip, built on Generation 12.1 architecture, also referred to as Xe Graphics, on a 10 nm process node manufactured by Intel. Its die size is 95 mm². The NVIDIA GeForce GTX 1080 uses the GP104 chip, built on the Pascal architecture, on a 16 nm process node from TSMC, with a die size of 314 mm². The GTX 1080 contains 7,200 million transistors, giving a transistor density of 22.9 million per mm², while the Intel part has no transistor count listed in the database.

The compute configurations differ substantially. The Iris Xe MAX has 768 shading units, 48 texture mapping units, and 24 raster output units. The GTX 1080 has 2,560 shading units, 160 TMUs, and 64 ROPs. These numbers explain the large gaps in texture rate (79.20 GTexel/s for Intel versus 277.3 GTexel/s for NVIDIA) and pixel rate (39.60 GPixel/s versus 110.9 GPixel/s). The FP32 throughput is 2.534 TFLOPS for the Intel part versus 8.873 TFLOPS for the NVIDIA part. For FP16, the Intel GPU achieves 5.069 TFLOPS using a 2:1 ratio, while the GTX 1080 manages only 138.6 GFLOPS with a 1:64 ratio, meaning the Intel GPU is far more capable in half-precision work, but the NVIDIA GPU dominates in single precision.

Memory architecture also differs. The Intel GPU uses 4 GB of LPDDR4X on a 128-bit bus, with a memory clock of 2133 MHz and 68.26 GB/s bandwidth. The GTX 1080 uses 8 GB of GDDR5X on a 256-bit bus, with a memory clock of 1251 MHz and 320.3 GB/s bandwidth. The NVIDIA card has twice the capacity, twice the bus width, and nearly five times the bandwidth. The Intel GPU’s memory clock is higher, but the effective bandwidth is constrained by the narrower bus.

The bus interface differs as well: the Iris Xe MAX uses PCIe 4.0 x8, while the GTX 1080 uses PCIe 3.0 x16. The Intel part has no display outputs, while the GTX 1080 offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a. Both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Neither has dedicated ray tracing or tensor cores.

Specification Differences

The table below highlights the fields where the two GPUs differ, based exclusively on the recorded data.

| Specification | Intel Iris Xe MAX | NVIDIA GeForce GTX 1080 |

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

| Process node | 10 nm (Intel) | 16 nm (TSMC) |

| Die size | 95 mm² | 314 mm² |

| Transistors | Not listed | 7,200 million |

| Transistor density | Not listed | 22.9M / mm² |

| Base clock | 300 MHz | 1607 MHz |

| Boost clock | 1650 MHz | 1733 MHz |

| Memory clock | 2133 MHz, 4.3 Gbps effective | 1251 MHz, 10 Gbps effective |

| Memory size | 4 GB | 8 GB |

| Memory type | LPDDR4X | GDDR5X |

| Memory bus width | 128 bit | 256 bit |

| Memory bandwidth | 68.26 GB/s | 320.3 GB/s |

| Shading units | 768 | 2560 |

| TMUs | 48 | 160 |

| ROPs | 24 | 64 |

| Pixel rate | 39.60 GPixel/s | 110.9 GPixel/s |

| Texture rate | 79.20 GTexel/s | 277.3 GTexel/s |

| FP32 | 2.534 TFLOPS | 8.873 TFLOPS |

| FP16 | 5.069 TFLOPS (2:1) | 138.6 GFLOPS (1:64) |

| TDP | 25 W | 180 W |

| Slot width | IGP | Dual-slot |

| Power connectors | None listed | 1x 8-pin |

| Suggested PSU | 200 W | 450 W |

| Bus interface | PCIe 4.0 x8 | PCIe 3.0 x16 |

| Display outputs | No outputs | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a |

| Dimensions | Not listed | 267 mm length, 112 mm height, 40 mm width |

| Release date | 2020-10-30 | 2016-05-26 |

| Predecessor | Graphics | GeForce 900 |

| Successor | Alchemist | GeForce 20 |

| Launch MSRP | Not listed | 599 USD |

The Verdict

The data is unambiguous for raw OpenCL compute performance: the NVIDIA GeForce GTX 1080 is the stronger GPU by a wide margin. Its 51,204 OpenCL score is 72% higher than the Intel Iris Xe MAX’s 14,315, and its FP32 throughput, texture rate, pixel rate, and memory bandwidth all dwarf the Intel part. The GTX 1080 also has twice the VRAM (8 GB versus 4 GB), a wider memory bus, and a much higher boost clock. For users running OpenCL workloads, the GTX 1080 is the clear choice.

However, the Iris Xe MAX has its own advantages in the recorded specifications. It consumes only 25 W versus 180 W, occupies an IGP slot instead of a dual-slot card, requires a 200 W PSU versus 450 W, and has no power connectors. It is also a 2020 release, four years newer than the GTX 1080, and uses a smaller die (95 mm² versus 314 mm²) with a more advanced 10 nm process. Its FP16 throughput of 5.069 TFLOPS exceeds the GTX 1080’s 138.6 GFLOPS by a factor of 36, making it far more suitable for half-precision compute. The Intel part also has no display outputs, which suggests it is not intended for conventional desktop graphics output.

In a direct comparison, the GTX 1080 wins the only benchmark, and its architectural specifications align with that result. The Iris Xe MAX is not a competitor in this head-to-head for OpenCL or traditional rasterization performance. But for low-power, integrated form-factor use cases where FP16 throughput matters more than FP32, the Intel GPU presents a different trade-off. The verdict from the data: pick the GTX 1080 for any workload that demands raw single-precision compute, memory bandwidth, or display connectivity. Pick the Iris Xe MAX only in scenarios where power draw, physical footprint, and half-precision capability are the deciding factors, and where the lack of display outputs is acceptable.

Where Each One Wins

The NVIDIA GeForce GTX 1080 wins in every category that the database measures for compute throughput. It has a 3.6x higher OpenCL score, a 3.5x higher FP32 rate, a 3.5x higher texture rate, a 2.8x higher pixel rate, and a 4.7x higher memory bandwidth. Its 8 GB VRAM doubles the Intel part’s 4 GB, which matters for large datasets or high-resolution textures. The GTX 1080 also offers display outputs, making it usable as a primary graphics card, whereas the Iris Xe MAX has none. The NVIDIA card’s 1,733 MHz boost clock is also higher than the Intel’s 1,650 MHz, though the base clock difference is more pronounced (1,607 MHz versus 300 MHz).

The Intel Iris Xe MAX wins in power efficiency and form factor. Its 25 W TDP is 155 W lower than the GTX 1080’s 180 W. Its suggested PSU of 200 W versus 450 W means it can operate in far more constrained systems. Its IGP slot width and lack of power connectors indicate it is designed for embedded or mobile integration rather than expansion slots. The FP16 advantage is the most striking numerical win: 5.069 TFLOPS versus 138.6 GFLOPS, a 36.6x difference. For applications that rely on half-precision arithmetic, such as certain inference workloads, the Intel part is technically superior in that specific metric. The Iris Xe MAX also uses a newer process node (10 nm versus 16 nm) and a smaller die, which correlates with its lower power draw.

In summary, the GTX 1080 is the winner for conventional rendering, OpenCL compute, and any task that requires high memory bandwidth or large VRAM. The Iris Xe MAX wins only for low-power, compact deployments and for FP16-centric compute, with the caveat that its lack of display outputs limits its use as a standalone GPU. The benchmark record shows one decisive victory for NVIDIA, and the specification sheet reinforces that conclusion.