GPU Comparison

Intel Iris Xe MAX Graphics and AMD Radeon RX 7900 XT occupy opposite ends of the hardware spectrum, yet the benchmark data reveals a peculiar head-to-head result. The Iris Xe MAX is a 25 W integrated-class part from Intel’s DG1 chip, built on a 10 nm process, while the RX 7900 XT is a 300 W dual-slot add-in board based on the Navi 31 chip and RDNA 3.0 architecture. The only shared benchmark in the data set is Geekbench OpenCL, where the Intel part wins outright, despite the AMD card holding a massive lead in raw specifications across every other measurable category. This page analyzes the available benchmark scores, architectural differences, and practical implications strictly from the provided facts.

Where Each One Wins

The data splits cleanly: the Intel Iris Xe MAX Graphics wins the only direct head-to-head comparison, while the AMD Radeon RX 7900 XT dominates every other benchmark category in its own test suite. In the Geekbench OpenCL test, the Intel part scores 14,315 against the AMD card’s 12,756, a 12.2% advantage. This is a surprising result given the architectural gap, but it is the sole data point where the two chips are compared directly.

For the AMD Radeon RX 7900 XT, the wins are extensive across its own benchmark list. It posts a Geekbench Vulkan score of 71,068, which is more than five times its OpenCL score and indicates strong performance in compute-heavy graphics workloads. In Passmark tests, the GPU scores 29,009 in G3D (DirectX 10/11/12 combined performance), 16,828 in GPU Compute, and 1,232 in G2D (2D graphics). The DirectX-specific Passmark scores are 340 for DirectX 11, 322 for DirectX 9, 158 for DirectX 10, and 110 for DirectX 12. The 3DMark Steel Nomad DX12 score is 5,623.

The Intel part, by contrast, has only the single Geekbench OpenCL score of 14,315 in its benchmark list. Its percentile rank among all GPUs is 56, while the AMD card sits at 55. This means the Intel chip is statistically near the median of all GPUs, while the AMD card, despite its much higher specification sheet, is also near the median in the aggregate ranking. The nearest rivals for the Intel part include the AMD Radeon Vega 11 (14,352, -0.3%), NVIDIA GeForce GTX 1070 Ti (14,277, +0.3%), and NVIDIA GeForce GTX TITAN (14,373, -0.4%). For the AMD card, the nearest rivals include the AMD Radeon 660M (13,812, -0.5%), NVIDIA RTX A2000 Mobile (13,821, -0.5%), AMD Radeon RX 570X (13,871, -0.9%), and AMD Radeon RX 9070 XT (13,543, +1.5%).

The use-case split is therefore stark: if the only metric is the head-to-head Geekbench OpenCL test, the Intel part wins. For every other workload, DirectX variants, Vulkan, 3DMark, Passmark G3D, GPU Compute, and 2D, the AMD card is the clear performer, with scores that dwarf the Intel part’s single data point.

The Verdict

Based strictly on the data, the Intel Iris Xe MAX Graphics is the correct choice for users whose primary workload is the Geekbench OpenCL benchmark. It beats the AMD Radeon RX 7900 XT by 12.2% in that test, and its 56th percentile ranking is marginally better than the AMD card’s 55th percentile. However, this is an extremely narrow use case. The AMD card wins every other benchmark category it was tested in, with scores that are orders of magnitude higher in Vulkan (71,068 vs. no comparable score) and Passmark G3D (29,009 vs. no comparable score).

For any real-world scenario that involves gaming, 3D rendering, or compute tasks beyond a single OpenCL test, the data strongly favors the AMD Radeon RX 7900 XT. It has 5,376 shading units versus 768, 336 texture mapping units versus 48, 192 ROPs versus 24, and 84 ray tracing cores versus none. Its memory subsystem is 20 GB of GDDR6 on a 320-bit bus with 800 GB/s bandwidth, compared to 4 GB of LPDDR4X on a 128-bit bus with 68.26 GB/s. The pixel rate is 459.6 GPixel/s versus 39.60 GPixel/s, and the texture rate is 804.4 GTexel/s versus 79.20 GTexel/s.

The verdict is not subtle: the Intel part wins one benchmark, the AMD part wins the rest. For users who need a GPU for general compute or graphics, the AMD card is the obvious pick. For users who specifically need to maximize a Geekbench OpenCL score, the Intel part is technically superior. The Intel chip is also end-of-life, as is the AMD card, but the AMD part’s successor is listed as Navi IV, while the Intel part’s successor is Alchemist.

Head-to-Head Benchmarks

The only head-to-head benchmark in the data set is Geekbench OpenCL. Intel Iris Xe MAX Graphics scores 14,315, while AMD Radeon RX 7900 XT scores 12,756. The Intel part wins with a 12.2% delta. This is a notable outlier because the AMD card’s specification sheet suggests it should be far faster: it has 7 times the shading units, 7 times the TMUs, 8 times the ROPs, and nearly 12 times the memory bandwidth (800 GB/s vs. 68.26 GB/s). The AMD card’s FP32 compute is 51.48 TFLOPS versus 2.534 TFLOPS for the Intel part, a 20-fold difference.

Why the Intel part wins this specific test is not explained by the data, but the result is clear. The Intel card’s nearest rivals in overall score are the AMD Radeon Vega 11 (14,352, -0.3%) and NVIDIA GeForce GTX 1070 Ti (14,277, +0.3%), meaning the Intel part is essentially tied with those older GPUs. The AMD card’s nearest rivals are much lower-scoring integrated and mobile parts like the AMD Radeon 660M (13,812, -0.5%) and NVIDIA RTX A2000 Mobile (13,821, -0.5%), despite the RX 7900 XT having far more hardware resources.

In other benchmark categories, the AMD card’s scores are its own wins, but there are no direct head-to-head comparisons. For example, the 3DMark Steel Nomad DX12 score of 5,623 and the Passmark G3D score of 29,009 are impressive numbers, but they have no Intel counterpart in the data. The AMD card also scores 1,232 in Passmark G2D, which is a 2D graphics test. The Intel part has no such scores listed.

The biggest win for the AMD card, relative to its own average, is the Geekbench Vulkan score of 71,068, which is 5.6 times its OpenCL score. This suggests the RDNA 3.0 architecture is particularly strong in Vulkan workloads. The Intel part has no Vulkan score, so no comparison is possible.

FAQ

Q: Which GPU wins the Geekbench OpenCL benchmark?

A: The Intel Iris Xe MAX Graphics wins with a score of 14,315, beating the AMD Radeon RX 7900 XT’s 12,756 by 12.2%.

Q: How do the two GPUs compare in overall percentile ranking?

A: The Intel part is at the 56th percentile of all GPUs, while the AMD card is at the 55th percentile. They are statistically tied in aggregate ranking.

Q: What is the AMD Radeon RX 7900 XT’s best benchmark score?

A: The highest score listed for the AMD card is the Geekbench Vulkan test at 71,068. Its Passmark G3D score is 29,009, and its 3DMark Steel Nomad DX12 score is 5,623.

Q: Does the Intel Iris Xe MAX Graphics have any ray tracing cores?

A: No. The Intel part lists no RT cores, while the AMD card has 84 ray tracing cores.

Q: What memory configurations do the two GPUs use?

A: The Intel part has 4 GB of LPDDR4X on a 128-bit bus with 68.26 GB/s bandwidth. The AMD card has 20 GB of GDDR6 on a 320-bit bus with 800.0 GB/s bandwidth.

Q: Are both GPUs end-of-life products?

A: Yes. Both the Intel Iris Xe MAX Graphics and the AMD Radeon RX 7900 XT have a production status of “End-of-life.”

Architecture Differences

The two GPUs are built on fundamentally different architectures and process nodes. The Intel Iris Xe MAX Graphics uses the DG1 chip with Generation 12.1 architecture (Xe Graphics), fabricated on a 10 nm process at Intel. Its die size is 95 mm², and it has 768 shading units, 48 TMUs, and 24 ROPs. The memory interface is 128-bit with LPDDR4X, running at 2133 MHz (4.3 Gbps effective). The base clock is 300 MHz with a boost of 1650 MHz. It has no dedicated RT cores, and its FP32 compute is 2.534 TFLOPS with FP16 at 5.069 TFLOPS (2:1). The thermal design power is 25 W, and it is listed as an IGP (integrated graphics processor) with no display outputs. It uses a PCIe 4.0 x8 interface.

The AMD Radeon RX 7900 XT uses the Navi 31 chip with RDNA 3.0 architecture (codenamed Plum Bonito), fabricated on a 5 nm process at TSMC. The die size is 529 mm² with 57,700 million transistors, giving a transistor density of 109.1M per mm². It has 5,376 shading units, 336 TMUs, 192 ROPs, and 84 RT cores. The memory is 20 GB of GDDR6 on a 320-bit bus with a bandwidth of 800.0 GB/s, running at 2500 MHz (20 Gbps effective). Clocks are 1387 MHz base, 2025 MHz game, and 2394 MHz boost. FP32 compute is 51.48 TFLOPS, and FP16 is 103.0 TFLOPS (2:1). The TDP is 300 W, requiring a 700 W suggested PSU and two 8-pin power connectors. It is a dual-slot card measuring 276 mm by 110 mm by 51 mm, with display outputs including 1x HDMI 2.1a, 2x DisplayPort 2.1, and 1x USB Type-C. The bus interface is PCIe 4.0 x16.

Key architectural differences include the process node (10 nm vs. 5 nm), core counts (768 vs. 5,376 shading units), memory type and capacity (4 GB LPDDR4X vs. 20 GB GDDR6), and the presence of RT cores (84 on AMD, none on Intel). The AMD card supports DirectX 12 Ultimate (12_2) while the Intel part supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The AMD card’s FP32 throughput is 20.3 times higher than the Intel part, and its texture rate is 10.2 times higher. The power envelope difference is stark: 300 W versus 25 W. The Intel part has no display outputs, functioning strictly as a compute or auxiliary processor, while the AMD card is a full-featured add-in board. The release dates are also different, with the Intel part launching on 2020-10-30 and the AMD card on 2022-11-02.