Intel Iris Xe MAX Graphics vs NVIDIA GeForce RTX 5090 D V2 Comparison

Where Each One Wins

The recorded data places these two graphics solutions in entirely different performance strata, and their benchmark results reflect that divide. The NVIDIA GeForce RTX 5090 D V2 posts a single recorded 3DMark Steel Nomad DX12 score of 16,504, while the Intel Iris Xe MAX Graphics reaches 14,315 in Geekbench OpenCL. These are not competing in the same workload class. The RTX 5090 D V2 sits at the 59th percentile among all GPUs in the database, whereas the Iris Xe MAX lands at the 56th percentile. The percentile gap looks modest, but the underlying scores tell a different story: the NVIDIA part is a discrete, high-power desktop accelerator built for sustained rendering tasks, while the Intel part is an integrated-class solution from 2020 with a 25 W envelope.

The RTX 5090 D V2 wins in every scenario where the database has a comparable metric. Its nearest rivals in the recorded data include the NVIDIA T400 (average score 16,508, delta 0%), the AMD Radeon PRO W7500 (16,415, delta 0.5%), the NVIDIA RTX PRO 6000 Blackwell (16,408, delta 0.6%), and the AMD Radeon RX 5700 XT (16,361, delta 0.9%). That clustering shows the RTX 5090 D V2 is effectively tied with a set of professional and mid-range cards in this particular DX12 test, all within a percentage point of each other. The Intel Iris Xe MAX, by contrast, sits near the AMD Radeon Vega 11 (14,352, delta -0.3%), the NVIDIA GeForce GTX 1070 Ti (14,277, delta 0.3%), the NVIDIA GeForce GTX TITAN (14,373, delta -0.4%), and the AMD Radeon RX Vega 11 (14,385, delta -0.5%). That rival group is a mix of older integrated graphics and legacy discrete cards, all roughly a decade old or more.

The use-case split is therefore stark. The RTX 5090 D V2 is the only one of the two with a recorded DX12 benchmark, and it delivers a score that places it alongside workstation-oriented GPUs like the T400 and PRO W7500. The Iris Xe MAX has no recorded DX12 result; its only data point is an OpenCL score, which is a compute-oriented API rather than a gaming or rasterization workload. The NVIDIA card also carries hardware features the Intel part lacks entirely: dedicated ray tracing cores (170 of them) and tensor cores (680 of them). The Intel part has no recorded RT cores or tensor cores at all. For any workload that leverages those units, the RTX 5090 D V2 is the only option with hardware support.

The Iris Xe MAX wins only in the sense of efficiency relative to its own class, but the database does not record any head-to-head benchmark wins for either part. The wins counters show zero for both A and B, meaning no direct comparative test exists in the recorded data. What the data does show is that the Intel part achieves its 14,315 OpenCL score with a 25 W TDP, 4 GB of LPDDR4X memory, and a 10 nm Intel process node. The RTX 5090 D V2 requires 575 W, 24 GB of GDDR7, and a 5 nm TSMC node to deliver its 16,504 Steel Nomad score. The RTX 5090 D V2 is also listed as Active production while the Iris Xe MAX is End-of-life, so the NVIDIA part is the only one with ongoing availability.

The Verdict

From the recorded data, the choice is clear for anyone building or buying today: the NVIDIA GeForce RTX 5090 D V2 is the only part with a current production status, a DX12 benchmark result, and hardware ray tracing and tensor cores. The Intel Iris Xe MAX is end-of-life, has no outputs of its own, and its only benchmark is an OpenCL compute test. If the workload is DX12 gaming or any modern rendering API, the data points entirely to the RTX 5090 D V2. If the workload is OpenCL compute at low power, the Iris Xe MAX has a recorded score, but it is 15.3% lower than the RTX 5090 D V2's Steel Nomad result when comparing raw numbers, and the two tests are not even the same benchmark.

The RTX 5090 D V2's launch MSRP is 2,299 USD. That is the only price figure in the database, and it is stated once here for reference. The Iris Xe MAX has no launch MSRP recorded. The RTX 5090 D V2 also has a successor listed (GeForce 60), while the Iris Xe MAX's successor is Alchemist, which has already come and gone given the Intel part's end-of-life status. For a system builder, the RTX 5090 D V2 offers a 384-bit memory bus, 1.34 TB/s bandwidth, and 104.8 TFLOPS FP32 performance. The Intel part offers 68.26 GB/s bandwidth and 2.534 TFLOPS FP32. Those are not comparable figures; they are two orders of magnitude apart in compute throughput.

The only scenario where the Iris Xe MAX makes sense from the data is a low-power, embedded, or legacy deployment where OpenCL compute on a 25 W part is the requirement, and where no discrete GPU can be installed due to physical constraints. Its slot width is listed as IGP, meaning it is an integrated graphics processor, not a card. The RTX 5090 D V2 is a dual-slot, 304 mm long card requiring a 950 W suggested PSU. They do not occupy the same market segment, and the database records no direct head-to-head wins for either, so any comparison is based on the separate benchmark scores and hardware specifications alone.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries for these two parts. The wins counter for both is zero. That absence is itself informative: the RTX 5090 D V2 and Iris Xe MAX have never been tested in the same suite, because they are not intended for the same use case. Instead, the database records one benchmark per part. The RTX 5090 D V2 scores 16,504 in 3DMark Steel Nomad DX12. The Iris Xe MAX scores 14,315 in Geekbench OpenCL. These are different tests with different APIs, so a direct percentage comparison is not valid, but the magnitude of the gap is worth noting. The RTX 5090 D V2's score is 15.3% higher than the Intel part's score in absolute terms, and the NVIDIA part does it in a DX12 test while the Intel part's test is a generic compute workload.

Looking at the rival clusters, the RTX 5090 D V2 is within 0.9% of four other GPUs, with the closest being the NVIDIA T400 at 16,508 (delta 0%) and the farthest being the AMD Radeon RX 5700 XT at 16,361 (delta 0.9%). That means the RTX 5090 D V2's Steel Nomad result is statistically tied with a range of cards from entry-level professional to mid-range gaming. The Iris Xe MAX's OpenCL score of 14,315 sits between the AMD Radeon Vega 11 (14,352, delta -0.3%) and the NVIDIA GeForce GTX TITAN (14,373, delta -0.4%), with the NVIDIA GeForce GTX 1070 Ti slightly behind at 14,277 (delta 0.3%). The Intel part is essentially tied with a 2017-era GTX 1070 Ti and a 2013-era GTX TITAN in OpenCL compute, which contextualizes its performance level.

The biggest win for the RTX 5090 D V2 is not in any single benchmark delta, but in the breadth of its specifications. It has 21,760 shading units versus 768 for the Intel part, 680 TMUs versus 48, and 176 ROPs versus 24. Its pixel rate is 423.6 GPixel/s versus 39.60 GPixel/s, and its texture rate is 1,636.8 GTexel/s versus 79.20 GTexel/s. Those are 10.7x, 14.2x, 7.3x, and 20.7x differences respectively. The memory bandwidth gap is 19.6x (1.34 TB/s versus 68.26 GB/s). The FP32 compute gap is 41.3x (104.8 TFLOPS versus 2.534 TFLOPS). The RTX 5090 D V2 also has 170 RT cores and 680 tensor cores, while the Intel part has none recorded. The only area where the Intel part is not drastically behind is in its FP16 ratio, where it achieves 5.069 TFLOPS at a 2:1 ratio, meaning it can double-rate FP16 relative to FP32, but even that figure is still far below the RTX 5090 D V2's 104.8 TFLOPS FP16 at a 1:1 ratio.

FAQ

Q: Which GPU has a higher recorded benchmark score?

A: The NVIDIA GeForce RTX 5090 D V2 scores 16,504 in 3DMark Steel Nomad DX12, while the Intel Iris Xe MAX Graphics scores 14,315 in Geekbench OpenCL. These are different tests, so the scores are not directly comparable, but the NVIDIA part's absolute number is 15.3% higher.

Q: Does the Intel Iris Xe MAX support ray tracing?

A: The database records no RT cores for the Intel part. The NVIDIA RTX 5090 D V2 has 170 RT cores and 680 tensor cores, so it is the only one of the two with dedicated hardware for those workloads.

Q: Is the Intel Iris Xe MAX still in production?

A: No. The production status is listed as End-of-life, with a release date of 2020-10-30. The NVIDIA RTX 5090 D V2 is listed as Active, with a release date of 2025-08-14.

Q: What is the memory configuration difference?

A: The RTX 5090 D V2 has 24 GB of GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth. The Iris Xe MAX has 4 GB of LPDDR4X on a 128-bit bus with 68.26 GB/s bandwidth.

Q: Which GPU has a smaller power requirement?

A: The Intel Iris Xe MAX has a 25 W TDP and a suggested PSU of 200 W. The NVIDIA RTX 5090 D V2 has a 575 W TDP and a suggested PSU of 950 W. The Intel part is an integrated graphics processor, while the NVIDIA part is a dual-slot discrete card.

Q: Are there any direct head-to-head benchmark results between these two?

A: No. The database records zero head-to-head wins for either part, and the only benchmarks are separate tests: Steel Nomad DX12 for the RTX 5090 D V2 and Geekbench OpenCL for the Iris Xe MAX.

Architecture Differences

The two parts come from different manufacturers, foundries, and process nodes. The NVIDIA GeForce RTX 5090 D V2 uses the GB202 chip on a 5 nm TSMC process, built on the Blackwell 2.0 architecture from the GeForce 50-series. The Intel Iris Xe MAX uses the DG1 chip on a 10 nm Intel process, built on Generation 12.1 architecture from the Xe Graphics generation. The transistor counts differ wildly: the NVIDIA part has 92,200 million transistors on a 750 mm² die, giving a density of 122.9 million transistors per mm². The Intel part has no recorded transistor count, but its die size is 95 mm², which is 7.9x smaller than the NVIDIA die.

The memory subsystems are architecturally distinct. The RTX 5090 D V2 uses GDDR7 with a 384-bit bus, while the Iris Xe MAX uses LPDDR4X with a 128-bit bus. The NVIDIA part's memory clock is listed as 1750 MHz with 28 Gbps effective, while the Intel part runs at 2133 MHz with 4.3 Gbps effective. The bus interfaces also differ: PCIe 5.0 x16 for the NVIDIA part versus PCIe 4.0 x8 for the Intel part. Display outputs are present on the RTX 5090 D V2 (1x HDMI 2.1b and 3x DisplayPort 2.1b), while the Iris Xe MAX has no outputs recorded, meaning it is designed for embedded use with no direct display connection.

The compute architectures diverge on ray tracing and tensor hardware. The RTX 5090 D V2 has 170 RT cores and 680 tensor cores, enabling hardware-accelerated ray tracing and AI workloads. The Intel part has no RT cores or tensor cores recorded. The shading unit counts are 21,760 versus 768, a 28.3x difference. The FP32 and FP16 ratios also differ: the NVIDIA part achieves 104.8 TFLOPS for both FP32 and FP16 at a 1:1 ratio, while the Intel part achieves 2.534 TFLOPS FP32 and 5.069 TFLOPS FP16 at a 2:1 ratio, meaning the Intel part trades FP32 throughput for double-rate FP16. The API support is similar in terms of DirectX and Vulkan versions (both support DirectX 12 and Vulkan 1.4), but the NVIDIA part supports DirectX 12 Ultimate (12_2) while the Intel part supports DirectX 12 (12_1).

Specification Differences

The most consequential specification gap is power. The RTX 5090 D V2 has a 575 W TDP with a 1x 16-pin power connector and a 950 W suggested PSU, while the Iris Xe MAX has a 25 W TDP, no power connector, and a 200 W suggested PSU. The slot width also differs: the NVIDIA part is dual-slot, while the Intel part is listed as IGP, meaning it is not a slot card at all. The RTX 5090 D V2 measures 304 mm in length, 137 mm in height, and 48 mm in width; the Intel part has no recorded dimensions.

Memory is another major split. The RTX 5090 D V2 has 24 GB of GDDR7 with a 384-bit bus and 1.34 TB/s bandwidth. The Iris Xe MAX has 4 GB of LPDDR4X with a 128-bit bus and 68.26 GB/s bandwidth. The clock speeds are also far apart: the NVIDIA part has a base clock of 2017 MHz and a boost clock of 2407 MHz, while the Intel part has a base clock of 300 MHz and a boost clock of 1650 MHz. The RTX 5090 D V2's memory clock is 1750 MHz (28 Gbps effective), versus 2133 MHz (4.3 Gbps effective) for the Intel part.

The rendering throughput metrics show the scale of the difference. Pixel rate is 423.6 GPixel/s versus 39.60 GPixel/s. Texture rate is 1,636.8 GTexel/s versus 79.20 GTexel/s. FP32 compute is 104.8 TFLOPS versus 2.534 TFLOPS. FP16 compute is 104.8 TFLOPS (1:1) versus 5.069 TFLOPS (2:1). The bus interface is PCIe 5.0 x16 versus PCIe 4.0 x8. The production status is Active versus End-of-life. The release dates are 2025-08-14 versus 2020-10-30. The RTX 5090 D V2 has a launch MSRP of 2,299 USD; the Intel part has none recorded. The NVIDIA part's predecessor is GeForce 40 and successor is GeForce 60; the Intel part's predecessor is Graphics and successor is Alchemist. The only specifications where the two match are the OpenGL version (4.6 for both) and the Vulkan version (1.4 for both).