AMD Radeon RX 7900 XTX vs NVIDIA GeForce RTX 5090 D V2 Comparison

The Verdict

The data in this comparison is unusually one-sided. The NVIDIA GeForce RTX 5090 D V2 holds the only recorded head-to-head benchmark victory, and it is a decisive one. In the 3DMark Steel Nomad DX12 test, the RTX 5090 D V2 scores 16504 against the AMD Radeon RX 7900 XTX's 6841, a margin of 58.5% in NVIDIA's favor. For any user whose primary workload is high-end DX12 gaming or compute that scales with raw FP32 throughput, the selection is clear: the RTX 5090 D V2 is the stronger card in this pairing.

The AMD Radeon RX 7900 XTX does not win any of the recorded head-to-head tests. However, the benchmark database shows a much broader set of measurements for the AMD card, including OpenCL, Vulkan, and multiple Passmark tests. The RTX 5090 D V2's entry is limited to a single test, so its average benchmark score of 16504 comes from that one result. The RX 7900 XTX's average benchmark score across ten tests is 19410, which is higher than the RTX 5090 D V2's average, but that figure is pulled up by strong showings in Vulkan (85612) and OpenCL (50465) workloads that have no direct RTX 5090 D V2 counterpart in the database.

The verdict depends on workload priorities. If the task is specifically the kind of DX12 gaming load captured by 3DMark Steel Nomad, the RTX 5090 D V2 is the only logical pick. If the task involves a mix of compute APIs where the RX 7900 XTX has recorded results, the AMD card demonstrates capability in those areas, though no direct comparison exists for those tests. The RTX 5090 D V2 also carries a much higher launch MSRP of 2,299 USD versus the RX 7900 XTX's 999 USD, but the performance gap in the one shared test is substantial. The AMD card is end-of-life, while the NVIDIA card is active in production. For a new purchase targeting top-tier DX12 performance, the RTX 5090 D V2 is the data-backed choice.

Architecture Differences

The two cards come from different architectural generations with distinct design philosophies. The AMD Radeon RX 7900 XTX uses the Navi 31 chip built on RDNA 3.0 architecture, codenamed Plum Bonito, and belongs to the Navi III (RX 7000) generation. The NVIDIA GeForce RTX 5090 D V2 uses the GB202 chip built on Blackwell 2.0 architecture and belongs to the GeForce 50 generation.

Both are fabricated on a 5 nm process at TSMC, but the dies differ significantly in scale. The RTX 5090 D V2 packs 92,200 million transistors on a 750 mm² die, while the RX 7900 XTX has 57,700 million transistors on a 529 mm² die. The transistor density reflects this: the NVIDIA chip reaches 122.9M transistors per mm², while the AMD chip has 109.1M per mm².

The compute resources diverge sharply. The RTX 5090 D V2 has 21760 shading units, 680 texture mapping units, and 170 RT cores, plus 680 tensor cores. The RX 7900 XTX has 6144 shading units, 384 TMUs, and 96 RT cores, with no tensor core count recorded. The NVIDIA card also has 176 ROPs versus 192 on the AMD card, a rare specification where AMD leads.

Memory configurations share the same capacity and bus width: both have 24 GB and a 384 bit bus. However, the memory technology differs. The RTX 5090 D V2 uses GDDR7 at 1750 MHz with 28 Gbps effective speed, delivering 1.34 TB/s of bandwidth. The RX 7900 XTX uses GDDR6 at 2500 MHz with 20 Gbps effective speed, delivering 960.0 GB/s. The NVIDIA card has roughly 40% more memory bandwidth.

Clock behavior also differs. The RX 7900 XTX has a base clock of 1929 MHz, a boost clock of 2498 MHz, and a game clock of 2365 MHz. The RTX 5090 D V2 has a base clock of 2017 MHz and a boost clock of 2407 MHz, with no game clock recorded. The AMD card's boost clock is slightly higher, but the NVIDIA card's massive shading unit count produces far higher aggregate throughput. The RTX 5090 D V2 achieves 104.8 TFLOPS FP32 and 104.8 TFLOPS FP16 at a 1:1 ratio. The RX 7900 XTX achieves 61.39 TFLOPS FP32 and 122.8 TFLOPS FP16 at a 2:1 ratio.

Power delivery differs as well. The RTX 5090 D V2 has a 575 W TDP, uses a single 16-pin connector, and recommends a 950 W power supply. The RX 7900 XTX has a 355 W TDP, uses two 8-pin connectors, and recommends a 750 W power supply. The bus interface also differs: the NVIDIA card uses PCIe 5.0 x16, while the AMD card uses PCIe 4.0 x16.

Where Each One Wins

The recorded head-to-head data gives the RTX 5090 D V2 a single, overwhelming win in 3DMark Steel Nomad DX12. This test is a modern DX12 workload, and the NVIDIA card's score of 16504 versus 6841 for the RX 7900 XTX represents a 58.5% advantage. This is the only directly comparable benchmark between the two cards in the database.

Beyond that single test, the RX 7900 XTX has a wider set of recorded benchmarks that show its strengths in other API categories, even though no RTX 5090 D V2 scores exist for those tests. The AMD card scores 85612 in Geekbench Vulkan and 50465 in Geekbench OpenCL, indicating strong compute performance in those APIs. In Passmark tests, the RX 7900 XTX scores 31238 in G3D, 17689 in GPU Compute, 1267 in G2D, 356 in DirectX 11, 330 in DirectX 9, 166 in DirectX 10, and 131 in DirectX 12.

The RTX 5090 D V2's advantage in raw FP32 throughput is substantial: 104.8 TFLOPS versus 61.39 TFLOPS for the AMD card. Its texture rate is also far higher at 1,636.8 GTexel/s versus 959.2 GTexel/s. The pixel rate slightly favors AMD at 479.6 GPixel/s versus 423.6 GPixel/s, but this does not translate into a benchmark win in the recorded data.

For users focused on the single tested workload, DX12 gaming at high settings, the RTX 5090 D V2 is the clear winner. For users who value the AMD card's recorded performance in Vulkan and OpenCL compute, the RX 7900 XTX has demonstrated capability, but the absence of matching RTX 5090 D V2 results for those tests means no direct comparison can be drawn.

FAQ

Q: Which card wins the only directly comparable benchmark?

A: The NVIDIA GeForce RTX 5090 D V2 wins 3DMark Steel Nomad DX12 with a score of 16504 against the AMD Radeon RX 7900 XTX's 6841, a 58.5% margin.

Q: Do both cards have the same memory capacity?

A: Yes, both have 24 GB of memory on a 384 bit bus, but the RTX 5090 D V2 uses GDDR7 with 1.34 TB/s bandwidth, while the RX 7900 XTX uses GDDR6 with 960.0 GB/s.

Q: Which card has more shading units?

A: The RTX 5090 D V2 has 21760 shading units, far more than the RX 7900 XTX's 6144. The NVIDIA card also has 680 tensor cores, while no tensor core count is recorded for the AMD card.

Q: What is the TDP difference between the two cards?

A: The RTX 5090 D V2 has a 575 W TDP and recommends a 950 W power supply, while the RX 7900 XTX has a 355 W TDP and recommends a 750 W power supply.

Q: Is the RX 7900 XTX still in production?

A: No, the RX 7900 XTX is listed as end-of-life, while the RTX 5090 D V2 is active in production.

Q: Which card has a higher average benchmark score?

A: The RX 7900 XTX has an average benchmark score of 19410 across ten tests, while the RTX 5090 D V2 has an average of 16504 from a single test. However, the only shared test heavily favors the NVIDIA card.

Head-to-Head Benchmarks

The database contains exactly one head-to-head benchmark between these two cards: 3DMark Steel Nomad DX12. The results are stark. The NVIDIA GeForce RTX 5090 D V2 scores 16504, while the AMD Radeon RX 7900 XTX scores 6841. The delta is 58.5% in NVIDIA's favor, meaning the RTX 5090 D V2 delivers nearly two and a half times the score of the AMD card in this workload.

This result aligns with the raw compute specifications. The RTX 5090 D V2's FP32 throughput of 104.8 TFLOPS is roughly 70% higher than the RX 7900 XTX's 61.39 TFLOPS. The texture rate tells a similar story: 1,636.8 GTexel/s versus 959.2 GTexel/s, a 70% advantage. Memory bandwidth also favors NVIDIA at 1.34 TB/s versus 960.0 GB/s. The RX 7900 XTX does have a higher pixel rate at 479.6 GPixel/s versus 423.6 GPixel/s, and a higher boost clock at 2498 MHz versus 2407 MHz, but these advantages do not overcome the NVIDIA card's massive lead in shading units and memory bandwidth.

The RX 7900 XTX's percentile ranking among all GPUs is 64, while the RTX 5090 D V2 sits at 59. This is notable: despite losing the head-to-head test by a wide margin, the AMD card has a higher percentile ranking, likely because its ten recorded benchmarks include strong Vulkan and OpenCL results that lift its standing. The nearest rivals for the RX 7900 XTX include the NVIDIA GeForce GTX 1060 3 GB at a 0.4% delta, the NVIDIA Quadro K5200 at -1%, the AMD FirePro D300 at -1.2%, and the NVIDIA TITAN Xp at 1.2%. The RTX 5090 D V2's nearest rivals are the NVIDIA T400 at a 0% delta, the AMD Radeon PRO W7500 at 0.5%, the NVIDIA RTX PRO 6000 Blackwell at 0.6%, and the AMD Radeon RX 5700 XT at 0.9%. These rival groups indicate that the RTX 5090 D V2's single-test average places it in a different performance tier than its actual capabilities might suggest, given that its rivals in the database are low-end workstation cards.

The single benchmark result, however, is unambiguous. In DX12 Steel Nomad, the RTX 5090 D V2 is the faster card by a substantial margin, and no other direct comparison exists to temper that conclusion.

Specification Differences

The two cards differ in nearly every major specification category. The RX 7900 XTX uses the Navi 31 chip with RDNA 3.0 architecture, while the RTX 5090 D V2 uses the GB202 chip with Blackwell 2.0 architecture. The AMD card is end-of-life; the NVIDIA card is active.

Transistor counts and die sizes differ substantially. The RTX 5090 D V2 has 92,200 million transistors on a 750 mm² die, while the RX 7900 XTX has 57,700 million on a 529 mm² die. Transistor density favors NVIDIA at 122.9M per mm² versus 109.1M per mm².

Clock speeds differ in configuration. The RX 7900 XTX has a 1929 MHz base clock, 2498 MHz boost, and 2365 MHz game clock. The RTX 5090 D V2 has a 2017 MHz base clock and 2407 MHz boost, with no game clock recorded. Memory clocks also differ: the AMD card runs at 2500 MHz with 20 Gbps effective, while the NVIDIA card runs at 1750 MHz with 28 Gbps effective.

Compute resources show the largest divergence. The RTX 5090 D V2 has 21760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The RX 7900 XTX has 6144 shading units, 384 TMUs, 192 ROPs, and 96 RT cores, with no tensor cores recorded. The AMD card leads in ROP count but trails everywhere else.

Performance rates reflect these differences. The RTX 5090 D V2 achieves 104.8 TFLOPS FP32 and 104.8 TFLOPS FP16 at a 1:1 ratio. The RX 7900 XTX achieves 61.39 TFLOPS FP32 and 122.8 TFLOPS FP16 at a 2:1 ratio. Texture rates are 1,636.8 GTexel/s for NVIDIA versus 959.2 GTexel/s for AMD. Pixel rates are 423.6 GPixel/s for NVIDIA versus 479.6 GPixel/s for AMD.

Memory capacity and bus width are the same at 24 GB and 384 bit, but the memory type and bandwidth differ. The RTX 5090 D V2 uses GDDR7 with 1.34 TB/s bandwidth; the RX 7900 XTX uses GDDR6 with 960.0 GB/s.

Power and connectivity specifications also differ. The RTX 5090 D V2 has a 575 W TDP, a single 16-pin connector, and recommends a 950 W power supply. The RX 7900 XTX has a 355 W TDP, two 8-pin connectors, and recommends a 750 W power supply. The NVIDIA card uses PCIe 5.0 x16; the AMD card uses PCIe 4.0 x16. Both are dual-slot cards, but dimensions differ: the RTX 5090 D V2 measures 304 mm by 137 mm by 48 mm, while the RX 7900 XTX measures 287 mm by 110 mm by 51 mm. Display outputs also differ, with the AMD card offering 1x HDMI 2.1a, 2x DisplayPort 2.1, and 1x USB Type-C, while the NVIDIA card offers 1x HDMI 2.1b and 3x DisplayPort 2.1b. The release dates are far apart: the RX 7900 XTX launched on 2022-11-02, the RTX 5090 D V2 on 2025-08-14.