NVIDIA GeForce RTX 4080 SUPER vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded data shows a complete sweep for the NVIDIA GeForce RTX 5090 D across all ten benchmark tests, with zero wins for the RTX 4080 SUPER. The largest margin comes in the 3DMark Steel Nomad DX12 test, where the 5090 D scores 14,326 against 6,600 for the 4080 SUPER, a 117.1% advantage. That is more than double the performance in a modern DirectX 12 workload, and it sets the tone for the entire comparison.

In compute-oriented tests, the 5090 D remains dominant but with smaller gaps. Geekbench OpenCL shows 310,674 versus 219,065, a 41.8% lead. Geekbench Vulkan follows closely with 376,915 versus 260,075, a 44.9% advantage. The PassMark GPU Compute test records 28,396 against 19,822, which is 43.3% higher. These results indicate that the 5090 D scales strongly across different compute APIs, not just in one proprietary benchmark.

DirectX legacy tests show narrower but still decisive margins. PassMark DirectX 9 scores 434 versus 381, a 13.9% lead. PassMark DirectX 10 scores 231 versus 193, a 19.7% advantage. PassMark DirectX 11 scores 371 versus 301, a 23.3% lead. The DirectX 12 test shows a much larger gap: 219 versus 134, a 63.4% advantage. This pattern suggests the 5090 D benefits increasingly from newer API features and higher raw throughput.

The overall graphics score in PassMark G3D is 44,065 versus 34,245, a 28.7% lead. The 2D test, PassMark G2D, shows 1,487 versus 1,270, a 17.1% advantage. Even in the 2D test, the 5090 D pulls ahead, though the margin is smaller than in 3D workloads.

Average benchmark scores confirm the hierarchy. The 5090 D averages 77,712 across all recorded tests, while the 4080 SUPER averages 54,209. That is a 43.3% higher average for the 5090 D. The percentile rankings also differ: the 5090 D sits at the 92nd percentile among all GPUs, while the 4080 SUPER sits at the 86th percentile. The 5090 D also has a different competitive neighborhood. Its nearest rivals in the database are AMD Radeon RX 6650M XT at 76,904 (1.1% behind), AMD Radeon RX 6850M XT at 78,940 (1.6% ahead), and two NVIDIA Tesla P100 variants at 79,396 and 79,605 (2.1% and 2.4% ahead respectively). The 4080 SUPER's nearest rivals include the RTX 4080 at 54,247 (0.1% behind), AMD Radeon Pro W5700X at 54,828 (1.1% ahead), AMD Radeon RX 6750 GRE 12 GB at 55,698 (2.7% ahead), and AMD Radeon 8060S at 55,757 (2.8% ahead).

The Verdict

The data is unambiguous. The RTX 5090 D wins every single recorded benchmark against the RTX 4080 SUPER, with margins ranging from 13.9% to 117.1%. The 5090 D is the faster card in every measured category, from legacy DirectX 9 to modern DirectX 12, from OpenCL to Vulkan, from 2D to compute.

Users who need maximum performance in DirectX 12 workloads should choose the 5090 D. The 117.1% lead in 3DMark Steel Nomad is the single largest gap in the entire comparison, and it shows that the 5090 D excels in the most demanding modern rendering scenarios. Users who prioritize compute performance should also choose the 5090 D, as it leads by roughly 42% to 45% across OpenCL, Vulkan, and GPU Compute tests.

Users who are constrained by physical space or power delivery should consider the 4080 SUPER. The 4080 SUPER is a triple-slot card, but the 5090 D is dual-slot. The 4080 SUPER has a 320 W TDP and a recommended 700 W PSU, while the 5090 D has a 575 W TDP and a recommended 950 W PSU. The 4080 SUPER is also shorter at 310 mm versus 304 mm for the 5090 D, and it is narrower at 61 mm versus 48 mm. However, these physical differences do not change the performance outcome.

Architecture Differences

The two cards come from different architectural generations. The RTX 5090 D uses the GB202 chip built on Blackwell 2.0 architecture, while the RTX 4080 SUPER uses the AD103 chip built on Ada Lovelace architecture. Both are manufactured by TSMC on a 5 nm process, so the process node is identical.

The transistor counts differ substantially. The 5090 D packs 92,200 million transistors on a 750 mm² die, while the 4080 SUPER has 45,900 million transistors on a 379 mm² die. The transistor density is similar: 122.9M per mm² for the 5090 D versus 121.1M per mm² for the 4080 SUPER. The 5090 D has roughly double the transistors on roughly double the die area.

Memory technology also differs. The 5090 D uses GDDR7 memory with 32 GB capacity, a 512-bit bus, and 1.79 TB/s bandwidth. The 4080 SUPER uses GDDR6X memory with 16 GB capacity, a 256-bit bus, and 736.3 GB/s bandwidth. The 5090 D has double the memory capacity, double the bus width, and more than double the bandwidth.

The core configurations are significantly different. The 5090 D has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The 4080 SUPER has 10,240 shading units, 320 TMUs, 112 ROPs, 80 RT cores, and 320 tensor cores. The 5090 D has more than double the shading units, TMUs, RT cores, and tensor cores, while having 64 more ROPs.

The 5090 D also has a newer PCIe interface: PCIe 5.0 x16 versus PCIe 4.0 x16 on the 4080 SUPER. Display outputs differ as well. The 5090 D has 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the 4080 SUPER has 1x HDMI 2.1 and 3x DisplayPort 1.4a. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Specification Differences

The core clock speeds differ. The 5090 D has a base clock of 2017 MHz and a boost clock of 2407 MHz. The 4080 SUPER has a base clock of 2295 MHz and a boost clock of 2550 MHz. The 4080 SUPER actually has higher clock speeds, but the 5090 D compensates with far more cores.

Memory clocks also differ. The 5090 D runs at 1750 MHz with 28 Gbps effective, while the 4080 SUPER runs at 1438 MHz with 23 Gbps effective. The 5090 D has higher memory clocks and a wider bus, resulting in 1.79 TB/s versus 736.3 GB/s bandwidth.

The compute rates reflect the core count differences. The 5090 D delivers 104.8 TFLOPS FP32 and 104.8 TFLOPS FP16 (1:1). The 4080 SUPER delivers 52.22 TFLOPS FP32 and 52.22 TFLOPS FP16 (1:1). The 5090 D has exactly double the FP32 and FP16 throughput.

Pixel and texture rates also differ. The 5090 D has a pixel rate of 423.6 GPixel/s and a texture rate of 1,636.8 GTexel/s. The 4080 SUPER has a pixel rate of 285.6 GPixel/s and a texture rate of 816.0 GTexel/s. The 5090 D is 48.3% faster in pixel rate and 100.6% faster in texture rate.

Power specifications differ significantly. The 5090 D has a TDP of 575 W with a suggested PSU of 950 W. The 4080 SUPER has a TDP of 320 W with a suggested PSU of 700 W. Both use a single 16-pin power connector.

Physical dimensions also differ. The 5090 D is 304 mm long, 137 mm tall, and 48 mm wide. The 4080 SUPER is 310 mm long, 140 mm tall, and 61 mm wide. The 4080 SUPER is slightly longer and taller but significantly wider. However, the 5090 D is dual-slot while the 4080 SUPER is triple-slot.

Production status differs. The 5090 D is listed as Active, while the 4080 SUPER is End-of-life. The 5090 D launched on 2025-01-29, and the 4080 SUPER launched on 2024-01-30. The 5090 D has a launch MSRP of 2,299 USD, and the 4080 SUPER has a launch MSRP of 999 USD.

FAQ

Q: Which GPU wins more benchmark tests in the database?

A: The RTX 5090 D wins all 10 recorded head-to-head tests. The RTX 4080 SUPER wins none.

Q: What is the largest performance gap between the two cards?

A: The largest gap is in 3DMark Steel Nomad DX12, where the RTX 5090 D scores 14,326 versus 6,600 for the RTX 4080 SUPER, a 117.1% advantage.

Q: How do the memory configurations compare?

A: The RTX 5090 D has 32 GB of GDDR7 memory with a 512-bit bus and 1.79 TB/s bandwidth. The RTX 4080 SUPER has 16 GB of GDDR6X memory with a 256-bit bus and 736.3 GB/s bandwidth.

Q: Which card has higher clock speeds?

A: The RTX 4080 SUPER has higher base and boost clocks: 2295 MHz and 2550 MHz respectively. The RTX 5090 D has 2017 MHz base and 2407 MHz boost.

Q: What are the power requirements for each card?

A: The RTX 5090 D has a 575 W TDP and a suggested 950 W PSU. The RTX 4080 SUPER has a 320 W TDP and a suggested 700 W PSU.

Q: Which card has more shading units and tensor cores?

A: The RTX 5090 D has 21,760 shading units and 680 tensor cores. The RTX 4080 SUPER has 10,240 shading units and 320 tensor cores.

Where Each One Wins

The RTX 5090 D wins in every benchmark category. Its most decisive victories are in DirectX 12 workloads, where it leads by 117.1% in 3DMark Steel Nomad and 63.4% in PassMark DirectX 12. These tests favor the 5090 D's larger core count, newer architecture, and higher memory bandwidth. Users running modern DirectX 12 games or applications should expect the 5090 D to deliver roughly double the performance in the most demanding scenarios.

The 5090 D also wins decisively in compute workloads. Geekbench OpenCL shows a 41.8% lead, Geekbench Vulkan shows a 44.9% lead, and PassMark GPU Compute shows a 43.3% lead. These consistent margins across different compute APIs indicate that the 5090 D's advantage is not limited to a single vendor-specific path. Users doing GPU compute, machine learning inference, or scientific workloads will see substantial gains.

The 5090 D wins in legacy DirectX tests as well, but by smaller margins. PassMark DirectX 9 shows a 13.9% lead, DirectX 10 shows a 19.7% lead, and DirectX 11 shows a 23.3% lead. These older APIs may not fully utilize the 5090 D's newer features, but the raw core count still provides a meaningful advantage.

The RTX 4080 SUPER wins in no benchmark tests. However, it has advantages in other areas that may matter to certain users. It has a lower TDP of 320 W versus 575 W, a lower suggested PSU of 700 W versus 950 W, and a triple-slot cooler versus dual-slot. It is also 6 mm longer at 310 mm versus 304 mm, but 13 mm narrower at 61 mm versus 48 mm. Users with power or space constraints may prefer the 4080 SUPER, but the performance data consistently favors the 5090 D.

The 5090 D also holds advantages in production status and future longevity. It is listed as Active, while the 4080 SUPER is End-of-life. The 5090 D supports DisplayPort 2.1b, while the 4080 SUPER supports DisplayPort 1.4a. The 5090 D uses PCIe 5.0 x16, while the 4080 SUPER uses PCIe 4.0 x16. These newer standards may offer better compatibility with future hardware.