NVIDIA GeForce RTX 5080 SUPER vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded database contains a single direct head-to-head benchmark between these two cards: 3DMark Steel Nomad DX12. In that test, the NVIDIA GeForce RTX 5090 D scores 14,326 points, while the NVIDIA GeForce RTX 5080 SUPER scores 3,075 points. The 5090 D wins with a delta of -78.5 percent, which indicates the 5080 SUPER trails by that margin. This is a decisive victory for the 5090 D, delivering roughly 4.7 times the raw 3DMark score of the 5080 SUPER.

The RTX 5080 SUPER has an average benchmark score of 3,075 across all recorded tests, which consists solely of that single 3DMark result. The RTX 5090 D, by contrast, has a much broader benchmark profile. Its average score across all recorded tests is 77,712, a figure heavily influenced by its additional test results beyond the 3DMark run. The 5090 D also achieves a percentile ranking of 92 against all GPUs in the database, while the 5080 SUPER sits at the 19th percentile. This means the 5090 D outperforms the vast majority of recorded graphics cards, while the 5080 SUPER ranks among the lower portion. The percentile gap of 73 points is substantial and reinforces the head-to-head result.

Looking at the 5090 D's other benchmark scores, it shows strong performance across different APIs and workloads. In Geekbench OpenCL, it records 310,674 points, and in Geekbench Vulkan, it reaches 376,915 points. Passmark results for the 5090 D include a G3D score of 44,065, a GPU compute score of 28,396, a G2D score of 1,487, and DirectX 9 through 12 scores of 434, 371, 231, and 219 respectively. The 5080 SUPER has no recorded results in these other test suites, so a direct comparison is only possible for the 3DMark Steel Nomad test. The data confirms that the 5090 D is the clear performance leader in every measurable category where both cards have entries.

Architecture Differences

Both cards are built on the Blackwell 2.0 architecture and use a 5 nm process node from TSMC, but they employ different physical chips. The RTX 5080 SUPER uses the GB203 chip, while the RTX 5090 D uses the GB202 chip. The GB202 is substantially larger, with a die size of 750 mm² compared to 378 mm² for the GB203. Transistor counts differ accordingly: the 5090 D contains 92,200 million transistors, while the 5080 SUPER has 45,600 million. Transistor density is similar between the two, at 122.9 million per mm² for the 5090 D and 120.6 million per mm² for the 5080 SUPER.

Core configuration differences are large. The 5090 D has 21,760 shading units, 680 texture mapping units, and 176 render output units. The 5080 SUPER has 10,752 shading units, 336 texture mapping units, and 112 render output units. This means the 5090 D has roughly twice the shader count, twice the TMUs, and 1.6 times the ROPs. Ray tracing cores follow the same pattern: the 5090 D has 170 RT cores versus 84 on the 5080 SUPER. Tensor cores also double, with 680 on the 5090 D versus 336 on the 5080 SUPER.

Clock speeds tell a different story. The 5080 SUPER runs at a base clock of 2,295 MHz and a boost clock of 2,617 MHz. The 5090 D has a lower base clock of 2,017 MHz and a boost clock of 2,407 MHz. Despite lower clocks, the 5090 D delivers far higher throughput due to its larger core count. The FP32 compute rating for the 5090 D is 104.8 TFLOPS, nearly double the 56.28 TFLOPS of the 5080 SUPER. FP16 performance matches FP32 on both cards at a 1:1 ratio. Texture and pixel rates also favor the 5090 D: 1,636.8 GTexel/s versus 879.3 GTexel/s, and 423.6 GPixel/s versus 293.1 GPixel/s.

Memory architecture is another major divergence. The 5090 D uses a 512-bit memory bus with 32 GB of GDDR7 memory, delivering 1.79 TB/s of bandwidth. The 5080 SUPER uses a 256-bit bus with 24 GB of GDDR7 memory, providing 1.02 TB/s. Memory clocks differ as well, with the 5090 D running at 1,750 MHz (28 Gbps effective) and the 5080 SUPER at 2,000 MHz (32 Gbps effective). The 5080 SUPER has a higher memory clock, but the 5090 D's wider bus gives it a 75 percent bandwidth advantage.

Both cards are dual-slot designs with a single 16-pin power connector and a PCIe 5.0 x16 interface. Display outputs are identical in type: one HDMI 2.1b and three DisplayPort 2.1b connectors. The 5090 D specifies a suggested PSU of 950 W, while the 5080 SUPER has no suggested PSU field in the data. Power draw differs, with the 5090 D rated at 575 W and the 5080 SUPER at 415 W. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Release dates differ by roughly 11 months, with the 5090 D launching on January 29, 2025, and the 5080 SUPER on December 31, 2025. The 5090 D lists its predecessor as GeForce 40 and its successor as GeForce 60, while the 5080 SUPER has no predecessor or successor recorded.

Where Each One Wins

The 5090 D wins in every head-to-head category recorded in the database. Its 3DMark Steel Nomad DX12 score of 14,326 is 78.5 percent higher than the 5080 SUPER's 3,075. This makes the 5090 D the clear choice for workloads that stress raw DX12 rendering performance. The 5090 D also dominates in memory-heavy scenarios due to its 32 GB capacity and 1.79 TB/s bandwidth, which is useful for large texture sets or high-resolution rendering. Its 104.8 TFLOPS FP32 compute and 1,636.8 GTexel/s texture rate indicate strong performance in compute-intensive tasks such as physics simulation or video processing.

The 5080 SUPER has no recorded benchmark wins. Its higher boost clock of 2,617 MHz versus 2,407 MHz on the 5090 D gives it a clock speed advantage, but this does not translate into a performance lead in the available data. The 5080 SUPER also has a lower power draw of 415 W versus 575 W, which suggests it may be easier to cool and power, but the database does not include thermal or efficiency benchmarks to quantify this. Its smaller physical footprint, 40 mm width versus 48 mm on the 5090 D, could be relevant for space-constrained builds, but no performance metric captures this. In all recorded benchmarks, the 5090 D is the winner.

The percentile data reinforces this split. The 5090 D sits at the 92nd percentile among all GPUs, while the 5080 SUPER sits at the 19th. This means the 5090 D is positioned near the top of the database's performance distribution, while the 5080 SUPER is near the bottom. The 5090 D's nearest rivals in the database include the AMD Radeon RX 6650M XT at 76,904 average score (1.1 percent higher), the AMD Radeon RX 6850M XT at 78,940 (1.6 percent lower), the NVIDIA Tesla P100 PCIe 12 GB at 79,396 (2.1 percent lower), and the NVIDIA Tesla P100 PCIe 16 GB at 79,605 (2.4 percent lower). The 5080 SUPER's nearest rivals include the NVIDIA Quadro P1000 at 3,163 average score (2.8 percent higher), the NVIDIA GeForce 820A at 2,983 (3.1 percent lower), the Intel Arc Pro B60 at 3,182 (3.4 percent higher), and the NVIDIA GeForce GTX 860M at 2,967 (3.6 percent higher). These rival comparisons show the 5080 SUPER competing in a much lower performance tier.

Specification Differences

The following specifications differ between the two cards:

• Chip: GB203 (5080 SUPER) versus GB202 (5090 D)
• Transistors: 45,600 million versus 92,200 million
• Die size: 378 mm² versus 750 mm²
• Transistor density: 120.6M / mm² versus 122.9M / mm²
• Base clock: 2,295 MHz versus 2,017 MHz
• Boost clock: 2,617 MHz versus 2,407 MHz
• Memory clock: 2,000 MHz 32 Gbps effective versus 1,750 MHz 28 Gbps effective
• Memory size: 24 GB versus 32 GB
• Memory bus width: 256 bit versus 512 bit
• Memory bandwidth: 1.02 TB/s versus 1.79 TB/s
• Shading units: 10,752 versus 21,760
• TMUs: 336 versus 680
• ROPs: 112 versus 176
• RT cores: 84 versus 170
• Tensor cores: 336 versus 680
• Pixel rate: 293.1 GPixel/s versus 423.6 GPixel/s
• Texture rate: 879.3 GTexel/s versus 1,636.8 GTexel/s
• FP32: 56.28 TFLOPS versus 104.8 TFLOPS
• FP16: 56.28 TFLOPS (1:1) versus 104.8 TFLOPS (1:1)
• TDP: 415 W versus 575 W
• Width: 40 mm versus 48 mm
• Suggested PSU: not listed versus 950 W
• Release date: December 31, 2025 versus January 29, 2025
• Predecessor: not listed versus GeForce 40
• Successor: not listed versus GeForce 60
• Launch MSRP: 999 USD versus 2,299 USD

The two cards share the same architecture, process node, foundry, series, generation, slot width, power connector type, bus interface, display outputs, and API support. Both are dual-slot, use a single 16-pin connector, and measure 304 mm in length and 137 mm in height. Both are listed as Active production status.

FAQ

Q: Which card has more graphics memory?

A: The NVIDIA GeForce RTX 5090 D has 32 GB of GDDR7 memory, while the NVIDIA GeForce RTX 5080 SUPER has 24 GB of GDDR7 memory.

Q: What is the performance difference in 3DMark Steel Nomad DX12?

A: The RTX 5090 D scores 14,326 points, and the RTX 5080 SUPER scores 3,075 points. The 5090 D leads by 78.5 percent.

Q: Which card has a higher boost clock?

A: The RTX 5080 SUPER has a boost clock of 2,617 MHz, which is higher than the RTX 5090 D's 2,407 MHz boost clock.

Q: How does the memory bandwidth compare?

A: The RTX 5090 D provides 1.79 TB/s of bandwidth over a 512-bit bus, while the RTX 5080 SUPER provides 1.02 TB/s over a 256-bit bus.

Q: What are the power requirements?

A: The RTX 5090 D has a TDP of 575 W and lists a suggested PSU of 950 W. The RTX 5080 SUPER has a TDP of 415 W and does not list a suggested PSU in the data.

Q: Which card has more ray tracing cores?

A: The RTX 5090 D has 170 RT cores, while the RTX 5080 SUPER has 84 RT cores. The 5090 D also has 680 tensor cores versus 336 on the 5080 SUPER.

Q: How do the cards rank among all GPUs?

A: The RTX 5090 D is at the 92nd percentile, while the RTX 5080 SUPER is at the 19th percentile based on recorded benchmark scores.