NVIDIA GeForce RTX 5080 vs NVIDIA GeForce RTX 5090 D Comparison

The NVIDIA GeForce RTX 5090 D and the NVIDIA GeForce RTX 5080 are two members of the same GeForce 50-series family, built on the same Blackwell 2.0 architecture and released on the same date. They are, however, positioned very differently within the lineup. The database shows the RTX 5090 D takes a clean sweep across every recorded benchmark in the head-to-head comparison, winning all ten tests. The RTX 5080 is not without merit, but the data reveals a consistent and often large performance gap that separates the two cards in nearly every workload category.

Head-to-Head Benchmarks

The most decisive result comes from the 3DMark Steel Nomad DX12 test, a modern DirectX 12 workload. Here the RTX 5090 D scores 14,326 points against the RTX 5080's 8,637 points, a difference of 65.9%. This is the largest single delta in the comparison and shows the 5090 D's advantage in a demanding contemporary rendering scenario is substantial. The gap is not a narrow margin; it is a wide canyon that appears right at the top of the workload list.

Compute-oriented benchmarks also show a strong lead for the 5090 D. In Geekbench OpenCL, the 5090 D records 310,674 points while the 5080 manages 235,901 points, a 31.7% advantage. The Vulkan test is even more pronounced, with the 5090 D hitting 376,915 points versus 255,450 points for the 5080, a 47.5% lead. These results suggest that the 5090 D's larger GPU configuration scales well into heterogeneous compute tasks, not just pure rasterization.

Passmark's suite provides a more granular view across different DirectX versions. The DirectX 12 test shows a 45% lead for the 5090 D, with scores of 219 versus 151. DirectX 11 shows a smaller but still clear 14.5% gap, 371 versus 324. DirectX 10 and DirectX 9 results are closer in percentage terms, with 11.1% and 11.6% leads respectively, but the 5090 D wins both. The passmark_g3d score, a general 3D graphics metric, comes in at 44,065 for the 5090 D versus 36,565 for the 5080, a 20.5% difference. The passmark_gpu_compute test shows a 30.3% advantage, 28,396 versus 21,789.

Even the 2D test, passmark_g2d, goes to the 5090 D, with a 5.1% lead. This is the smallest margin in the entire comparison, but it is still a win. The RTX 5090 D wins all ten head-to-head tests. The RTX 5080 does not win a single one. The margin varies from a modest 5.1% in 2D work to a commanding 65.9% in the Steel Nomad DX12 test. The pattern is clear: the 5090 D is faster across the board, and the advantage tends to grow in more modern and more compute-heavy workloads.

Where Each One Wins

The RTX 5090 D wins in every recorded category, but the size of the win tells a story about where the card's strengths lie. The largest margins are in the newest and most demanding workloads. The Steel Nomad DX12 test, the Vulkan compute test, and the DirectX 12 passmark test all show advantages of 45% or more. These are the tests that most heavily utilize modern GPU features, parallel processing, and raw throughput. The data implies the 5090 D is built to excel in exactly the kind of work that next-generation games and applications will demand.

The RTX 5080, while losing every test, is not a weak card in absolute terms. Its average benchmark score of 56,083 places it in the 87th percentile of all GPUs in the database. It sits close to rivals like the AMD Radeon 8060S, which scores 55,757, a difference of just 0.6%. The 5080 also edges out the AMD Radeon RX 6750 GRE 12 GB by 0.7%. The 5080 is a capable performer that competes with upper-mid-range and high-end cards from other vendors. The problem is not that the 5080 is slow, it is that the 5090 D is significantly faster.

The 5090 D, by contrast, sits in the 92nd percentile of all GPUs. Its average score of 77,712 is 1.1% ahead of the AMD Radeon RX 6650M XT and 1.6% behind the AMD Radeon RX 6850M XT. It trails the NVIDIA Tesla P100 PCIe 12 GB by 2.1% and the Tesla P100 PCIe 16 GB by 2.4%. These nearest rivals are a mix of mobile and data-center parts, which makes direct comparison less straightforward, but the 5090 D's percentile ranking confirms it is a top-tier product in the database's overall standings.

For a user deciding between these two cards, the use-case split is stark. If the workload is any of the tested scenarios, the 5090 D is the faster option. The 5080's closest competitive positioning is against other cards in its own price class, not against the 5090 D. The 5080 makes more sense for someone who wants a high-end Blackwell GPU but does not need the absolute peak of performance. The 5090 D is for those who want the maximum recorded performance in every single benchmark category.

Architecture Differences

Both cards are built on the Blackwell 2.0 architecture and use a 5 nm process from TSMC, but they are very different chips. The RTX 5090 D uses the GB202 chip, while the RTX 5080 uses the GB203 chip. The GB202 is a much larger die at 750 mm², compared to the GB203's 378 mm². Transistor counts differ accordingly: the 5090 D has 92,200 million transistors, while the 5080 has 45,600 million. Transistor density is similar, with the 5090 D at 122.9 million per mm² and the 5080 at 120.6 million per mm², indicating the process is being used to similar efficiency on both chips.

The compute resources scale up dramatically on the 5090 D. It has 21,760 shading units, 680 texture mapping units, and 176 ROPs. The 5080 has 10,752 shading units, 336 TMUs, and 112 ROPs. The 5090 D has exactly double the shading units and double the TMUs. Ray tracing cores also double, with 170 on the 5090 D versus 84 on the 5080. Tensor cores follow the same pattern, 680 versus 336. This doubling of core counts is the primary driver of the performance difference seen in the benchmarks.

Clock speeds tell a different story. The RTX 5080 actually runs at higher clocks. Its base clock is 2,295 MHz and its boost clock is 2,617 MHz, compared to the 5090 D's base of 2,017 MHz and boost of 2,407 MHz. The 5080 also has faster memory clocks, with its GDDR7 running at 1,875 MHz or 30 Gbps effective, versus the 5090 D's 1,750 MHz or 28 Gbps effective. Despite the lower clock speeds, the 5090 D's massive core count and wider memory bus deliver far higher throughput. The 5090 D achieves 104.8 TFLOPS of FP32 performance, while the 5080 reaches 56.28 TFLOPS.

Memory configuration is another major divider. The 5090 D has 32 GB of GDDR7 on a 512-bit bus, yielding 1.79 TB/s of bandwidth. The 5080 has 16 GB of GDDR7 on a 256-bit bus, yielding 960.0 GB/s of bandwidth. The 5090 D has double the memory capacity and nearly double the bandwidth. This is relevant for high-resolution textures and large datasets. The 5090 D also has higher pixel and texture rates, 423.6 GPixel/s and 1,636.8 GTexel/s, versus 293.1 GPixel/s and 879.3 GTexel/s on the 5080.

Power requirements differ substantially. The 5090 D has a TDP of 575 W and a suggested PSU of 950 W, while the 5080 has a TDP of 360 W and a suggested PSU of 750 W. Both use a single 16-pin power connector and are dual-slot cards. Physical dimensions are identical in length and height, 304 mm and 137 mm, but the 5090 D is thicker at 48 mm versus 40 mm. The launch MSRP of the RTX 5090 D is 2,299 USD. The launch MSRP of the RTX 5080 is 999 USD.

FAQ

Q: Which card wins in DirectX 12 workloads?

A: The RTX 5090 D wins decisively. In the 3DMark Steel Nomad DX12 test, it scores 14,326 versus 8,637, a 65.9% advantage. In the Passmark DirectX 12 test, it scores 219 versus 151, a 45% lead.

Q: Is the RTX 5080 competitive with the RTX 5090 D in any benchmark?

A: The RTX 5080 does not win any of the ten recorded head-to-head benchmarks. The closest margin is in the Passmark G2D test, where the 5090 D leads by only 5.1%. In every other test, the 5090 D leads by at least 11.1%.

Q: How do the memory configurations compare?

A: The RTX 5090 D has 32 GB of GDDR7 on a 512-bit bus with 1.79 TB/s of bandwidth. The RTX 5080 has 16 GB of GDDR7 on a 256-bit bus with 960.0 GB/s of bandwidth. The 5090 D has double the capacity and roughly double the bandwidth.

Q: Which card has higher clock speeds?

A: The RTX 5080 has higher clock speeds. Its base clock is 2,295 MHz and boost is 2,617 MHz, while the 5090 D has a base of 2,017 MHz and boost of 2,407 MHz. The 5080 also has faster memory clocks at 30 Gbps effective versus 28 Gbps effective.

Q: What are the power requirements for each card?

A: The RTX 5090 D has a TDP of 575 W and a suggested PSU of 950 W. The RTX 5080 has a TDP of 360 W and a suggested PSU of 750 W. Both use a single 16-pin power connector.

Q: How does each card rank against all other GPUs in the database?

A: The RTX 5090 D is in the 92nd percentile of all GPUs, with an average benchmark score of 77,712. The RTX 5080 is in the 87th percentile, with an average score of 56,083.

Specification Differences

The two cards differ in nearly every core specification. The chip is GB202 on the 5090 D versus GB203 on the 5080. Die size is 750 mm² versus 378 mm². Transistor count is 92,200 million versus 45,600 million. Shading units are 21,760 versus 10,752. TMUs are 680 versus 336. ROPs are 176 versus 112. Ray tracing cores are 170 versus 84. Tensor cores are 680 versus 336. FP32 performance is 104.8 TFLOPS versus 56.28 TFLOPS. Pixel rate is 423.6 GPixel/s versus 293.1 GPixel/s. Texture rate is 1,636.8 GTexel/s versus 879.3 GTexel/s.

Memory size is 32 GB versus 16 GB. Memory bus width is 512 bit versus 256 bit. Memory bandwidth is 1.79 TB/s versus 960.0 GB/s. Base clock is 2017 MHz versus 2295 MHz. Boost clock is 2407 MHz versus 2617 MHz. Memory clock is 1750 MHz 28 Gbps effective versus 1875 MHz 30 Gbps effective. TDP is 575 W versus 360 W. Suggested PSU is 950 W versus 750 W. Card width is 48 mm versus 40 mm. Launch MSRP is 2,299 USD versus 999 USD.

The two cards share several specifications. Both use the Blackwell 2.0 architecture, a 5 nm TSMC process, GDDR7 memory, dual-slot cooling, a single 16-pin power connector, PCIe 5.0 x16, and the same display outputs of 1x HDMI 2.1b and 3x DisplayPort 2.1b. They also share identical length and height dimensions, 304 mm and 137 mm, and the same API support for DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. Both are currently marked as Active in production status.

The Verdict

The data is unambiguous. The RTX 5090 D wins all ten recorded head-to-head benchmarks, with advantages ranging from 5.1% to 65.9%. Its average benchmark score of 77,712 is significantly higher than the 5080's 56,083. The 5090 D also ranks higher in the overall database, sitting in the 92nd percentile versus the 5080's 87th. For any user who needs the highest recorded performance in these workloads, the 5090 D is the clear choice.

The RTX 5080 is still a high-performing card in its own right. Its 87th percentile ranking and its close competition with cards like the AMD Radeon 8060S and the AMD Radeon RX 6750 GRE 12 GB show it is a solid option in the broader market. Its lower TDP of 360 W and lower suggested PSU of 750 W make it a more manageable card for system integration. Its higher clock speeds and faster memory clock are notable, but they do not overcome the core count and memory bandwidth advantages of the 5090 D.

The choice between these two cards comes down to whether the user needs the absolute top-tier performance that the 5090 D delivers or whether the 5080's still-strong performance and lower power envelope are sufficient. The benchmark results show no scenario where the 5080 outperforms the 5090 D. The 5090 D is the faster card in every measured category. The 5080 is the more restrained option, but the performance gap is consistent and often very large. For maximum performance, the data points firmly to the RTX 5090 D.