NVIDIA GeForce RTX 3080 Ti vs NVIDIA GeForce RTX 5090 Mobile Comparison

The NVIDIA GeForce RTX 5090 Mobile and NVIDIA GeForce RTX 3080 Ti represent two distinct generations of laptop and desktop graphics, respectively. The benchmark data reveals a clear generational shift, with the newer 5090 Mobile winning 7 out of 10 head-to-head tests, yet the older 3080 Ti still holds specific advantages. This analysis breaks down where each GPU excels, based strictly on the provided benchmark scores and specifications.

Where Each One Wins

The RTX 5090 Mobile establishes dominance in modern, graphics-intensive workloads. Its largest victory comes in the Passmark DirectX 12 test, where it scores 138 versus the 3080 Ti's 110, a 25.5% advantage. This suggests the Blackwell architecture's optimizations for contemporary API features translate into substantial real-world performance gains. Similarly, in Passmark DirectX 11, the 5090 Mobile leads with 269 points against 223, a 20.6% margin, indicating strong performance in the most common gaming API. The 3DMark Steel Nomad DX12 benchmark reinforces this trend, with the 5090 Mobile posting 5871 points compared to 5077 for the 3080 Ti, a 15.6% lead that points to superior handling of modern game engines.

The 5090 Mobile also excels in compute-heavy scenarios that leverage its newer architecture. In Geekbench OpenCL, it scores 201834 versus 170037, an 18.7% advantage, and in Geekbench Vulkan, it maintains a lead with 198405 points against 192697, a 3% margin. The Passmark G3D score of 30034 for the 5090 Mobile versus 26896 for the 3080 Ti (11.7% higher) further confirms its overall graphical superiority. Even in legacy DirectX 9 workloads, the 5090 Mobile wins decisively with 324 points against 274, an 18.2% improvement, showing that its advantages extend backward through older API generations.

Conversely, the RTX 3080 Ti carves out wins in specific, narrower scenarios. Its most significant victory is in Passmark GPU Compute, where it scores 15282 against the 5090 Mobile's 13401, a 12.3% advantage. This is a notable result, as it suggests the older card's raw compute throughput, derived from its higher FP32 rating of 34.10 TFLOPS, remains competitive or superior in certain non-gaming tasks. The 3080 Ti also edges out the 5090 Mobile in Passmark G2D, scoring 1091 versus 1057, a 3.1% margin, indicating better performance in 2D desktop and application rendering. Its final win is marginal, taking the Passmark DirectX 10 test with 184 points against 183, a 0.5% difference that is essentially a statistical tie.

The Verdict

The data paints a clear picture for gamers and users of modern graphics APIs: the NVIDIA GeForce RTX 5090 Mobile is the superior choice. Its wins in DirectX 11, DirectX 12, and 3DMark Steel Nomad, coupled with a higher Passmark G3D score, demonstrate that it is significantly faster for current and upcoming game titles. The 25.5% lead in DirectX 12 is particularly compelling, as this API is the standard for new AAA releases. For anyone building a high-performance laptop for gaming, the 5090 Mobile's data-driven advantages make it the logical pick.

However, the RTX 3080 Ti is not without its merits. Its decisive 12.3% win in Passmark GPU Compute suggests it may be better suited for specific professional or compute-oriented workloads that are not primarily graphics-focused. Its higher FP32 performance of 34.10 TFLOPS versus 31.80 TFLOPS for the 5090 Mobile supports this, indicating raw number-crunching power. For users with a desktop system who prioritize compute tasks and do not require the latest features, the 3080 Ti remains a viable option, even though it is end-of-life.

The overarching verdict is that the 5090 Mobile is the better all-around graphics processor, winning the majority of benchmarks and posting a higher average benchmark score of 45152 compared to 41187 for the 3080 Ti. Its 84th percentile ranking versus the 3080 Ti's 83rd percentile further solidifies its position as a top-tier performer. The choice ultimately depends on the primary use case, with the 5090 Mobile for graphics and the 3080 Ti for specific compute scenarios.

Head-to-Head Benchmarks

The most significant win for the RTX 5090 Mobile is in the Passmark DirectX 12 test. Here, it scores 138 points, a 25.5% improvement over the 3080 Ti's 110 points. This is a massive generational leap and indicates that the 5090 Mobile's architecture is far more efficient at processing the complex draw calls and features present in DirectX 12 titles. This result is the strongest argument for the newer card.

Following closely is the Passmark DirectX 11 test, where the 5090 Mobile's 269 points eclipse the 3080 Ti's 223 points by 20.6%. This win is crucial because DirectX 11 remains the baseline for a vast library of games. The 5090 Mobile's ability to outperform the 3080 Ti by a fifth in this workload demonstrates that its advantages are not limited to bleeding-edge APIs.

In the 3DMark Steel Nomad DX12 test, the 5090 Mobile scores 5871, which is 15.6% higher than the 3080 Ti's 5077. This is a synthetic benchmark designed to stress modern GPUs, and the result confirms that the 5090 Mobile has a substantial edge in sustained, demanding rendering scenarios. The Geekbench OpenCL result is another decisive victory, with the 5090 Mobile scoring 201834 against 170037, an 18.7% lead that highlights its superior compute potential in a cross-platform API.

The RTX 3080 Ti's biggest win is in Passmark GPU Compute, where it scores 15282, outperforming the 5090 Mobile's 13401 by 12.3%. This is a counter-intuitive result given the newer architecture, but it shows that the 3080 Ti's higher peak FP32 throughput can be leveraged in specific compute-heavy tasks. The 3080 Ti also wins the Passmark G2D test with 1091 points versus 1057, a 3.1% margin that suggests better performance in 2D environments, though the real-world impact is minimal.

FAQ

Q: Which GPU is faster in DirectX 12 games?

A: The NVIDIA GeForce RTX 5090 Mobile is significantly faster, scoring 138 in Passmark DirectX 12 compared to the 3080 Ti's 110, a 25.5% advantage.

Q: Does the older RTX 3080 Ti beat the new RTX 5090 Mobile in any benchmark?

A: Yes, the RTX 3080 Ti wins three tests: Passmark GPU Compute (15282 vs 13401), Passmark G2D (1091 vs 1057), and Passmark DirectX 10 (184 vs 183).

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

A: The RTX 5090 Mobile scores 5871, which is 15.6% higher than the RTX 3080 Ti's 5077, indicating a clear lead in modern, high-fidelity rendering.

Q: How do the two GPUs compare in overall average performance?

A: The RTX 5090 Mobile has a higher average benchmark score of 45152, while the RTX 3080 Ti averages 41187, showing a broad performance advantage for the newer card.

Q: Which GPU has a higher raw compute score in OpenCL?

A: The RTX 5090 Mobile leads in Geekbench OpenCL with a score of 201834, which is 18.7% higher than the RTX 3080 Ti's 170037.

Q: Based on the data, which GPU is better for compute tasks?

A: The RTX 3080 Ti wins the Passmark GPU Compute test by 12.3%, suggesting it may be better for certain non-graphics compute workloads, despite its lower overall average score.

Architecture Differences

The two GPUs are built on fundamentally different architectures and manufacturing processes. The RTX 5090 Mobile uses the GB203 chip based on the Blackwell 2.0 architecture, fabricated on a 5 nm process by TSMC. In contrast, the RTX 3080 Ti uses the GA102 chip based on the older Ampere architecture, fabricated on an 8 nm process by Samsung. This process difference is a primary driver of the performance gap, with the 5090 Mobile's 5 nm node allowing for a transistor density of 120.6M / mm², far exceeding the 3080 Ti's 45.1M / mm².

The core configurations also differ. The 5090 Mobile has 10496 shading units, 328 TMUs, and 112 ROPs, while the 3080 Ti has 10240 shading units, 320 TMUs, and 112 ROPs. The newer card also has 82 RT cores and 328 tensor cores, compared to 80 RT cores and 320 tensor cores on the 3080 Ti. These differences, while seemingly small, contribute to the 5090 Mobile's superior benchmark performance in graphics workloads. The memory subsystems are distinct as well, with the 5090 Mobile using 24 GB of GDDR7 on a 256-bit bus, while the 3080 Ti uses 12 GB of GDDR6X on a wider 384-bit bus.

The bus interface has also advanced. The 5090 Mobile supports PCIe 5.0 x16, while the 3080 Ti uses PCIe 4.0 x16. This newer interface provides higher bandwidth for data transfer between the GPU and the rest of the system. The 5090 Mobile's TDP of 95 W is dramatically lower than the 3080 Ti's 350 W, highlighting a major efficiency improvement with the Blackwell architecture. The 3080 Ti is a dual-slot card with a 12-pin power connector and a suggested PSU of 750 W, whereas the 5090 Mobile is an IGP with no power connectors, designed for portable devices.

Specification Differences

The most obvious specification difference is memory capacity and type. The RTX 5090 Mobile has 24 GB of GDDR7 memory, while the RTX 3080 Ti has 12 GB of GDDR6X. Despite the larger capacity, the 5090 Mobile's memory bandwidth is 896.0 GB/s, slightly lower than the 3080 Ti's 912.4 GB/s, due to the narrower 256-bit bus compared to the 384-bit bus. The memory clocks also differ, with the 5090 Mobile running at 1750 MHz (28 Gbps effective) and the 3080 Ti at 1188 MHz (19 Gbps effective).

Clock speeds and power are also significantly different. The RTX 5090 Mobile has a base clock of 990 MHz and a boost clock of 1515 MHz, while the RTX 3080 Ti has a higher base clock of 1365 MHz and a boost clock of 1665 MHz. This means the 3080 Ti runs at higher frequencies, but the 5090 Mobile's efficiency comes from its newer architecture. The TDP is a stark contrast: 95 W for the 5090 Mobile versus 350 W for the 3080 Ti, a 255 W difference. The 3080 Ti also has a specified length of 285 mm, a height of 112 mm, and a width of 40 mm, while the 5090 Mobile has no listed dimensions as it is an integrated part of a laptop.

The release dates and production statuses are also noteworthy. The RTX 5090 Mobile was released on 2025-03-26 and is currently Active in production, while the RTX 3080 Ti was released on 2021-05-30 and is End-of-life. The 3080 Ti has a launch MSRP of 1,199 USD, while the 5090 Mobile has no listed MSRP as it is a mobile component. Finally, the display outputs differ, with the 5090 Mobile being "Portable Device Dependent" and the 3080 Ti featuring 1x HDMI 2.1 and 3x DisplayPort 1.4a.