AMD Radeon Pro Vega 20 vs NVIDIA GeForce RTX 3080 Ti Mobile Comparison

The AMD Radeon Pro Vega 20 and the NVIDIA GeForce RTX 3080 Ti Mobile occupy different corners of the mobile GPU market, and the benchmark data shows a stark performance gulf between them. The NVIDIA part is a clear winner in the head-to-head tests, but the AMD part holds its own in a different performance percentile, making the choice dependent on the specific workload.

Head-to-Head Benchmarks

The direct comparison between these two GPUs is limited to two compute-oriented tests, and the results are decisive. In the Geekbench OpenCL test, the NVIDIA GeForce RTX 3080 Ti Mobile scores 117,866, while the AMD Radeon Pro Vega 20 manages 26,679. This translates to a deltaPct of -77.4% for the AMD part, meaning the NVIDIA GPU delivers over four times the raw compute performance in this particular benchmark. The gap is similarly pronounced in the Geekbench Vulkan test, where the RTX 3080 Ti Mobile scores 107,502 against the Radeon Pro Vega 20’s 26,410, a deltaPct of -75.4%. These are not marginal differences; the NVIDIA part is substantially ahead in both cross-platform compute APIs.

Looking at the broader benchmark landscape, the two GPUs sit in different performance strata. The AMD Radeon Pro Vega 20 achieves an average benchmark score of 27,839, while the NVIDIA GeForce RTX 3080 Ti Mobile posts an average of 25,740. Interestingly, despite the NVIDIA part’s massive wins in OpenCL and Vulkan, its average score is slightly lower, which suggests the AMD GPU performs better in other tests not directly compared here, or that the NVIDIA part’s average is dragged down by its less impressive results in certain DirectX or PassMark tests. The percentileVsAllGpus figures reflect this nuance: the AMD part sits at the 73rd percentile, while the NVIDIA part is at the 71st. This means the Radeon Pro Vega 20 outperforms a slightly larger fraction of all GPUs in the database, even though it loses decisively to the RTX 3080 Ti Mobile in the head-to-head tests.

The rival data further contextualizes these positions. The AMD Radeon Pro Vega 20’s nearest rival is the AMD Radeon RX 7800M, which scores an average of 27,883, a deltaPct of -0.2% from the Vega 20. This indicates the Vega 20 is essentially tied with a modern mid-range Radeon part. Other close rivals include the AMD Radeon Pro W5500X (27,973, -0.5%) and the NVIDIA GeForce GTX 980 Ti (28,020, -0.6%). Notably, the Vega 20 is actually 1% ahead of the NVIDIA GeForce RTX 3090 in average score, despite the RTX 3090 being a flagship desktop part. For the RTX 3080 Ti Mobile, the nearest rival is the AMD Radeon Pro W5700 (25,726, 0.1%), showing a near-tie, with the AMD FirePro D700 (25,842, -0.4%) and AMD Radeon RX 6700M (25,633, 0.4%) also in close proximity. This suggests the mobile RTX 3080 Ti is comparable to workstation-class AMD parts from different generations.

FAQ

Q: How much faster is the NVIDIA GeForce RTX 3080 Ti Mobile in the Geekbench OpenCL test?

A: The RTX 3080 Ti Mobile scores 117,866, which is 77.4% higher than the Radeon Pro Vega 20’s score of 26,679. This means the NVIDIA part is roughly 4.4 times faster in this specific compute benchmark.

Q: Does the AMD Radeon Pro Vega 20 outperform the RTX 3080 Ti Mobile in any benchmark?

A: In the two head-to-head tests (Geekbench OpenCL and Vulkan), the AMD part loses both. However, the AMD GPU has a higher average benchmark score (27,839 vs 25,740) and a higher percentile rank (73rd vs 71st), indicating it performs better across the full suite of tests in the database.

Q: What is the significance of the deltaPct values for the nearest rivals?

A: The deltaPct shows the percentage difference between the average score of the GPU in question and its nearest rival. For the Radeon Pro Vega 20, a -0.2% deltaPct against the RX 7800M means the two are essentially performance equals. For the RTX 3080 Ti Mobile, a 0.1% deltaPct against the Radeon Pro W5700 indicates a statistical tie in average score.

Q: Which GPU has a higher transistor count?

A: The NVIDIA GeForce RTX 3080 Ti Mobile has 22,000 million transistors, while the transistor count for the AMD Radeon Pro Vega 20 is not listed in the data. The NVIDIA part also has a die size of 496 mm², whereas the AMD die size is not provided.

Q: Are both GPUs still in production?

A: No. Both are listed as "End-of-life" in their production status. The AMD Radeon Pro Vega 20 was released on 2018-11-13, and the NVIDIA GeForce RTX 3080 Ti Mobile was released on 2022-01-24.

Q: What are the API support differences between the two?

A: The NVIDIA part supports DirectX 12 Ultimate (12_2), while the AMD part supports DirectX 12 (12_1). Both support OpenGL 4.6, but the NVIDIA part supports Vulkan 1.4, whereas the AMD part supports Vulkan 1.3.

The Verdict

The data presents a clear, if lopsided, picture. The NVIDIA GeForce RTX 3080 Ti Mobile is the definitive winner in raw compute performance, as evidenced by its 77.4% lead in OpenCL and 75.4% lead in Vulkan. Any workload that leverages these APIs will see a massive advantage with the NVIDIA part. The benchmark results indicate this is not a close contest in those specific tests.

However, the AMD Radeon Pro Vega 20 should not be dismissed. Its higher average benchmark score (27,839 vs 25,740) and superior percentile rank (73rd vs 71st) suggest it is the better-rounded performer across a wider variety of tests, even if it loses the two most common cross-platform compute benchmarks. The deltaPct against its nearest rival, the RX 7800M, is a negligible -0.2%, showing it remains competitive with newer mid-range hardware.

For users whose primary concern is compute performance in OpenCL or Vulkan, the RTX 3080 Ti Mobile is the only rational choice from this data. For those who need a GPU that performs consistently well across a broad suite of benchmarks, the Radeon Pro Vega 20 offers a compelling profile, especially considering its higher percentile ranking. The NVIDIA part is for maximizing performance in specific compute tasks, while the AMD part is for balanced, general-purpose use. The RTX 3080 Ti Mobile also has the advantage of a much larger memory pool (16 GB vs 4 GB), which could be critical for large datasets.

Specification Differences

The two GPUs differ across nearly every major specification category. The NVIDIA GeForce RTX 3080 Ti Mobile uses the GA103 chip and is built on an 8 nm process at Samsung, while the AMD Radeon Pro Vega 20 uses the Vega 12 chip and a 14 nm process at GlobalFoundries. The NVIDIA part has 22,000 million transistors on a 496 mm² die, whereas the AMD transistor count and die size are not listed.

Memory is a major differentiator: the NVIDIA part has 16 GB of GDDR6 memory on a 256-bit bus, delivering 512.0 GB/s of bandwidth. The AMD part has 4 GB of HBM2 memory on a 1024-bit bus, delivering 189.4 GB/s. The memory clock differs as well, with the NVIDIA part running at 2000 MHz (16 Gbps effective) and the AMD part at 740 MHz (1480 Mbps effective).

The core configurations are vastly different. The NVIDIA part has 7424 shading units, 232 TMUs, and 96 ROPs, while the AMD part has 1280 shading units, 80 TMUs, and 32 ROPs. The NVIDIA part also includes 58 RT cores and 232 tensor cores, which the AMD part lacks entirely. The clock speeds are similar, with the NVIDIA part boosting to 1260 MHz and the AMD part to 1283 MHz. The TDP differs, with the NVIDIA part at 115 W and the AMD part at 100 W. The bus interface also differs, with the NVIDIA part using PCIe 4.0 x16 and the AMD part using PCIe 3.0 x16.

Architecture Differences

The architectural divide is fundamental. The AMD Radeon Pro Vega 20 is based on GCN 5.0, a design that predates dedicated ray tracing and tensor cores. It uses a 14 nm process and supports DirectX 12 (12_1), Vulkan 1.3, and OpenGL 4.6. Its FP32 performance is 3.284 TFLOPS, and its FP16 performance is 6.569 TFLOPS (2:1 ratio). The shading unit count is 1280, with a pixel rate of 41.06 GPixel/s and a texture rate of 102.6 GTexel/s.

The NVIDIA GeForce RTX 3080 Ti Mobile is built on the Ampere architecture, which introduces dedicated RT cores for ray tracing and tensor cores for AI workloads. It uses an 8 nm process and supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6. Its FP32 and FP16 performance are both 18.71 TFLOPS (1:1 ratio), a significant departure from the AMD part’s 2:1 FP16 ratio. The NVIDIA part has 7424 shading units, 58 RT cores, and 232 tensor cores, with a pixel rate of 121.0 GPixel/s and a texture rate of 292.3 GTexel/s. The die size of 496 mm² is also a notable difference, reflecting the complexity of the Ampere architecture.

Where Each One Wins

The NVIDIA GeForce RTX 3080 Ti Mobile wins decisively in compute-oriented benchmarks. Its Geekbench OpenCL score of 117,866 and Vulkan score of 107,502 are over four times the AMD part’s scores. This makes it the clear choice for applications that rely heavily on these APIs, such as scientific computing, machine learning inference, and rendering tasks that use OpenCL or Vulkan. The 16 GB memory capacity and 512.0 GB/s bandwidth also position it well for large datasets and high-resolution textures.

The AMD Radeon Pro Vega 20 wins in the overall benchmark average and percentile ranking. Its average score of 27,839 is higher than the NVIDIA part’s 25,740, and its 73rd percentile rank is above the NVIDIA part’s 71st. This suggests the AMD GPU performs better in a broader range of tests, potentially including DirectX and PassMark suites where the NVIDIA part scores lower (e.g., PassMark DirectX 9 at 201 and DirectX 12 at 88). The AMD part also has a lower TDP (100 W vs 115 W), which could be advantageous in thermally constrained laptops. Its HBM2 memory, while smaller in capacity, offers a very wide 1024-bit bus, which may be beneficial for certain bandwidth-sensitive workloads that fit within the 4 GB frame buffer. The Vega 20’s release date of 2018 also suggests it may be found in older, more affordable systems, though pricing data is not available.