AMD Radeon Pro Vega 20 vs NVIDIA GeForce GTX 980 Ti Comparison

The NVIDIA GeForce GTX 980 Ti and AMD Radeon Pro Vega 20 are two end-of-life GPUs from different eras, with the former launching in the GeForce 900 generation and the latter in the Radeon Pro Mac Vega Series. Their average benchmark scores are remarkably close, with the GTX 980 Ti posting 28020 and the Pro Vega 20 posting 27839, a difference of just 0.7%. This places both cards at the 73rd percentile among all GPUs, indicating they occupy nearly identical performance tiers despite their architectural and generational differences.

Head-to-Head Benchmarks

The head-to-head results reveal a stark split in API-specific performance. In the Geekbench Metal test, the AMD Radeon Pro Vega 20 decisively wins with a score of 30427 against the GTX 980 Ti's 19520. This represents a 35.8% advantage for the AMD card, a substantial margin that underscores its strength in Apple's Metal API, which is consistent with its positioning in the Radeon Pro Mac lineup. The data shows that in Metal-centric workloads, the Pro Vega 20 is clearly the superior option.

However, the tables turn dramatically in the other two benchmark tests. In Geekbench OpenCL, the NVIDIA GeForce GTX 980 Ti delivers a commanding performance, scoring 43513 compared to the Pro Vega 20's 26679. This translates to a 63.1% lead for NVIDIA, demonstrating a massive advantage in OpenCL compute workloads. The gap widens even further in the Geekbench Vulkan test, where the GTX 980 Ti scores 47724 against the Pro Vega 20's 26410, an 80.7% difference. This near-doubling of performance in Vulkan indicates that the GTX 980 Ti has significantly stronger driver optimization and hardware efficiency for this modern graphics API.

The aggregate picture from these three tests shows the GTX 980 Ti winning two out of three head-to-head matchups. Its victories in OpenCL and Vulkan are so lopsided that they more than compensate for the Metal loss, which is reflected in the overall average benchmark scores. The GTX 980 Ti's average of 28020 edges out the Pro Vega 20's 27839, confirming that NVIDIA holds a slight overall edge in this comparison. The delta percentages in the nearestRivals data reinforce this, listing the GTX 980 Ti as 0.7% ahead of the Pro Vega 20, while the Pro Vega 20 sits 0.6% behind the GTX 980 Ti.

Where Each One Wins

The benchmark results clearly delineate distinct use-case scenarios for each card. The AMD Radeon Pro Vega 20 is the clear winner in Metal-based environments. Its 35.8% lead in the Geekbench Metal test makes it the preferable choice for macOS users or professionals working within Apple's ecosystem where Metal is the primary graphics and compute API. This aligns with its "Radeon Pro Mac" generation designation, suggesting it was engineered for seamless integration in that platform. For tasks like Metal-accelerated rendering, video processing, or any workflow that leverages this API, the data shows the Pro Vega 20 is the superior performer.

Conversely, the NVIDIA GeForce GTX 980 Ti dominates in OpenCL and Vulkan workloads. The 63.1% advantage in OpenCL makes it a powerhouse for general-purpose GPU compute tasks, such as scientific simulations, data processing, or any application that utilizes OpenCL for acceleration. Its 80.7% lead in Vulkan is even more pronounced, making it the definitive choice for Vulkan-based gaming, real-time 3D rendering, or any modern graphics application that uses this low-overhead API. The GTX 980 Ti's architecture appears to be exceptionally well-suited for these cross-platform APIs, delivering performance that far exceeds its newer rival.

In terms of raw compute throughput, the specifications corroborate these benchmark findings. The GTX 980 Ti offers 6.060 TFLOPS of FP32 performance, while the Pro Vega 20 provides 3.284 TFLOPS, a differential that explains the NVIDIA card's dominance in compute-heavy OpenCL tests. The Pro Vega 20 does have a significant advantage in FP16 performance, delivering 6.569 TFLOPS with a 2:1 ratio, but the benchmark data does not show a test that specifically leverages this capability, so its practical impact in this comparison remains unquantified.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process technologies. The NVIDIA GeForce GTX 980 Ti uses the GM200 chip based on the Maxwell 2.0 architecture, manufactured on a 28 nm process at TSMC. This chip contains 8,000 million transistors on a 601 mm² die, yielding a transistor density of 13.3M per mm². In contrast, the AMD Radeon Pro Vega 20 uses the Vega 12 chip based on the GCN 5.0 architecture, manufactured on a 14 nm process at GlobalFoundries. The fact pack does not list transistor count or die size for the Vega 12, but the process node difference is notable, with AMD using a smaller 14 nm process compared to NVIDIA's 28 nm.

The memory subsystems are also starkly different. The GTX 980 Ti features 6 GB of GDDR5 memory on a 384-bit bus, delivering a bandwidth of 336.6 GB/s. The Pro Vega 20, by contrast, has 4 GB of HBM2 memory on a much wider 1024-bit bus, but with a lower bandwidth of 189.4 GB/s due to its lower memory clock of 740 MHz. This means the NVIDIA card has significantly higher memory bandwidth, which likely contributes to its superior performance in memory-intensive benchmark tests. The GTX 980 Ti also has a substantially larger memory capacity, which can be advantageous for large datasets or high-resolution textures.

Compute resources differ considerably as well. The GTX 980 Ti has 2816 shading units, 176 texture mapping units, and 96 ROPs, while the Pro Vega 20 has 1280 shading units, 80 TMUs, and 32 ROPs. The pixel rate of the GTX 980 Ti is 103.3 GPixel/s compared to the Pro Vega 20's 41.06 GPixel/s, and the texture rate is 189.4 GTexel/s versus 102.6 GTexel/s. These numbers indicate that the GTX 980 Ti has roughly double the raw compute resources in most categories, which is consistent with its higher FP32 throughput and its dominant performance in OpenCL and Vulkan. Both cards support DirectX 12 (12_1) and OpenGL 4.6, but the GTX 980 Ti supports Vulkan 1.4 while the Pro Vega 20 supports Vulkan 1.3.

Power and physical characteristics also diverge. The GTX 980 Ti has a TDP of 250 W and requires a 600 W power supply, using a dual-slot design with 1x 6-pin and 1x 8-pin power connectors. It measures 267 mm in length, 111 mm in height, and 40 mm in width. The Pro Vega 20 is far more power-efficient with a TDP of 100 W and an IGP (integrated graphics processor) slot width, with dimensions listed as "Portable Device Dependent." This suggests the Pro Vega 20 is designed for mobile or compact systems, while the GTX 980 Ti is a full-size desktop card.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 980 Ti has a slightly higher average benchmark score of 28020, compared to the AMD Radeon Pro Vega 20's 27839, a 0.7% difference.

Q: How does the GTX 980 Ti perform in Vulkan compared to the Pro Vega 20?

A: The GTX 980 Ti scores 47724 in the Geekbench Vulkan test, which is 80.7% higher than the Pro Vega 20's score of 26410, giving NVIDIA a massive advantage in this API.

Q: What is the memory configuration difference between the two cards?

A: The GTX 980 Ti uses 6 GB of GDDR5 memory on a 384-bit bus with 336.6 GB/s bandwidth, while the Pro Vega 20 uses 4 GB of HBM2 memory on a 1024-bit bus with 189.4 GB/s bandwidth.

Q: In which benchmark does the AMD Radeon Pro Vega 20 outperform the GTX 980 Ti?

A: The Pro Vega 20 wins in the Geekbench Metal test, scoring 30427 against the GTX 980 Ti's 19520, a 35.8% advantage for AMD.

Q: What are the process nodes for each GPU?

A: The GTX 980 Ti is manufactured on a 28 nm process at TSMC, while the Pro Vega 20 is manufactured on a 14 nm process at GlobalFoundries.

Q: Which card has higher FP32 compute performance?

A: The GTX 980 Ti offers 6.060 TFLOPS of FP32 performance, nearly double the Pro Vega 20's 3.284 TFLOPS, which aligns with its superior OpenCL benchmark results.

The Verdict

The data presents a clear picture for different user profiles. The AMD Radeon Pro Vega 20 is the choice for users whose primary workloads run on the Metal API. Its 35.8% lead in Geekbench Metal, combined with its 100 W TDP and IGP form factor, makes it an ideal solution for compact systems or Mac environments where Metal is the standard interface. It offers lower power consumption and is designed for portable or integrated use cases, as indicated by its "Portable Device Dependent" dimensions.

The NVIDIA GeForce GTX 980 Ti is the superior option for general compute and modern graphics APIs. Its 63.1% lead in OpenCL and 80.7% lead in Vulkan demonstrate overwhelming dominance in these cross-platform workloads. For users running OpenCL-based compute tasks or Vulkan-based games and applications, the GTX 980 Ti delivers more than double the performance in Vulkan scenarios. Its larger 6 GB memory capacity and higher bandwidth of 336.6 GB/s also provide practical advantages for handling larger assets and data sets.

The aggregate benchmark scores show the GTX 980 Ti holds a slight overall advantage, with an average score of 28020 versus 27839. Both cards sit at the 73rd percentile among all GPUs, and their nearestRivals lists show they compete directly with modern cards like the AMD Radeon RX 7800M and the NVIDIA GeForce RTX 3090 in average score terms. The GTX 980 Ti's higher TDP of 250 W is a trade-off for its performance, requiring a 600 W power supply and dual-slot cooling, whereas the Pro Vega 20's 100 W TDP makes it significantly more power-efficient. Ultimately, the choice hinges on the target platform and API: pick the Pro Vega 20 for Metal-centric workflows and the GTX 980 Ti for everything else.