AMD Radeon RX 6750 XT vs NVIDIA GeForce GTX 980 Ti Comparison

The GeForce GTX 980 Ti and Radeon RX 6750 XT represent two distinct eras of GPU design, with the data showing a clear generational shift in performance. Across the four shared benchmark tests, the AMD Radeon RX 6750 XT claims three decisive victories, while the NVIDIA card manages a single, equally decisive win in one specific workload. The average benchmark scores reflect this split, with the GTX 980 Ti averaging 28,020 points and the RX 6750 XT averaging 26,011 points, though the individual test results tell a far more nuanced story than the aggregate suggests.

Head-to-Head Benchmarks

The most striking result in the head-to-head comparison is the Geekbench Metal test, where the RX 6750 XT posts a score of 137,249 against the GTX 980 Ti’s 19,520. This represents a deltaPct of -85.8% from the NVIDIA card’s perspective, meaning the AMD card is roughly 603% faster in this specific Apple-centric compute API. The magnitude of this gap is extraordinary, far exceeding any other differential in the dataset, and likely reflects architectural improvements in how the newer RDNA 2.0 design handles Metal’s compute and graphics workloads.

In the 3DMark Steel Nomad DX12 test, the RX 6750 XT again dominates, scoring 2,548 versus the GTX 980 Ti’s 1,321. The deltaPct of -48.2% indicates the AMD card is nearly twice as fast in this modern DirectX 12 gaming benchmark, a difference that aligns with the NVIDIA card’s older Maxwell architecture lacking full support for the latest DX12 feature sets. Similarly, in Geekbench Vulkan, the RX 6750 XT’s 98,089 score more than doubles the GTX 980 Ti’s 47,724, with a deltaPct of -51.3%. Vulkan is a cross-platform API that can expose raw GPU capabilities, and the data here shows the AMD card delivering over two times the performance.

The single win for the GTX 980 Ti comes in Geekbench OpenCL, where it scores 43,513 against the RX 6750 XT’s 16,360. This deltaPct of 166% means the NVIDIA card is 2.66 times faster in this compute workload. This is a peculiar result, given that the RX 6750 XT has a higher raw FP32 throughput, but it suggests that driver optimization or specific workload characteristics in the OpenCL test favor the older architecture. This win is substantial enough to skew the average benchmark scores, explaining why the GTX 980 Ti holds a higher aggregate despite losing three of four direct comparisons.

Where Each One Wins

The RX 6750 XT is the clear victor in modern graphics APIs and gaming-oriented workloads. Its wins in 3DMark Steel Nomad DX12 and Geekbench Vulkan point to a card that handles contemporary DirectX 12 and Vulkan titles with significantly greater efficiency. The massive Metal score also indicates superior performance in macOS environments or applications that leverage Metal for compute, which is a niche where the GTX 980 Ti appears severely outdated.

The GTX 980 Ti’s win in Geekbench OpenCL is its sole bright spot, suggesting that for legacy OpenCL compute tasks, the older Maxwell architecture retains a substantial advantage. This could be relevant for older scientific software, certain rendering pipelines, or any application that has not migrated to Vulkan or DirectX 12. The data shows that this advantage is not marginal; it is a 166% delta, which is a dominant result for the NVIDIA card. However, this win is isolated to one test, and the overall trend clearly favors the AMD card in any scenario that uses modern APIs.

For users prioritizing gaming performance in current titles, the RX 6750 XT is the only rational choice based on the 3DMark results. For those with legacy OpenCL compute dependencies, the GTX 980 Ti’s performance in that specific metric is unmatched by the AMD card in this comparison. The percentile rankings support this split: the GTX 980 Ti sits at the 73rd percentile among all GPUs, while the RX 6750 XT sits at the 71st percentile, indicating that despite the AMD card’s modern wins, the aggregate benchmark suite still places them in a similar overall tier.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process nodes. The GTX 980 Ti uses the GM200 chip based on Maxwell 2.0 architecture, fabricated on a 28 nm process at TSMC. This is a massive die at 601 mm², housing 8,000 million transistors, which yields a transistor density of 13.3 million per square millimeter. In contrast, the RX 6750 XT uses the Navi 22 chip based on RDNA 2.0, fabricated on a 7 nm process, also at TSMC. The AMD die is significantly smaller at 335 mm² but packs 17,200 million transistors, achieving a density of 51.3 million per square millimeter, nearly four times the density of the NVIDIA chip.

The core configurations differ substantially. The GTX 980 Ti has 2,816 shading units, 176 texture mapping units, and 96 raster operation pipelines. The RX 6750 XT has 2,560 shading units, 160 TMUs, and only 64 ROPs. Despite having fewer units across the board, the AMD card achieves higher fill rates: 166.4 GPixel/s versus 103.3 GPixel/s for pixel rate, and 416.0 GTexel/s versus 189.4 GTexel/s for texture rate. The FP32 performance also favors AMD, with 13.31 TFLOPS against 6.060 TFLOPS, more than double the compute throughput.

The RX 6750 XT also introduces hardware features absent from the GTX 980 Ti, specifically 40 ray tracing cores. The Maxwell architecture has no RT cores, while RDNA 2.0 includes them for hardware-accelerated ray tracing. The AMD card also supports FP16 compute at 26.62 TFLOPS with a 2:1 ratio, whereas the GTX 980 Ti has no listed FP16 capability. The NVIDIA card supports DirectX 12 (12_1), while the AMD card supports DirectX 12 Ultimate (12_2), indicating support for newer DX12 features like ray tracing and variable rate shading.

Specification Differences

The memory subsystems represent a major divergence. The GTX 980 Ti comes with 6 GB of GDDR5 memory on a 384-bit bus, delivering 336.6 GB/s of bandwidth at an effective 7 Gbps. The RX 6750 XT doubles the capacity to 12 GB of GDDR6 memory on a narrower 192-bit bus, yet achieves higher bandwidth at 432.0 GB/s due to the faster 18 Gbps effective memory speed. This means the AMD card offers both more capacity and higher bandwidth, which is critical for modern games with large texture pools.

Clock speeds show the generational leap in efficiency. The GTX 980 Ti has a base clock of 1000 MHz and a boost of 1076 MHz, while the RX 6750 XT runs at a 2150 MHz base and 2600 MHz boost, with a game clock of 2495 MHz. The AMD card’s much higher clocks are enabled by the smaller 7 nm process node, which allows higher frequencies at similar power levels. Both cards have a TDP of 250 W and require the same power connectors (1x 6-pin + 1x 8-pin) and a 600 W suggested PSU.

The bus interface differs, with the GTX 980 Ti using PCIe 3.0 x16 and the RX 6750 XT using PCIe 4.0 x16. Display outputs also vary: the NVIDIA card offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2, while the AMD card provides 1x HDMI 2.1 and 3x DisplayPort 1.4a, supporting newer display standards. Physical dimensions are nearly identical, with both cards at 267 mm length, 40 mm width, and a dual-slot form factor, with only a 1 mm difference in height (111 mm vs 110 mm). The release dates span nearly seven years, with the GTX 980 Ti launching in June 2015 at a launch MSRP of 649 USD, and the RX 6750 XT launching in March 2022 at a launch MSRP of 549 USD.

FAQ

Q: Which card has a higher average benchmark score?

A: The NVIDIA GeForce GTX 980 Ti has an average benchmark score of 28,020, which is higher than the AMD Radeon RX 6750 XT’s 26,011, despite losing three of four head-to-head tests.

Q: How much faster is the RX 6750 XT in 3DMark Steel Nomad DX12?

A: The RX 6750 XT scores 2,548 in this test, while the GTX 980 Ti scores 1,321, resulting in a deltaPct of -48.2% for the NVIDIA card, meaning the AMD card is approximately 93% faster.

Q: Does the GTX 980 Ti win any benchmark against the RX 6750 XT?

A: Yes, the GTX 980 Ti wins Geekbench OpenCL with a score of 43,513 against the RX 6750 XT’s 16,360, a deltaPct of 166%, making it roughly 2.66 times faster in that workload.

Q: What is the difference in memory capacity and bandwidth?

A: The GTX 980 Ti has 6 GB of GDDR5 memory with 336.6 GB/s bandwidth, while the RX 6750 XT has 12 GB of GDDR6 memory with 432.0 GB/s bandwidth, giving the AMD card both double the capacity and higher throughput.

Q: Which card supports ray tracing hardware?

A: Only the AMD Radeon RX 6750 XT has 40 ray tracing cores; the NVIDIA GeForce GTX 980 Ti has no RT cores listed, as its Maxwell 2.0 architecture predates hardware ray tracing support.

Q: How do the transistor densities compare?

A: The GTX 980 Ti has a transistor density of 13.3 million per square millimeter on a 28 nm process, while the RX 6750 XT achieves 51.3 million per square millimeter on a 7 nm process, a nearly four-fold difference.

The Verdict

The data presents a clear split based on workload. The AMD Radeon RX 6750 XT is the superior card for modern gaming and contemporary compute APIs, as evidenced by its dominant wins in 3DMark Steel Nomad DX12, Geekbench Vulkan, and Geekbench Metal. Its 12 GB memory capacity, higher bandwidth, double the FP32 throughput, and ray tracing support make it the appropriate choice for anyone running current games or applications that leverage DirectX 12 Ultimate, Vulkan, or Metal. The 3DMark result alone, where it nearly doubles the GTX 980 Ti’s score, should steer gamers toward the AMD card.

The NVIDIA GeForce GTX 980 Ti, however, is not without merit in this comparison. Its 166% victory in Geekbench OpenCL indicates that for legacy compute workloads relying on that API, the older Maxwell architecture is vastly superior. Additionally, its higher average benchmark score and 73rd percentile ranking, versus the RX 6750 XT’s 71st percentile, show that in an aggregate sense, the NVIDIA card still holds its own in mixed workloads. The GTX 980 Ti also has a higher launch MSRP of 649 USD compared to the RX 6750 XT’s 549 USD, though both cards are end-of-life products.

For a user building a system for modern gaming or new content creation, the RX 6750 XT is the only logical pick based on the benchmark data. For a user with specific OpenCL compute requirements, the GTX 980 Ti offers a unique advantage that the AMD card cannot match. The data does not support a universal winner; instead, it identifies each card’s strengths in distinct performance domains, with the RX 6750 XT being the more future-proof and versatile option overall.