NVIDIA GeForce GTX 980 Ti vs NVIDIA GeForce RTX 3070 Ti Comparison

The NVIDIA GeForce RTX 3070 Ti and the NVIDIA GeForce GTX 980 Ti represent two distinct eras of GPU design, separated by a generational gap that is starkly visible in the benchmark data. The data shows a decisive performance shift, with the newer Ampere-based card delivering results that are often more than double those of the older Maxwell flagship. This analysis will quantify that gap across the available tests, explore the architectural reasons behind it, and determine which card suits which workload based purely on the facts.

Head-to-Head Benchmarks

The head-to-head results are unambiguous, with the RTX 3070 Ti winning all three shared tests by a massive margin. In the 3DMark Steel Nomad DX12 test, the RTX 3070 Ti scores 3478 points against the GTX 980 Ti’s 1321 points. This translates to a delta of 163.3% — meaning the newer card is over one and a half times faster in this modern DirectX 12 rasterization workload. The gap here is not incremental; it represents a complete paradigm shift in performance capability.

The compute-oriented benchmarks reveal an even wider chasm. In Geekbench OpenCL, the RTX 3070 Ti posts a score of 119718, while the GTX 980 Ti manages only 43513. The deltaPct of 175.1% indicates that the Ampere card delivers nearly three times the raw compute throughput. This is a clear signal that the RTX 3070 Ti is not just a gaming card with a refresh; its compute resources are on an entirely different level. The Geekbench Vulkan result is the most lopsided of all, with the RTX 3070 Ti scoring 139541 versus the GTX 980 Ti’s 47724, a delta of 192.4%. Vulkan is a low-level API that scales well with hardware parallelism, and the data here suggests the RTX 3070 Ti has a profound advantage in parallel execution.

The per-test deltas are telling. The smallest victory for the RTX 3070 Ti is in the DX12 test (163.3%), while the largest is in Vulkan (192.4%). This pattern implies that the advantage grows as the API becomes more efficient at utilizing the underlying hardware. The GTX 980 Ti, with its older Maxwell architecture, simply cannot feed its execution units as effectively in these modern workloads. The overall average benchmark score reinforces the narrative: the RTX 3070 Ti sits at 29945, while the GTX 980 Ti is at 28020. The 3DMark result alone accounts for a significant portion of that difference, but the compute scores push the average farther apart.

The Verdict

The data is unequivocal: the NVIDIA GeForce RTX 3070 Ti is the superior performer in every measured category. If the choice is between these two cards for any modern workload, the RTX 3070 Ti is the only rational pick from a performance standpoint. It wins 100% of the head-to-head tests (3 out of 3), with deltas ranging from 163.3% to 192.4%. The GTX 980 Ti’s 0 wins in the head-to-head comparison are a stark summary of its competitive position.

The RTX 3070 Ti also holds a higher percentile rank among all GPUs, sitting at the 75th percentile versus the GTX 980 Ti’s 73rd. While both are near the top of the historical performance curve, the newer card is positioned more favorably. The RTX 3070 Ti’s nearest rival is the NVIDIA GeForce RTX 5070 Mobile, with a deltaPct of just 0.1%, meaning they are essentially tied in average score. The GTX 980 Ti’s closest competitor is the AMD Radeon Pro W5500X, a workstation card, with a deltaPct of 0.2%. This placement underscores that the GTX 980 Ti’s peers are older or specialized hardware, while the RTX 3070 Ti is competing with modern mobile and desktop offerings.

For a user prioritizing raw frame rates in new games or engaging in GPU-accelerated compute, the RTX 3070 Ti is the definitive choice. The GTX 980 Ti should only be considered if it is the only available option, as its performance is demonstrably inferior in every test. Its legacy is that of a capable card from its era, but the data shows it has been thoroughly outclassed.

FAQ

Q: How much faster is the RTX 3070 Ti in the 3DMark Steel Nomad DX12 test?

A: The RTX 3070 Ti scores 3478 points compared to the GTX 980 Ti’s 1321 points. This represents a performance lead of 163.3% for the newer card.

Q: Is the RTX 3070 Ti a better choice for compute workloads?

A: Yes. In Geekbench OpenCL, the RTX 3070 Ti scores 119718 versus 43513 for the GTX 980 Ti, a delta of 175.1%. The Vulkan test shows an even larger gap, with the RTX 3070 Ti leading by 192.4%.

Q: Did the GTX 980 Ti win any benchmark in the head-to-head comparison?

A: No. The RTX 3070 Ti won all three shared tests: 3DMark Steel Nomad DX12, Geekbench OpenCL, and Geekbench Vulkan. The GTX 980 Ti recorded zero wins.

Q: How do their average benchmark scores compare?

A: The RTX 3070 Ti has an average benchmark score of 29945, while the GTX 980 Ti’s average is 28020. This difference is driven by the newer card’s massive wins in the compute tests.

Q: What is the percentile ranking difference between the two cards?

A: The RTX 3070 Ti is in the 75th percentile of all GPUs, while the GTX 980 Ti is in the 73rd percentile. Both are high-ranked, but the newer card holds a slight edge in overall standing.

Q: Which card has a higher launch MSRP?

A: The GTX 980 Ti had a launch MSRP of 649 USD, which is higher than the RTX 3070 Ti’s launch MSRP of 599 USD. The newer card is both cheaper at launch and significantly faster.

Specification Differences

The two cards differ in nearly every core specification. The RTX 3070 Ti has 8 GB of GDDR6X memory on a 256-bit bus, delivering a bandwidth of 608.3 GB/s. The GTX 980 Ti has 6 GB of GDDR5 memory on a wider 384-bit bus, but its bandwidth is only 336.6 GB/s. This means the newer card has more memory, faster memory, and higher bandwidth despite a narrower bus.

The shader configuration is also drastically different. The RTX 3070 Ti packs 6144 shading units, 192 TMUs, and 96 ROPs. In contrast, the GTX 980 Ti has 2816 shading units, 176 TMUs, and 96 ROPs. The RTX 3070 Ti has more than double the shading units and more TMUs, while ROP count is identical. Clock speeds favor the newer card as well, with a base clock of 1575 MHz and boost of 1770 MHz versus the GTX 980 Ti’s 1000 MHz base and 1076 MHz boost.

The power characteristics differ too. The RTX 3070 Ti has a TDP of 290 W and uses a single 12-pin power connector, while the GTX 980 Ti has a TDP of 250 W and uses a 6-pin plus 8-pin configuration. Both cards suggest a 600 W power supply. The bus interface is another differentiator: the RTX 3070 Ti uses PCIe 4.0 x16, while the GTX 980 Ti uses PCIe 3.0 x16. Display outputs also differ, with the newer card offering 1x HDMI 2.1 and 3x DisplayPort 1.4a, while the older card has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2.

Architecture Differences

The architectural gap is fundamental. The RTX 3070 Ti is built on the Ampere architecture using the GA104 chip, fabricated on an 8 nm process at Samsung. The GTX 980 Ti uses the Maxwell 2.0 architecture with the GM200 chip, built on a 28 nm process at TSMC. This process node difference is massive, allowing the RTX 3070 Ti to pack 17,400 million transistors into a 392 mm² die, achieving a transistor density of 44.4M per mm². The GTX 980 Ti, by comparison, has only 8,000 million transistors on a much larger 601 mm² die, with a density of just 13.3M per mm².

The most significant functional difference is the presence of specialized hardware. The RTX 3070 Ti includes 48 RT cores for ray tracing and 192 tensor cores for AI-accelerated workloads. The GTX 980 Ti has no such hardware, with null values for both RT and tensor cores. This means the newer card can accelerate ray-traced effects and tensor-based features like DLSS, while the older card cannot. The FP32 compute performance reflects this: the RTX 3070 Ti delivers 21.75 TFLOPS, while the GTX 980 Ti delivers only 6.060 TFLOPS. The RTX 3070 Ti also supports FP16 at a 1:1 ratio with FP32, while the GTX 980 Ti has no FP16 data available.

The API support also differs. The RTX 3070 Ti supports DirectX 12 Ultimate (12_2), which includes features like mesh shaders and variable rate shading. The GTX 980 Ti is limited to DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4, but the newer card's hardware is designed for the more advanced DX12 Ultimate feature set. The memory type and bandwidth, already noted, are a direct consequence of the architectural generation: GDDR6X versus GDDR5.

Where Each One Wins

The RTX 3070 Ti wins in every benchmarked category, making its use-case list straightforward. It is the superior choice for any modern DirectX 12 gaming, as evidenced by its 163.3% lead in 3DMark Steel Nomad. It is also the clear winner for compute-heavy tasks, with a 175.1% lead in OpenCL and a 192.4% lead in Vulkan. Any workload that can leverage Vulkan's low-level parallelism will see the largest benefit. Additionally, the presence of RT and tensor cores means the RTX 3070 Ti can handle ray tracing and AI-based features that the GTX 980 Ti simply cannot process.

The GTX 980 Ti has no benchmark wins to claim. Its only potential advantage lies in its legacy status. With 6 GB of memory and a 384-bit bus, it was a high-end card for its time, but the data shows it is not competitive with the RTX 3070 Ti. Its higher launch MSRP of 649 USD compared to the RTX 3070 Ti's 599 USD further cements the newer card's superiority in both price and performance. The GTX 980 Ti's nearest rivals are workstation and older mobile GPUs like the AMD Radeon Pro W5500X and AMD Radeon Pro Vega 20, indicating its performance tier is now relegated to legacy or niche applications. There is no workload in the provided data where the GTX 980 Ti comes out ahead.