NVIDIA GeForce RTX 2080 Ti

The NVIDIA GeForce RTX 2080 Ti is the flagship of the GeForce 20-series, built on the Turing architecture with a 12 nm process from TSMC. It packs 18,600 million transistors on a 754 mm² die, with 4,352 shading units, 272 texture mapping units, and 88 ROPs. The card operates at a base clock of 1350 MHz and a boost clock of 1545 MHz, delivering 13.45 TFLOPS of FP32 performance and 26.90 TFLOPS of FP16 (2:1). Its average benchmark score across the database is 30,459, placing it in the 74th percentile of all GPUs tested. The RTX 2080 Ti carries a launch MSRP of 999 USD.

How It Compares

The RTX 2080 Ti’s average benchmark score of 30,459 is effectively identical to the AMD Radeon RX 590, which scores 30,457 — a delta of 0%. This indicates that, in the aggregate of all benchmark workloads, the two cards are statistically indistinguishable, despite the architectural differences between Turing and Polaris.

Against the AMD Radeon RX 590 GME, the RTX 2080 Ti trails by 1% (RX 590 GME average score 30,755). This is a narrow margin, suggesting that the GME variant holds a slight edge in the overall benchmark suite, though the difference is within typical run-to-run variance.

The RTX 2080 Ti is 1% ahead of the AMD Radeon RX 570, which posts an average score of 30,158. While the lead is small, it confirms that the 2080 Ti outperforms the older RX 570 across the mixed workload set.

Finally, the RTX 2080 Ti leads the NVIDIA GeForce RTX 5060 by 1.2% (RTX 5060 average score 30,109). This is the largest delta among the four nearest rivals, though still a modest advantage. The comparison is notable because the RTX 5060 is a much newer product, yet the 2080 Ti maintains a slight edge in aggregate performance.

Ray Tracing and Feature Set

The RTX 2080 Ti is equipped with 68 dedicated RT cores and 544 tensor cores, marking NVIDIA’s first mainstream implementation of hardware-accelerated ray tracing. These RT cores offload bounding volume hierarchy traversal and ray intersection calculations, enabling real-time ray-traced effects in supported titles. The tensor cores are designed for AI-accelerated workloads, including neural network inference and deep learning-based features, though specific software implementations are not enumerated in the data.

In terms of API support, the card is rated for DirectX 12 Ultimate (feature level 12_2), which includes support for hardware ray tracing, variable rate shading, and mesh shaders. It also supports OpenGL 4.6 and Vulkan 1.4, ensuring broad compatibility with modern graphics APIs. The combination of RT and tensor cores, along with DirectX 12 Ultimate, positions the 2080 Ti as a forward-looking product for its generation, even though it is now end-of-life.

Memory Subsystem

The RTX 2080 Ti is equipped with 11 GB of GDDR6 memory on a 352-bit bus, yielding a peak bandwidth of 616.0 GB/s. The memory operates at 1750 MHz (14 Gbps effective). This configuration provides a substantial memory pool for high-resolution textures and large scene data. At 4K and beyond, the combination of 11 GB capacity and 616 GB/s bandwidth helps mitigate stutter from texture swapping, though the exact impact depends on the workload. The 352-bit bus width is wider than many contemporary cards, contributing to the high bandwidth figure. For multi-monitor setups or VR, the memory subsystem offers ample headroom, as evidenced by the card’s strong performance in compute and rasterization benchmarks.

FAQ

Q: What is the average benchmark score of the RTX 2080 Ti?

A: The average benchmark score across all tests in the database is 30,459, placing it in the 74th percentile of all GPUs.

Q: How does the RTX 2080 Ti compare to the AMD Radeon RX 590?

A: The RTX 2080 Ti scores 30,459 versus the RX 590’s 30,457, a delta of 0% — effectively a tie.

Q: What APIs does the RTX 2080 Ti support?

A: It supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the memory configuration of the RTX 2080 Ti?

A: It has 11 GB of GDDR6 on a 352-bit bus, with 616.0 GB/s bandwidth and a 1750 MHz (14 Gbps effective) memory clock.

Q: How many RT and tensor cores does the RTX 2080 Ti have?

A: It features 68 RT cores and 544 tensor cores.

Q: What is the TDP and power connector requirement?

A: The TDP is 250 W, and it requires two 8-pin power connectors, with a suggested PSU of 600 W.

Benchmark Performance

The RTX 2080 Ti’s individual benchmark results show a mixed picture across different workloads. In the 3DMark Steel Nomad DX12 test, it scores 3,537 points, a strong result for a GPU of its generation. The Geekbench OpenCL score is 134,929, while the Vulkan score is 132,930 — both indicate robust compute performance across different APIs. In Passmark tests, the card achieves 21,548 in G3D (the overall 3D graphics score) and 10,056 in GPU compute. The DirectX-specific Passmark scores are 153 (DX10), 190 (DX11), 84 (DX12), and 235 (DX9). The low DX12 score relative to DX11 and DX9 is notable, but the average benchmark score aggregates all tests.

The aggregate average of 30,459 places the RTX 2080 Ti in a tight cluster with its nearest rivals. The AMD Radeon RX 590 is essentially tied at 30,457 (0% delta). The RX 590 GME is 1% faster, while the RX 570 is 1% slower, and the RTX 5060 is 1.2% slower. These deltas are all within a few percent, suggesting that the RTX 2080 Ti’s overall performance is competitive with these cards, despite being from an older generation. The percentile ranking of 74 indicates that it outperforms three-quarters of all GPUs in the database, which is a solid standing for a card released in 2018.

When interpreting these results, it is important to note that the average benchmark score is a composite of many tests, some of which may favor specific architectures. The RTX 2080 Ti’s strong compute scores (Geekbench, Passmark GPU compute) and competitive rasterization performance keep it relevant even as newer cards enter the market. However, the 1% deficit to the RX 590 GME and the 1.2% lead over the RTX 5060 are marginal, meaning real-world performance will vary by application and driver maturity. The data shows that the RTX 2080 Ti remains a capable performer, but it no longer leads its price segment — indeed, its launch MSRP of 999 USD is far above the current market positioning of its rivals, though pricing is not a factor in these benchmark comparisons.

In summary, the RTX 2080 Ti delivers a balanced set of scores that place it in the upper quartile of GPUs, with a near-tie against the RX 590 and a slight edge over the RX 570 and RTX 5060. Its memory subsystem and feature set were advanced for its time, and the benchmark data confirms that it still holds its own against more recent mid-range offerings.