AMD Radeon RX 6700 XT vs NVIDIA GeForce RTX 2080 Ti Comparison

The NVIDIA GeForce RTX 2080 Ti and AMD Radeon RX 6700 XT are both end-of-life cards that now compete in the used and clearance markets, but they were built for different eras. The 2080 Ti launched in late 2018 as NVIDIA’s flagship Turing part, while the 6700 XT arrived in early 2021 as a mid-range RDNA 2 offering. The benchmark data shows a clear overall winner in raw performance, but the 6700 XT holds its own in a few specific workloads, making the choice less obvious than the average scores suggest.

Head-to-Head Benchmarks

The most decisive result comes from the 3DMark Steel Nomad DX12 test, where the RTX 2080 Ti scores 3537 against the RX 6700 XT’s 2435. That is a 45.3% lead for NVIDIA, the largest margin of any shared benchmark. This is a modern, demanding DX12 workload, and the 2080 Ti’s older architecture still crushes the newer AMD card there. The Geekbench Vulkan test tells a similar story: 132930 for the 2080 Ti versus 93500 for the 6700 XT, a 42.2% advantage. NVIDIA also dominates Geekbench OpenCL with a score of 128171 compared to 44152, a massive 190.3% delta that suggests the 2080 Ti has far stronger compute throughput in that specific API.

The gap narrows considerably in the Passmark suite. In DirectX 10, the 2080 Ti wins 153 to 124, a 23.4% margin. DirectX 11 shows 190 versus 173, a 9.8% win for NVIDIA. DirectX 12 is closer still: 84 to 77, a 9.1% edge. The Passmark G3D score, which aggregates overall gaming performance, gives the 2080 Ti 21548 against 19786, an 8.9% lead. Even the GPU compute test favors NVIDIA at 10056 versus 9336, a 7.7% difference.

The RX 6700 XT secures exactly two wins, and both are narrow. In Passmark DirectX 9, AMD scores 242 against NVIDIA’s 235, a 2.9% edge. In the 2D test, AMD wins 951 to 927, a 2.5% margin. These are legacy workloads and low-stakes metrics, but they show the 6700 XT is not entirely outclassed. The overall win count is 8-2 in favor of the 2080 Ti, and the average benchmark score tells the same story: 29783 for NVIDIA versus 27425 for AMD, an 8.6% gap.

Architecture Differences

The two cards are built on fundamentally different designs. The RTX 2080 Ti uses the TU102 chip on TSMC’s 12 nm process, with 18,600 million transistors spread across a 754 mm² die. That yields a transistor density of 24.7 million per mm². The RX 6700 XT uses the Navi 22 chip on TSMC’s 7 nm process, packing 17,200 million transistors onto a much smaller 335 mm² die, giving a density of 51.3 million per mm². AMD’s newer node nearly doubles the transistor density, which is why the 6700 XT can achieve comparable performance from a much smaller chip.

The 2080 Ti is a Turing architecture part with 4352 shading units, 272 texture mapping units, 88 ROPs, 68 ray tracing cores, and 544 tensor cores. The 6700 XT is RDNA 2.0 with 2560 shading units, 160 TMUs, 64 ROPs, and 40 ray tracing cores, but it has no tensor cores at all. Clock speeds are dramatically different: the 2080 Ti runs at a 1350 MHz base and 1545 MHz boost, while the 6700 XT starts at 2321 MHz and boosts to 2581 MHz, with a 2424 MHz game clock. AMD compensates for fewer cores with much higher frequencies.

Memory configurations also differ. The 2080 Ti has 11 GB of GDDR6 on a 352-bit bus, providing 616 GB/s of bandwidth. The 6700 XT has 12 GB of GDDR6 on a 192-bit bus, yielding 384 GB/s. Despite having one more gigabyte, the AMD card has significantly less memory bandwidth. The NVIDIA card also supports PCIe 3.0 x16, while the 6700 XT uses the newer PCIe 4.0 x16 interface. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, and both feature ray tracing cores, though the 2080 Ti has 68 of them versus 40 on the AMD card.

Power draw and physical design are also distinct. The 2080 Ti has a 250 W TDP and requires two 8-pin power connectors with a 600 W suggested PSU. The 6700 XT has a 230 W TDP, needs one 6-pin and one 8-pin connector, and asks for a 550 W PSU. Both are dual-slot cards at 267 mm in length, but the 2080 Ti is 116 mm tall and 35 mm wide, while the 6700 XT is 110 mm tall and 40 mm wide. The 2080 Ti also includes a USB Type-C display output, which the AMD card lacks.

Where Each One Wins

The RTX 2080 Ti wins in nearly every modern and compute-oriented workload. Its 45.3% lead in 3DMark Steel Nomad and 42.2% lead in Geekbench Vulkan make it the clear choice for current DX12 titles and Vulkan-based games. The 190.3% OpenCL margin is extreme, suggesting the 2080 Ti is far better suited for GPU compute tasks like rendering or scientific workloads that depend on that API. In the Passmark suite, the 2080 Ti takes every DirectX 10, 11, and 12 test, so it is also the better option for older DirectX games that still see heavy use.

The RX 6700 XT wins in exactly two categories: DirectX 9 and 2D performance. The 2.9% lead in DirectX 9 (242 vs 235) means legacy titles from the early 2000s will run marginally better on AMD. The 2.5% edge in the 2D test (951 vs 927) affects desktop compositing and 2D applications, but this is a minor practical advantage. The 6700 XT also has 12 GB of VRAM versus 11 GB, which could matter in texture-heavy scenarios, although the 2080 Ti’s 616 GB/s bandwidth versus 384 GB/s is a substantial counterweight. For pure gaming across a broad library, the data favors the 2080 Ti in 8 of 10 shared tests.

FAQ

Q: Which card is faster in DirectX 12 gaming?

A: The RTX 2080 Ti wins the Passmark DirectX 12 test with 84 points versus 77 for the RX 6700 XT, a 9.1% advantage. In the more demanding 3DMark Steel Nomad DX12 benchmark, the 2080 Ti leads by a much larger 45.3%.

Q: Does the RX 6700 XT win any benchmarks?

A: Yes, it wins two of the ten shared tests. It scores 242 in Passmark DirectX 9 versus 235 for the 2080 Ti, a 2.9% lead, and it wins the 2D test 951 to 927, a 2.5% margin.

Q: How do the average benchmark scores compare?

A: The RTX 2080 Ti has an average benchmark score of 29783, putting it in the 75th percentile of all GPUs. The RX 6700 XT averages 27425, which places it in the 72nd percentile. The 2080 Ti holds an 8.6% advantage.

Q: Which card has more memory bandwidth?

A: The RTX 2080 Ti has 616 GB/s of bandwidth from its 11 GB GDDR6 memory on a 352-bit bus. The RX 6700 XT has 384 GB/s from 12 GB GDDR6 on a 192-bit bus.

Q: Are there differences in ray tracing hardware?

A: Yes. The RTX 2080 Ti has 68 ray tracing cores, while the RX 6700 XT has 40. The NVIDIA card also includes 544 tensor cores, which the AMD card lacks entirely.

Q: Which card requires a more powerful power supply?

A: The RTX 2080 Ti has a 250 W TDP and a suggested 600 W PSU, along with two 8-pin connectors. The RX 6700 XT has a 230 W TDP and a suggested 550 W PSU, using one 6-pin and one 8-pin connector.

Specification Differences

The two cards differ across nearly every major specification. The process node is 12 nm for the 2080 Ti versus 7 nm for the 6700 XT. Transistor counts are close at 18,600 million versus 17,200 million, but die size is vastly different: 754 mm² versus 335 mm². Transistor density is 24.7M/mm² versus 51.3M/mm². The 2080 Ti has 4352 shading units, 272 TMUs, 88 ROPs, 68 RT cores, and 544 tensor cores, while the 6700 XT has 2560 shading units, 160 TMUs, 64 ROPs, and 40 RT cores, with no tensor cores.

Clock speeds are reversed. The 2080 Ti runs at 1350 MHz base and 1545 MHz boost, while the 6700 XT runs at 2321 MHz base, 2581 MHz boost, and a 2424 MHz game clock. Memory is 11 GB at 1750 MHz (14 Gbps effective) on the 2080 Ti versus 12 GB at 2000 MHz (16 Gbps effective) on the 6700 XT. The 2080 Ti has a 352-bit bus and 616 GB/s bandwidth; the 6700 XT has a 192-bit bus and 384 GB/s. Pixel rate is 136.0 GPixel/s versus 165.2 GPixel/s, texture rate is 420.2 GTexel/s versus 413.0 GTexel/s, and FP32 is 13.45 TFLOPS versus 13.21 TFLOPS.

Power specifications differ: 250 W TDP with two 8-pin connectors and a 600 W PSU suggestion for NVIDIA, versus 230 W TDP with one 6-pin and one 8-pin connector and a 550 W PSU for AMD. The bus interface is PCIe 3.0 on the 2080 Ti and PCIe 4.0 on the 6700 XT. Display outputs are the same except the 2080 Ti adds a USB Type-C port. Physical dimensions are 267 x 116 x 35 mm for NVIDIA and 267 x 110 x 40 mm for AMD. The launch MSRP for the 2080 Ti was 999 USD, while the 6700 XT launched at 479 USD.

The Verdict

The data is unambiguous on raw performance: the RTX 2080 Ti is the faster card. It wins 8 of 10 shared benchmarks, including every modern API test, and it holds a 45.3% lead in the most demanding DX12 workload. Its average benchmark score is 29783 versus 27425, an 8.6% gap that places it in the 75th percentile of all GPUs compared to the 72nd percentile for the 6700 XT. The 2080 Ti also has 544 tensor cores, which the AMD card lacks, and nearly twice the memory bandwidth at 616 GB/s versus 384 GB/s.

However, the RX 6700 XT is not a poor choice. It wins DirectX 9 and 2D tests, has 12 GB of VRAM versus 11 GB, runs on a more efficient 7 nm process with a smaller die, and requires less power with a 230 W TDP versus 250 W. Its nearest rivals include the RTX 3090, from which it trails by only 0.5%, indicating it performs in a higher class than its mid-range positioning suggests. The 2080 Ti’s nearest rivals include the RTX 3070 Ti and RX 6800, both of which it trails by 0.5% and 1%, respectively.

For a builder prioritizing raw gaming performance and compute capability, the RTX 2080 Ti is the clear winner from the benchmark data. For someone who values a more modern, efficient process node, slightly more VRAM, and lower power draw, the RX 6700 XT is the pragmatic pick, even if it concedes most performance tests. The 2080 Ti’s 68 RT cores and 544 tensor cores give it a feature set that the 6700 XT cannot match, but the AMD card’s higher clocks and smaller footprint make it a more efficient package. The choice comes down to whether you need the 2080 Ti’s raw muscle or the 6700 XT’s efficiency and longer-lived process node.