AMD Radeon RX 470 vs NVIDIA GeForce RTX 2080 Ti Comparison

Head-to-Head Benchmarks

The benchmark data presents a starkly one-sided comparison between the NVIDIA GeForce RTX 2080 Ti and the AMD Radeon RX 470. Across the three shared tests, the RTX 2080 Ti secures a clean sweep, with wins in every single benchmark. The most dramatic margin appears in the 3DMark Steel Nomad DX12 test, where the RTX 2080 Ti scores 3537 against the RX 470's 842. This translates to a delta of 320.1%, meaning the NVIDIA card delivers more than four times the performance in this particular DirectX 12 workload. This is not a marginal lead; it is a generational chasm.

The gap narrows somewhat in compute-oriented tests, but the overall pattern remains consistent. In Geekbench OpenCL, the RTX 2080 Ti posts 128,171 points, while the RX 470 manages 33,568. The delta here is 281.8%, still a massive advantage for the Turing-based card. Moving to Geekbench Vulkan, the RTX 2080 Ti scores 132,930, eclipsing the RX 470's 39,884 by 233.3%. While the percentage lead shrinks across these three tests, it never falls below triple digits. The data indicates that regardless of the API or workload type, the RTX 2080 Ti maintains a commanding lead.

It is worth noting the context of these scores. The RTX 2080 Ti sits in the 75th percentile of all GPUs, with an average benchmark score of 29,783. Its closest rivals according to the data are the NVIDIA GeForce RTX 5070 Mobile (29,928, delta -0.5%) and the NVIDIA GeForce RTX 3070 Ti (29,945, delta -0.5%). This places the 2080 Ti in a performance tier that is essentially level with these modern and recent high-end parts. The RX 470, by contrast, sits in the 74th percentile with an average score of 28,996. While its percentile rank is surprisingly close, its nearest rivals tell a different story: the AMD Radeon RX 6800M (28,874, delta 0.4%), Intel Arc A370M (29,175, delta -0.6%), and AMD Radeon RX Vega M GH (29,197, delta -0.7%). These are primarily mobile and integrated-class parts, illustrating that the RX 470's average is buoyed by a different set of benchmarks than the ones shared with the RTX 2080 Ti.

The disparity in raw specifications aligns with the benchmark results. The RTX 2080 Ti features 4,352 shading units, 272 TMUs, and 88 ROPs, while the RX 470 has 2,048 shading units, 128 TMUs, and just 32 ROPs. The fill rates reflect this: the RTX 2080 Ti achieves 136.0 GPixel/s and 420.2 GTexel/s, versus the RX 470's 38.59 GPixel/s and 154.4 GTexel/s. In terms of raw floating-point compute, the RTX 2080 Ti delivers 13.45 TFLOPS of FP32 performance, nearly three times the RX 470's 4.940 TFLOPS.

The memory subsystem further widens the gap. The RTX 2080 Ti has 11 GB of GDDR6 memory on a 352-bit bus, yielding 616.0 GB/s of bandwidth. The RX 470 has 4 GB of GDDR5 on a 256-bit bus, providing 211.2 GB/s. This difference is critical for modern games and high-resolution textures. The RTX 2080 Ti's memory clock runs at 1750 MHz (14 Gbps effective), while the RX 470's memory runs at 1650 MHz (6.6 Gbps effective). The combination of more memory, a wider bus, and faster effective speed creates a bandwidth advantage that is directly observable in the benchmark deltas.

The Verdict

The data is unambiguous. The NVIDIA GeForce RTX 2080 Ti wins every head-to-head benchmark against the AMD Radeon RX 470. With a 320.1% lead in 3DMark Steel Nomad DX12, a 281.8% lead in Geekbench OpenCL, and a 233.3% lead in Geekbench Vulkan, there is no scenario in this dataset where the RX 470 comes out ahead. The RTX 2080 Ti is the definitive choice for anyone prioritizing raw performance, especially in DirectX 12 and compute-heavy workloads.

The RTX 2080 Ti's placement among its nearest rivals reinforces its standing. It is within 1% of the RTX 5070 Mobile and RTX 3070 Ti, and only 1% behind the AMD Radeon RX 6800. This means that despite being an older end-of-life product, the 2080 Ti still trades blows with much newer hardware. Its 75th percentile ranking and average score of 29,783 place it firmly in the high-performance tier.

For the RX 470, the story is about context rather than direct competition. Its 74th percentile ranking is respectable, and its average score of 28,996 is only about 2.6% lower than the 2080 Ti's. However, this average is skewed by benchmarks like Geekbench Metal (41,690) and a different set of tests that the RTX 2080 Ti does not share. When directly compared, the RX 470's nearest rivals include the RX 6800M and Intel Arc A370M, indicating it competes in a much lower performance bracket. The data shows the RX 470 as a capable part for its era, but it is not in the same league as the RTX 2080 Ti.

Where Each One Wins

The NVIDIA GeForce RTX 2080 Ti wins in every benchmark category where both cards have results. In 3DMark Steel Nomad DX12, a modern DirectX 12 test, the 320.1% delta suggests the RTX 2080 Ti is built for current-generation gaming workloads. Its 68 RT cores and 544 tensor cores, while not directly benchmarked here, are architectural features that the RX 470 completely lacks, pointing to advantages in ray tracing and AI-accelerated tasks that the RX 470 cannot execute at all.

In Geekbench OpenCL and Vulkan, the RTX 2080 Ti's leads of 281.8% and 233.3% respectively indicate superior general-purpose compute and cross-API performance. The RX 470 does have a Geekbench Metal score of 41,690, which is its only unique result, but this does not translate to a win against the RTX 2080 Ti since the NVIDIA card has no Metal result for comparison. The RX 470's wins, if any, would have to come from workloads not represented in this dataset, such as legacy DirectX 9 or 10 titles where its architecture might be more efficient relative to its specs, but no such data exists here.

The RTX 2080 Ti also wins on memory capacity and bandwidth. With 11 GB versus 4 GB, the NVIDIA card can handle larger textures and datasets without spilling into system memory. The 616.0 GB/s bandwidth versus 211.2 GB/s means that memory-bound tasks will disproportionately favor the RTX 2080 Ti. For users working with high-resolution assets or running modern games at high settings, the RTX 2080 Ti is the clear choice.

FAQ

Q: How much faster is the NVIDIA GeForce RTX 2080 Ti than the AMD Radeon RX 470 in 3DMark Steel Nomad DX12?

A: The RTX 2080 Ti scores 3537, while the RX 470 scores 842. This represents a delta of 320.1%, making the RTX 2080 Ti over four times faster in this DirectX 12 benchmark.

Q: Does the AMD Radeon RX 470 win any shared benchmark against the RTX 2080 Ti?

A: No. In the three shared tests (3DMark Steel Nomad DX12, Geekbench OpenCL, and Geekbench Vulkan), the RTX 2080 Ti wins all three. The wins tally is 3-0 in favor of NVIDIA.

Q: What is the difference in average benchmark scores between the two cards?

A: The RTX 2080 Ti has an average benchmark score of 29,783, while the RX 470 has an average of 28,996. The RTX 2080 Ti's average is about 2.7% higher, though the individual benchmark deltas are far larger.

Q: How does the RTX 2080 Ti compare to its nearest rivals?

A: The RTX 2080 Ti is within 0.5% of the RTX 5070 Mobile (29,928) and RTX 3070 Ti (29,945), and within 1% of the AMD Radeon RX 6800 (30,095). It is 2% ahead of the AMD FirePro W8000 (29,211).

Q: What is the memory bandwidth difference?

A: The RTX 2080 Ti has a memory bandwidth of 616.0 GB/s, while the RX 470 has 211.2 GB/s. The NVIDIA card provides nearly three times the bandwidth.

Q: Does the RX 470 have ray tracing or tensor cores?

A: No. The RX 470 has no RT cores and no tensor cores. The RTX 2080 Ti has 68 RT cores and 544 tensor cores.

Architecture Differences

The two cards are built on fundamentally different architectures from different eras. The NVIDIA GeForce RTX 2080 Ti uses the Turing architecture, manufactured on a 12 nm process at TSMC. Its chip, TU102, contains 18,600 million transistors on a die size of 754 mm², yielding a transistor density of 24.7M per mm². Turing introduced dedicated RT cores (68 on this card) for real-time ray tracing and tensor cores (544) for AI and deep learning workloads. These features are entirely absent from the AMD card.

The AMD Radeon RX 470 uses the GCN 4.0 architecture, also known as Polaris, manufactured on a 14 nm process at GlobalFoundries. Its chip, Ellesmere, contains 5,700 million transistors on a die size of 232 mm², with a transistor density of 24.6M per mm². GCN 4.0 is a refined version of AMD's Graphics Core Next design, which focuses on compute-heavy shader processing but lacks dedicated hardware for ray tracing or tensor operations.

The FP16 (half-precision) performance highlights a key difference. The RTX 2080 Ti achieves 26.90 TFLOPS FP16 using a 2:1 ratio relative to its FP32 performance, meaning it can double its throughput for half-precision workloads. The RX 470 also achieves 4.940 TFLOPS FP16, but at a 1:1 ratio, meaning it has no such acceleration. This gives the RTX 2080 Ti a significant advantage in workloads that can utilize FP16, such as certain machine learning inference tasks.

The API support also differs. The RTX 2080 Ti supports DirectX 12 Ultimate (12_2), while the RX 470 supports DirectX 12 (12_0). This means the NVIDIA card is compatible with the latest DirectX 12 Ultimate features, including ray tracing and mesh shaders. Both cards support OpenGL 4.6, but the RTX 2080 Ti supports Vulkan 1.4 while the RX 470 supports Vulkan 1.3.

Specification Differences

The most notable specification differences are in the core configuration and memory subsystem. The RTX 2080 Ti has 4,352 shading units, 272 TMUs, and 88 ROPs, compared to the RX 470's 2,048 shading units, 128 TMUs, and 32 ROPs. This means the RTX 2080 Ti has more than double the shader processors and more than double the texture units, while having nearly triple the ROPs.

Memory is a major divider. The RTX 2080 Ti has 11 GB of GDDR6 memory on a 352-bit bus, while the RX 470 has 4 GB of GDDR5 on a 256-bit bus. The bandwidth figures are 616.0 GB/s versus 211.2 GB/s. Clock speeds also differ, with the RTX 2080 Ti running at 1350 MHz base and 1545 MHz boost, while the RX 470 runs at 926 MHz base and 1206 MHz boost.

Power and physical specifications differ substantially. The RTX 2080 Ti has a TDP of 250 W and requires a 600 W power supply, using two 8-pin connectors. The RX 470 has a TDP of 120 W and requires a 300 W power supply, using a single 6-pin connector. Both are dual-slot cards, but the RTX 2080 Ti is longer at 267 mm (10.5 inches) versus 240 mm (9.4 inches) and taller at 116 mm versus 95 mm. Both have the same 35 mm width.

Display outputs differ slightly. The RTX 2080 Ti has 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C. The RX 470 has 1x HDMI 2.0b and 3x DisplayPort 1.4a. The process nodes are 12 nm for NVIDIA versus 14 nm for AMD, and the transistor counts are 18,600 million versus 5,700 million. The release dates are also different, with the RTX 2080 Ti launching on 2018-09-19 and the RX 470 on 2016-08-03. The launch MSRP is 999 USD for the RTX 2080 Ti and 179 USD for the RX 470.