AMD Radeon RX 570 vs NVIDIA GeForce RTX 2080 SUPER Comparison

AMD Radeon RX 570 and NVIDIA GeForce RTX 2080 SUPER occupy very different positions in the recorded benchmark database. The RTX 2080 SUPER wins every head-to-head test in our measurements, and the RX 570 trails by margins that range from 53.2% to 67.7% depending on the workload. However, the average benchmark score tells a more nuanced story: the RX 570 sits at 28766, which places it in the 74th percentile of all GPUs, while the RTX 2080 SUPER averages 24170, landing in the 69th percentile. That inversion is worth examining, because it suggests the two cards are being measured across different test suites, and the raw head-to-head results favor the newer, larger GPU by a wide margin.

Where Each One Wins

The RTX 2080 SUPER dominates every recorded head-to-head benchmark. In 3DMark Steel Nomad DX12, it scores 1882 against the RX 570's 880, a 53.2% advantage. In Geekbench OpenCL, the gap widens to 99226 versus 32084, a 67.7% lead. In Geekbench Vulkan, the RTX 2080 SUPER posts 111284 while the RX 570 manages 42752, a 61.6% margin. There is no test in the database where the RX 570 comes out ahead of the RTX 2080 SUPER.

The RX 570's wins, if they can be called that, are contextual rather than direct. Its average benchmark score of 28766 is higher than the RTX 2080 SUPER's 24170, which reflects the fact that the RX 570 has additional recorded tests that the RTX 2080 SUPER does not share. Specifically, the RX 570 has a Geekbench Metal score of 39349, which is not present in the RTX 2080 SUPER's benchmark list. That Metal result pulls the RX 570's average upward. The RTX 2080 SUPER, by contrast, has a suite that includes Passmark tests at lower scores, such as 140 for DirectX 10, 165 for DirectX 11, 75 for DirectX 12, 227 for DirectX 9, 920 for G2D, 19490 for G3D, and 8290 for GPU compute, which drag its average down.

In terms of pure compute throughput, the RTX 2080 SUPER is clearly the stronger part. Its FP32 rating is 11.15 TFLOPS versus 5.095 TFLOPS for the RX 570. Its texture rate is 348.5 GTexel/s versus 159.2 GTexel/s. Its pixel rate is 116.2 GPixel/s versus 39.81 GPixel/s. The RTX 2080 SUPER also has dedicated RT cores (48) and tensor cores (384), which the RX 570 lacks entirely. Those features do not appear in the head-to-head benchmark list, but their presence in the specification data indicates a capability gap beyond raw rasterization.

The Verdict

The data points to a clear conclusion: the RTX 2080 SUPER is the faster GPU in every test where both are measured. Its smallest win, 53.2% in 3DMark Steel Nomad DX12, is still a dominant margin. Its largest win, 67.7% in Geekbench OpenCL, shows that the gap grows in compute-heavy workloads. If a user is choosing between these two based on recorded performance, the RTX 2080 SUPER is the only option that wins.

The RX 570 is not without a role. Its average benchmark score is higher, and its percentile ranking (74th) is above the RTX 2080 SUPER's (69th). That is partly a function of the test set, but it also indicates that the RX 570 remains competitive in the broader database context, especially in Metal-based workloads where it records a 39349 score. For users who only care about the specific tests in the head-to-head list, the RX 570 is not competitive. For users who weight the full benchmark profile, the RX 570's higher average and percentile suggest it holds up better against the wider field of GPUs, even if it loses decisively to the RTX 2080 SUPER in direct comparison.

The RTX 2080 SUPER also carries newer hardware features. It supports DirectX 12 Ultimate (12_2) versus the RX 570's DirectX 12 (12_0). It lists Vulkan 1.4 support versus the RX 570's Vulkan 1.3. It has 8 GB of GDDR6 memory with 495.9 GB/s bandwidth, while the RX 570 has 4 GB of GDDR5 with 224.0 GB/s. Those are not just numbers; they translate into higher resolution capability and faster texture streaming in memory-intensive scenarios.

Head-to-Head Benchmarks

The largest recorded gap is in Geekbench OpenCL, where the RTX 2080 SUPER scores 99226 against the RX 570's 32084. That is a 67.7% deficit for the RX 570, meaning the RTX 2080 SUPER delivers roughly three times the OpenCL performance. This result aligns with the raw compute specifications: the RTX 2080 SUPER has 3072 shading units, 192 texture mapping units, and 64 ROPs, versus 2048 shading units, 128 TMUs, and 32 ROPs for the RX 570. The RTX 2080 SUPER also has a boost clock of 1815 MHz versus 1244 MHz, so it has both more units and a higher clock speed.

Geekbench Vulkan shows a 61.6% lead for the RTX 2080 SUPER, with scores of 111284 and 42752. This is the RTX 2080 SUPER's highest single score across all tests, and it confirms that the Turing architecture scales well in Vulkan workloads. The RX 570's Vulkan score of 42752 is its second-highest recorded result after its Metal score, which suggests the RX 570 is reasonably efficient in modern APIs, but the RTX 2080 SUPER simply has more headroom.

The smallest margin is in 3DMark Steel Nomad DX12, where the RTX 2080 SUPER leads 1882 to 880, a 53.2% difference. This is still a substantial win, but it is the closest the RX 570 gets in the head-to-head set. The RX 570's 880 score is less than half of the RTX 2080 SUPER's 1882, which indicates that even in a modern DX12 title, the older Polaris architecture cannot close the gap.

There is no measurement in which the RX 570 wins. The head-to-head record is 0 wins for the RX 570 and 3 wins for the RTX 2080 SUPER. In every case, the delta is negative from the RX 570's perspective, ranging from -53.2% to -67.7%.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon RX 570 has an average benchmark score of 28766, while the NVIDIA GeForce RTX 2080 SUPER averages 24170.

Q: What is the largest performance gap in the head-to-head tests?

A: The largest gap is in Geekbench OpenCL, where the RTX 2080 SUPER scores 99226 and the RX 570 scores 32084, a 67.7% difference.

Q: Does the RX 570 ever beat the RTX 2080 SUPER in any recorded test?

A: No. The head-to-head results show 3 wins for the RTX 2080 SUPER and 0 wins for the RX 570.

Q: How do the memory configurations compare?

A: The RTX 2080 SUPER has 8 GB of GDDR6 memory with a 256-bit bus and 495.9 GB/s bandwidth. The RX 570 has 4 GB of GDDR5 memory with a 256-bit bus and 224.0 GB/s bandwidth.

Q: What is the percentile ranking difference?

A: The RX 570 is in the 74th percentile of all GPUs, while the RTX 2080 SUPER is in the 69th percentile.

Q: Does the RTX 2080 SUPER support ray tracing?

A: Yes, the RTX 2080 SUPER includes 48 RT cores and 384 tensor cores. The RX 570 has no RT cores or tensor cores listed.

Architecture Differences

The two GPUs come from different architectural generations. The RX 570 uses AMD's GCN 4.0 architecture, built on a 14 nm process at GlobalFoundries. The chip is Polaris 20, with 5,700 million transistors on a 232 mm² die. The RTX 2080 SUPER uses NVIDIA's Turing architecture, built on a 12 nm process at TSMC. The chip is TU104, with 13,600 million transistors on a 545 mm² die.

The transistor density is similar: 24.6 million transistors per square millimeter for the RX 570 and 25.0 million per square millimeter for the RTX 2080 SUPER. But the RTX 2080 SUPER has more than twice the total transistor count, which reflects its larger die and more complex feature set.

The RTX 2080 SUPER is the only one of the two with dedicated ray tracing and tensor cores. It lists 48 RT cores and 384 tensor cores, while the RX 570 has no such entries. This is a fundamental architectural difference: the RTX 2080 SUPER is designed to accelerate ray-traced workloads and AI-based operations, while the RX 570 is purely a rasterization GPU.

The FP16 throughput also differs. The RX 570 has FP16 performance equal to its FP32 at 5.095 TFLOPS, a 1:1 ratio. The RTX 2080 SUPER has FP16 performance of 22.30 TFLOPS versus its FP32 of 11.15 TFLOPS, a 2:1 ratio. This means the RTX 2080 SUPER can double its throughput in FP16 workloads, a feature that is absent in the RX 570.

The API support differs as well. The RTX 2080 SUPER supports DirectX 12 Ultimate (12_2) and Vulkan 1.4. The RX 570 supports DirectX 12 (12_0) and Vulkan 1.3. Both support OpenGL 4.6.

Specification Differences

The RTX 2080 SUPER has a higher base clock (1650 MHz versus 1168 MHz) and a higher boost clock (1815 MHz versus 1244 MHz). Its memory runs at 1937 MHz with an effective data rate of 15.5 Gbps, compared to 1750 MHz and 7 Gbps effective for the RX 570.

Memory capacity and type differ: 8 GB GDDR6 versus 4 GB GDDR5. The bus width is the same at 256 bits, but the bandwidth is more than double: 495.9 GB/s versus 224.0 GB/s.

The compute units scale up significantly. The RTX 2080 SUPER has 3072 shading units, 192 TMUs, and 64 ROPs. The RX 570 has 2048 shading units, 128 TMUs, and 32 ROPs. Pixel rate and texture rate follow: 116.2 GPixel/s and 348.5 GTexel/s for the RTX 2080 SUPER, versus 39.81 GPixel/s and 159.2 GTexel/s for the RX 570.

Power requirements differ. The RTX 2080 SUPER has a TDP of 250 W and needs a 600 W power supply, using a 6-pin plus an 8-pin connector. The RX 570 has a TDP of 150 W, requires a 450 W power supply, and uses a single 6-pin connector.

Physical dimensions vary. The RTX 2080 SUPER is 267 mm long, 116 mm tall, and 35 mm wide. The RX 570 is 241 mm long. Both are dual-slot cards.

Display outputs differ. The RTX 2080 SUPER offers 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C. The RX 570 offers 1x DVI, 1x HDMI 2.0b, and 3x DisplayPort 1.4a.

Release dates are separated by over two years. The RX 570 launched on 2017-04-17, while the RTX 2080 SUPER launched on 2019-07-22. Both are now end-of-life products. The RX 570's predecessor is Arctic Islands and its successor is Vega. The RTX 2080 SUPER's predecessor is GeForce 10 and its successor is GeForce 30.