AMD Radeon Pro 570 vs AMD Radeon RX 6800M Comparison

Where Each One Wins

The benchmark data splits cleanly between these two mobile graphics parts. The AMD Radeon RX 6800M wins every single recorded head-to-head test, while the AMD Radeon Pro 570 does not register a single victory in the shared benchmark suite. That is a decisive sweep, but the magnitude of the win varies meaningfully across different workloads.

In the Geekbench Metal test, which exercises the GPU through Apple's Metal API, the RX 6800M scores 113,721 against the Pro 570's 39,945. That is a 64.9% advantage for the newer part. The OpenCL results tell a similar story: the RX 6800M posts 87,621 versus 27,702, a 68.4% lead. Vulkan shows the RX 6800M at 94,766 against 31,974, a 66.3% gap. Across all three compute-oriented benchmarks, the RX 6800M is consistently between roughly 2.8x and 3.2x faster than the Pro 570.

However, the database also records a broader set of benchmarks for the RX 6800M that the Pro 570 does not share. The RX 6800M scores 13,261 in Passmark G3D, 5,032 in Passmark GPU Compute, and 538 in Passmark G2D. Its DirectX 9, 10, 11, and 12 Passmark scores are 147, 101, 127, and 65 respectively. These additional data points show where the RX 6800M's strengths lie: it is a much more capable part in every measured category, from legacy DirectX 9 workloads to modern DirectX 12 and general-purpose compute.

The average benchmark score tells a more nuanced story. The Pro 570 averages 33,207 across its three recorded tests, while the RX 6800M averages 28,874 across its eleven recorded tests. This difference is partly due to the Passmark DirectX scores being much lower in absolute terms than the Geekbench scores, which drags down the RX 6800M's overall average. The percentile rankings reflect this: the Pro 570 sits at the 78th percentile of all GPUs, while the RX 6800M sits at the 74th percentile. The Pro 570's higher percentile despite losing every head-to-head test indicates that its three benchmarks are all strong relative to the broader GPU population, whereas the RX 6800M's Passmark results are weaker relative to its Geekbench results.

The nearest rivals for each part reinforce this positioning. The Pro 570's closest competitor is the NVIDIA GeForce RTX 3050 Mobile, which averages 33,170, a mere 0.1% difference. The NVIDIA T550 Mobile (33,161, 0.1%), NVIDIA P104-100 (32,982, 0.7%), and NVIDIA T600 Mobile (32,849, 1.1%) are all within striking distance. The RX 6800M, by contrast, sits near the AMD Radeon RX 570 (28,766, 0.4%), AMD Radeon RX 470 (28,996, -0.4%), AMD Radeon R9 M295X (28,580, 1%), and Intel Arc A370M (29,175, -1%). The RX 6800M's rivals are older or lower-tier parts, which suggests its average score is depressed by the Passmark tests rather than by any weakness in raw compute.

The Verdict

The data presents an unambiguous picture for most workloads. If the task involves Metal, OpenCL, or Vulkan compute, the AMD Radeon RX 6800M is the clear choice. It leads by 64.9%, 68.4%, and 66.3% respectively in those three shared tests. No interpretation of the numbers can rescue the Pro 570 in these categories; the RX 6800M is roughly three times faster in every shared benchmark.

The AMD Radeon Pro 570 does have one statistical advantage: its 78th percentile ranking versus the RX 6800M's 74th percentile. This reflects the Pro 570's consistency across its three measured tests, all of which land in a relatively narrow band (27,702 to 39,945). The RX 6800M's range is far wider (65 to 113,721), and its Passmark DirectX results are exceptionally low in absolute terms. For a user who cares about percentile placement relative to the entire GPU database, the Pro 570 looks better on paper. But that is a statistical artifact of the benchmark mix, not a performance advantage.

The RX 6800M's additional benchmark data shows it is also strong in DirectX-based workloads. Its Passmark G3D score of 13,261 is substantial, and its GPU Compute score of 5,032 indicates solid general-purpose performance. The DirectX 9 through 12 scores (147, 101, 127, 65) are harder to interpret without a comparison point, but they represent the RX 6800M's measured performance across those APIs.

For a user deciding between these two parts, the RX 6800M is the choice for any application that uses Metal, OpenCL, or Vulkan. The Pro 570's only claim is its higher percentile ranking, which may matter for database comparisons but does not translate into a single benchmark win. The RX 6800M also offers more memory (12 GB versus 4 GB), a newer architecture (RDNA 2.0 versus GCN 4.0), and support for DirectX 12 Ultimate versus DirectX 12 (12_0). The Pro 570 is end-of-life with a 2017 release date; the RX 6800M is also end-of-life but released in 2021.

Head-to-Head Benchmarks

The three shared benchmarks all follow the same pattern: the RX 6800M wins by a wide margin, and the Pro 570 trails by roughly two-thirds. The largest delta is in OpenCL, where the RX 6800M's 87,621 score beats the Pro 570's 27,702 by 68.4%. That is the single biggest gap in the head-to-head data. The RX 6800M is more than three times faster in this test, which typically reflects raw compute throughput rather than driver optimizations.

The Vulkan test shows a 66.3% lead for the RX 6800M (94,766 versus 31,974). Vulkan is a low-level API that tends to expose architectural differences directly, and the RDNA 2.0 architecture's newer design shows here. The Metal test has the smallest margin of the three at 64.9% (113,721 versus 39,945), but it is still a massive advantage. Metal is Apple's API, and the Pro 570 is specifically listed as part of the "Radeon Pro Mac (500 Series)" generation, so it might be expected to perform relatively better in Metal. The data does not support that expectation; the RX 6800M still wins decisively.

The Pro 570's best result relative to the RX 6800M is in Metal, where its score of 39,945 is 35.1% of the RX 6800M's 113,721. In OpenCL, the Pro 570 achieves only 31.6% of the RX 6800M's score. In Vulkan, it achieves 33.7%. So the Pro 570 is proportionally least far behind in Metal, but "least far behind" still means losing by nearly two-thirds.

The RX 6800M's additional benchmarks provide context for its overall capability. Its Passmark G3D score of 13,261 is its highest Passmark result, while its DirectX 12 score of 65 is its lowest. The DirectX 9 score of 147 is the highest among the four DirectX versions tested. This pattern suggests the RX 6800M handles older APIs relatively well but struggles with DirectX 12 in the Passmark suite, possibly due to driver characteristics or the specific workload in that test.

FAQ

Q: Which GPU wins more benchmarks in the head-to-head comparison?

A: The AMD Radeon RX 6800M wins all three shared benchmarks: Geekbench Metal (113,721 versus 39,945), Geekbench OpenCL (87,621 versus 27,702), and Geekbench Vulkan (94,766 versus 31,974). The Pro 570 records zero wins.

Q: What is the largest performance gap between the two GPUs?

A: The largest delta is in Geekbench OpenCL, where the RX 6800M leads by 68.4%. The Vulkan test shows a 66.3% gap, and the Metal test shows a 64.9% gap.

Q: How do their overall percentile rankings compare?

A: The Pro 570 sits at the 78th percentile of all GPUs, while the RX 6800M sits at the 74th percentile. This is despite the RX 6800M winning every head-to-head test, due to the RX 6800M's lower Passmark scores dragging down its average.

Q: What is the average benchmark score for each GPU?

A: The Pro 570 averages 33,207 across its three benchmarks (Metal, OpenCL, Vulkan). The RX 6800M averages 28,874 across its eleven benchmarks, which include three Geekbench tests and eight Passmark tests.

Q: Does the Pro 570 have any performance advantage in any shared test?

A: No. The RX 6800M wins all three shared benchmarks. The Pro 570's only statistical advantage is its higher percentile ranking (78 versus 74), which reflects its narrower and consistently high benchmark scores.

Q: Which GPU has better memory specifications?

A: The RX 6800M has 12 GB of GDDR6 memory with a 192-bit bus and 384.0 GB/s bandwidth. The Pro 570 has 4 GB of GDDR5 memory with a 256-bit bus and 217.0 GB/s bandwidth.

Architecture Differences

The two GPUs come from entirely different architectural generations. The AMD Radeon Pro 570 uses the Ellesmere chip based on GCN 4.0 architecture, built on a 14 nm process at GlobalFoundries. The RX 6800M uses the Navi 22 chip based on RDNA 2.0 architecture, built on a 7 nm process at TSMC. This process shrink from 14 nm to 7 nm is a major factor in the performance difference.

The transistor counts reflect the architectural leap. The Pro 570 contains 5,700 million transistors on a 232 mm² die, giving a transistor density of 24.6 million per mm². The RX 6800M contains 17,200 million transistors on a 335 mm² die, with a density of 51.3 million per mm². The RX 6800M has roughly three times the transistor count and more than double the density.

The compute resources differ substantially. The Pro 570 has 1,792 shading units, 112 texture mapping units, and 32 raster operations pipelines. The RX 6800M has 2,560 shading units, 160 TMUs, and 64 ROPs. The RX 6800M also has 40 ray tracing cores, which the Pro 570 lacks entirely. This is a fundamental feature difference: RDNA 2.0 includes dedicated ray tracing hardware, while GCN 4.0 has none.

Clock speeds also differ significantly. The Pro 570 runs at a 1000 MHz base and 1105 MHz boost. The RX 6800M runs at 2116 MHz base, 2300 MHz game clock, and 2390 MHz boost. The RX 6800M's boost clock is more than double the Pro 570's base clock. The memory clocks differ as well: the Pro 570 uses 1695 MHz GDDR5 (6.8 Gbps effective), while the RX 6800M uses 2000 MHz GDDR6 (16 Gbps effective).

The pixel and texture rates reflect these differences. The Pro 570 delivers 35.36 GPixel/s and 123.8 GTexel/s. The RX 6800M delivers 153.0 GPixel/s and 382.4 GTexel/s. The FP32 compute is 3.960 TFLOPS for the Pro 570 versus 12.24 TFLOPS for the RX 6800M. The FP16 rates diverge even more: 3.960 TFLOPS (1:1) for the Pro 570 versus 24.47 TFLOPS (2:1) for the RX 6800M.

The API support also differs. The Pro 570 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. The RX 6800M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RX 6800M's DirectX 12 Ultimate support includes features like ray tracing and mesh shaders that the Pro 570 cannot access.

Specification Differences

The two GPUs differ across nearly every recorded specification. The Pro 570 is from the "Radeon Pro Mac (500 Series)" generation, while the RX 6800M is from the "Navi Mobile (RX 6000M)" generation. The Pro 570's chip is Ellesmere; the RX 6800M's chip is Navi 22. The process nodes are 14 nm versus 7 nm, and the foundries are GlobalFoundries versus TSMC.

Memory is a major differentiator. The Pro 570 has 4 GB of GDDR5 on a 256-bit bus with 217.0 GB/s bandwidth. The RX 6800M has 12 GB of GDDR6 on a 192-bit bus with 384.0 GB/s bandwidth. Despite the narrower bus, the RX 6800M's faster GDDR6 memory delivers 77% more bandwidth.

The bus interface differs: PCIe 3.0 x16 for the Pro 570 versus PCIe 4.0 x16 for the RX 6800M. Both have IGP slot width and no power connectors, and both are listed as "Portable Device Dependent" for display outputs. The power draw is similar: 150 W for the Pro 570 and 145 W for the RX 6800M.

Release dates are four years apart: June 2017 for the Pro 570 and May 2021 for the RX 6800M. Both are end-of-life. The RX 6800M has a listed predecessor, "Polaris Mobile," while the Pro 570 has none. Neither has a successor listed, and neither has a launch MSRP in the database. The RX 6800M's shading units (2,560) exceed the Pro 570's (1,792) by 768, and its TMUs (160 versus 112) and ROPs (64 versus 32) are also higher. The RX 6800M's 40 ray tracing cores have no counterpart in the Pro 570.