AMD Radeon R9 M390X vs NVIDIA GeForce RTX 2060 SUPER Comparison

Head-to-Headed Benchmarks

The database contains one direct head-to-head benchmark result for these two GPUs, and it is a decisive victory for the NVIDIA GeForce RTX 2060 SUPER. In the Geekbench Vulkan test, the RTX 2060 SUPER scores 77,402 points, while the AMD Radeon M390X scores 16,951 points. That translates to a delta of -78.1%, meaning the NVIDIA GPU outperforms the AMD part by a staggering 60,451 points, a margin that dwarfs any other comparison in the database. The AMD card's average benchmark score across all tests, 20,662, is itself lower than the NVIDIA card's Vulkan score alone, which shows the scale of the gap in this specific API workload.

Looking at the broader average benchmark scores, the AMD Radeon R9 M390X actually holds a higher aggregate across all recorded tests, at 20,662, compared to the RTX 2060 SUPER's 18,093. That is a difference of 2,569 points in favor of the older AMD part. This apparent contradiction is explained by the different test suites: the AMD card's average is drawn from only two benchmark entries (Geekbench Metal and Geekbench Vulkan), while the NVIDIA card's average includes ten different tests, several of which are legacy DirectX and compute workloads that pull its average down. The RTX 2060 SUPER's Geekbench OpenCL score of 76,957 is its second-highest result, while its Passmark DirectX 9 score of 218 and DirectX 10 score of 111 are far lower.

In single-test comparison, the RTX 2060 SUPER's Vulkan advantage is consistent with its other compute-heavy results: it scores 76,957 in Geekbench OpenCL and 16,462 in Passmark G3D, while the AMD card's Vulkan score of 16,951 is its lowest recorded result. The RTX 2060 SUPER also posts a Passmark GPU Compute score of 6,721, a DirectX 12 score of 61, and a DirectX 11 score of 130. The AMD card has no equivalent DirectX results in the database, so cross-test comparisons are limited.

Percentile placement also differs meaningfully. The AMD Radeon R9 M390X sits at the 66th percentile among all GPUs, which places it ahead of the RTX 2060 SUPER's 62nd percentile despite the NVIDIA card's superior raw compute in most modern APIs. The RTX 2060 SUPER's nearest rivals in average score are the NVIDIA GeForce RTX 3060 Mobile (-0.4%), AMD Radeon Pro 5700 (-0.5%), AMD Radeon RX 460 (-1.5%), and Intel Arc A770M (-1.6%). The AMD M390X's nearest rivals are the AMD Radeon RX 5600 XT (-0.2%), Intel Arc A750 (+0.4%), Intel Arc B570 (+0.5%), and NVIDIA GeForce RTX 3070 Mobile (+0.6%), showing that the AMD card competes in a different performance tier despite its higher average score.

FAQ

Q: Which GPU is faster in the Geekbench Vulkan benchmark?

A: The NVIDIA GeForce RTX 2060 SUPER wins decisively, scoring 77,402 versus the AMD Radeon R9 M390X's 16,951, a delta of -78.1%.

Q: Does the AMD Radeon R9 M390X have a higher average benchmark score than the RTX 2060 SUPER?

A: Yes. The AMD card averages 20,662 across its recorded benchmarks, while the RTX 2060 SUPER averages 18,093. However, the NVIDIA card's average includes ten tests versus the AMD card's two, and the AMD card's score comes entirely from Metal and Vulkan tests.

Q: What is the RTX 2060 SUPER's best individual benchmark result?

A: Its Geekbench Vulkan score of 77,402 is the highest, followed closely by its Geekbench OpenCL score of 76,957. Its Passmark G3D score is 16,462.

Q: What is the AMD Radeon R9 M390X's best individual benchmark result?

A: Its Geekbench Metal score of 24,373 is its top score. Its Geekbench Vulkan score is 16,951, which is lower than its Metal score.

Q: How do the two GPUs compare in percentile ranking?

A: The AMD Radeon R9 M390X ranks at the 66th percentile of all GPUs, while the NVIDIA GeForce RTX 2060 SUPER ranks at the 62nd percentile.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce RTX 2060 SUPER has 2,176 shading units, while the AMD Radeon R9 M390X has 2,048.

Where Each One Wins

The AMD Radeon R9 M390X wins on aggregate benchmark average, with 20,662 versus 18,093. This advantage is driven by its two-test profile, which consists of Geekbench Metal (24,373) and Geekbench Vulkan (16,951). The Metal score is particularly strong, indicating that the AMD card is well-suited for Apple ecosystem workloads where Metal is the primary API. Its higher percentile ranking (66th versus 62nd) also suggests that despite the NVIDIA card's raw compute capability, the AMD card sits in a higher overall performance bracket when measured against all GPUs.

The NVIDIA GeForce RTX 2060 SUPER wins on raw compute throughput in Vulkan, OpenCL, and DirectX workloads. Its Vulkan score of 77,402 is more than four times the AMD card's Vulkan score. Its OpenCL score of 76,957 is similarly dominant. For users running modern cross-platform APIs like Vulkan, or compute-heavy tasks through OpenCL, the RTX 2060 SUPER is clearly the superior choice. The NVIDIA card also supports DirectX 12 Ultimate (12_2), while the AMD card only reaches DirectX 12 (12_0), which matters for games and applications that use the latest DirectX 12 features.

The AMD card wins on power efficiency and form factor. Its TDP is 75 W versus the RTX 2060 SUPER's 175 W, and it uses an MXM module with no power connectors, making it suitable for compact portable systems. The RTX 2060 SUPER is a dual-slot card requiring a single 8-pin power connector and a 450 W power supply. For a mobile or small-form-factor system, the AMD card is the practical choice despite its lower performance.

The NVIDIA card wins on memory capacity and bandwidth. It has 8 GB of GDDR6 at 448.0 GB/s, versus the AMD card's 4 GB of GDDR5 at 160.0 GB/s. This matters for texture-heavy workloads and high-resolution gaming, where the larger and faster memory pool avoids bandwidth bottlenecks.

Specification Differences

The two GPUs differ across nearly every core specification. The NVIDIA GeForce RTX 2060 SUPER has 2,176 shading units, 136 texture mapping units, and 64 render output units. The AMD Radeon R9 M390X has 2,048 shading units, 128 texture mapping units, and 32 render output units. The RTX 2060 SUPER also includes 34 ray tracing cores and 272 tensor cores, while the AMD card has no equivalent hardware.

Memory configuration differs substantially. The RTX 2060 SUPER uses 8 GB of GDDR6 on a 256-bit bus, achieving 448.0 GB/s of bandwidth. The AMD card uses 4 GB of GDDR5 on a 256-bit bus, achieving 160.0 GB/s. The NVIDIA card's memory clock is 1750 MHz (14 Gbps effective), while the AMD card's memory clock is 1250 MHz (5 Gbps effective).

Clock speeds also differ. The RTX 2060 SUPER has a base clock of 1470 MHz and a boost clock of 1650 MHz. The AMD card has no listed base or boost clock in the database.

Pixel and texture fill rates are much higher on the NVIDIA card: 105.6 GPixel/s and 224.4 GTexel/s versus the AMD card's 23.14 GPixel/s and 92.54 GTexel/s. The RTX 2060 SUPER's FP32 throughput is 7.181 TFLOPS and FP16 throughput is 14.36 TFLOPS (2:1 ratio), while the AMD card delivers 2.961 TFLOPS for both FP32 and FP16 (1:1 ratio).

Power requirements differ sharply. The RTX 2060 SUPER has a TDP of 175 W, uses a dual-slot cooler, requires one 8-pin power connector, and suggests a 450 W power supply. The AMD card has a TDP of 75 W, uses an MXM module, and requires no power connectors.

Physical dimensions are only listed for the NVIDIA card: 229 mm length, 113 mm height, 35 mm width. Display outputs also differ: the RTX 2060 SUPER offers 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C, while the AMD card's outputs are listed as portable device dependent.

The RTX 2060 SUPER supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the AMD card supports DirectX 12 (12_0) and Vulkan 1.2.170. Both support OpenGL 4.6.

Architecture Differences

The AMD Radeon R9 M390X is built on GCN 3.0 architecture using the Amethyst chip, manufactured on TSMC's 28 nm process. The die measures 366 mm² and contains 5,000 million transistors, resulting in a transistor density of 13.7 million per square millimeter. The NVIDIA GeForce RTX 2060 SUPER uses the Turing architecture with the TU106 chip, fabricated on TSMC's 12 nm process. Its die is 445 mm² and contains 10,800 million transistors, yielding a density of 24.3 million per square millimeter.

The process node difference is significant: 28 nm versus 12 nm, which explains the NVIDIA card's higher transistor count and density despite the larger die. The RTX 2060 SUPER's Turing architecture introduces dedicated ray tracing cores (34) and tensor cores (272), which are absent from the older GCN 3.0 design. These features enable hardware-accelerated ray tracing and AI-based operations, which the AMD card cannot perform in hardware.

The FP16 compute ratio also differs: the AMD card computes FP16 at 1:1 with FP32 (both 2.961 TFLOPS), while the NVIDIA card computes FP16 at 2:1 (14.36 TFLOPS versus 7.181 TFLOPS FP32). The NVIDIA card's Turing architecture is designed for more efficient mixed-precision workloads, while the AMD card's GCN 3.0 treats FP16 and FP32 equally.

The release timeline shows the AMD card launched on 2015-05-04 as part of the Gem System (R9 M300) generation, with its predecessor being Solar System and successor Polaris Mobile. The NVIDIA card launched on 2019-07-08 in the GeForce 20-series, with predecessor GeForce 10 and successor GeForce 30. The AMD card is end-of-life, as is the NVIDIA card. The RTX 2060 SUPER's launch MSRP is 399 USD.