AMD Radeon R9 M390X vs NVIDIA GeForce RTX 3060 Mobile Comparison

FAQ

Q: How does the AMD Radeon R9 M390X compare to the NVIDIA GeForce RTX 3060 Mobile in raw compute performance?

A: The recorded data shows a decisive gap. The R9 M390X delivers 2.961 TFLOPS of FP32 performance, while the RTX 3060 Mobile reaches 10.94 TFLOPS. In the only direct head-to-head benchmark available, Geekbench Vulkan, the RTX 3060 Mobile scores 80,344 versus 16,951 for the R9 M390X, which is a 78.9% advantage for the NVIDIA part.

Q: What is the memory configuration difference between these two GPUs?

A: The Radeon R9 M390X uses 4 GB of GDDR5 on a 256-bit bus, yielding 160.0 GB/s of bandwidth. The RTX 3060 Mobile uses 6 GB of GDDR6 on a 192-bit bus, yielding 336.0 GB/s. Despite the narrower bus, the RTX 3060 Mobile has more than twice the memory bandwidth and 2 GB more capacity.

Q: Which GPU has a higher transistor count and what does that indicate?

A: The RTX 3060 Mobile packs 12,000 million transistors on Samsung's 8 nm process, while the R9 M390X has 5,000 million transistors on TSMC's 28 nm node. The transistor density figures are 43.5M per mm² for the RTX 3060 Mobile versus 13.7M per mm² for the R9 M390X. This reflects the much newer manufacturing technology used for the NVIDIA chip.

Q: Do both GPUs support modern APIs?

A: The RTX 3060 Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The R9 M390X supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.2.170. Both are compatible with current operating systems, but the NVIDIA part has higher API version support, particularly for Vulkan.

Q: What are the average benchmark scores for each GPU?

A: The R9 M390X has an average benchmark score of 20,662, which places it at the 66th percentile of all GPUs in the database. The RTX 3060 Mobile has an average benchmark score of 18,159, placing it at the 62nd percentile. Interestingly, the older AMD part has a higher average score despite losing the head-to-head comparison, because the R9 M390X has more benchmark submissions in the database, including Geekbench Metal and Vulkan tests.

Q: Are these GPUs still in production?

A: Both are marked as end-of-life in the database. The R9 M390X was released in 2015 and its predecessor is the Solar System series, with the successor being Polaris Mobile. The RTX 3060 Mobile was released in 2021 and its predecessor is the GeForce 20 Mobile series. Neither has a successor listed for the NVIDIA side, and the AMD part has been replaced by newer mobile architectures.

The Verdict

The data paints a clear picture for most use cases. The NVIDIA GeForce RTX 3060 Mobile is the stronger performer in the one benchmark where both GPUs were measured head-to-head. Its Geekbench Vulkan score of 80,344 crushes the R9 M390X's 16,951, a margin of 78.9%. This is not a close contest in modern workloads that leverage Vulkan.

However, the average benchmark scores tell a different story. The R9 M390X sits at 20,662 average, which is 12% higher than the RTX 3060 Mobile's 18,159. The R9 M390X also ranks at the 66th percentile versus the 62nd percentile for the NVIDIA part. This suggests that in certain older workloads, particularly those using Metal or legacy APIs, the AMD GPU holds its own. The R9 M390X's Geekbench Metal score of 24,373 is a strong result, though no direct Metal comparison exists for the RTX 3060 Mobile in the database.

For a practical recommendation: if you are running modern games or compute workloads that use Vulkan or DirectX 12 Ultimate, the RTX 3060 Mobile is the clear choice. Its FP32 throughput of 10.94 TFLOPS is nearly four times that of the R9 M390X's 2.961 TFLOPS. The 6 GB VRAM capacity and 336.0 GB/s bandwidth also make it far more suitable for modern game assets and higher resolutions.

The R9 M390X is a legacy part from 2015. It belongs in older laptops where the MXM module form factor is required, and it can still handle lighter workloads or older titles. But for anyone choosing between these two today, the RTX 3060 Mobile is the only sensible pick for gaming or creative work. The R9 M390X would only be relevant for someone maintaining a vintage system or needing a drop-in replacement for a specific MXM chassis.

Head-to-Head Benchmarks

The database contains exactly one direct head-to-head benchmark between these two GPUs: Geekbench Vulkan. In this test, the NVIDIA GeForce RTX 3060 Mobile scores 80,344, while the AMD Radeon R9 M390X scores 16,951. The delta is 78.9% in favor of NVIDIA. This is an enormous gap, and it reflects the architectural differences between a 2015 GCN 3.0 part and a 2021 Ampere chip.

To put this in context, the R9 M390X's nearest rivals in the database include the AMD Radeon RX 5600 XT with an average score of 20,713 (0.2% ahead), the Intel Arc A750 at 20,582 (0.4% behind), and the Intel Arc B570 at 20,556 (0.5% behind). The NVIDIA GeForce RTX 3070 Mobile sits at 20,534, just 0.6% behind the R9 M390X. This cluster of GPUs, all within a narrow band around 20,500 to 20,700, shows that the R9 M390X performs roughly on par with midrange desktop GPUs from several generations later, at least in the aggregate benchmarks used for the average score.

The RTX 3060 Mobile's nearest rivals include the AMD Radeon Pro 5700 at 18,189 (0.2% behind the RTX 3060 Mobile), the NVIDIA GeForce RTX 2060 SUPER at 18,093 (0.4% behind), and the AMD Radeon RX 460 at 18,373 (1.2% ahead). The Intel Arc A770M is 1.2% ahead at 18,383. This places the RTX 3060 Mobile in a similar performance band to desktop GPUs like the RTX 2060 SUPER and the Radeon Pro 5700, which is impressive for a mobile part with an 80 W TDP.

The individual benchmark scores for the RTX 3060 Mobile show a wide spread across different test types. Its Geekbench OpenCL score is 79,483, and its Vulkan score is 80,344. In Passmark tests, it scores 13,230 in G3D, 5,718 in GPU Compute, 588 in G2D, 146 in DirectX 9, 110 in DirectX 11, 90 in DirectX 10, and 58 in DirectX 12. The 3DMark Steel Nomad DX12 score is 1,821. These numbers show that the RTX 3060 Mobile performs best in modern compute and API-heavy workloads, while its older DirectX 9 and 10 performance is comparatively modest.

Specification Differences

The most obvious difference is the manufacturing process. The R9 M390X uses a 28 nm node from TSMC, while the RTX 3060 Mobile uses Samsung's 8 nm process. This explains the transistor density gap: 13.7M per mm² for AMD versus 43.5M per mm² for NVIDIA. The die sizes are 366 mm² for the R9 M390X and 276 mm² for the RTX 3060 Mobile. Despite being physically larger, the AMD chip has fewer than half the transistors.

Memory is another major split. The R9 M390X has 4 GB of GDDR5 on a 256-bit bus with 160.0 GB/s bandwidth. The RTX 3060 Mobile has 6 GB of GDDR6 on a 192-bit bus with 336.0 GB/s bandwidth. The effective memory clock is 5 Gbps for AMD and 14 Gbps for NVIDIA.

The compute units differ substantially. The R9 M390X has 2,048 shading units, 128 TMUs, and 32 ROPs. The RTX 3060 Mobile has 3,840 shading units, 120 TMUs, and 48 ROPs. NVIDIA's part also includes 30 RT cores and 120 tensor cores, which the AMD GPU lacks entirely. Pixel rate is 23.14 GPixel/s for AMD versus 68.40 GPixel/s for NVIDIA. Texture rate is 92.54 GTexel/s versus 171.0 GTexel/s.

Power consumption is close: 75 W for the R9 M390X and 80 W for the RTX 3060 Mobile. Both use no external power connectors. The R9 M390X uses an MXM module slot, while the RTX 3060 Mobile has no listed slot width. Both use PCIe, but the AMD part is PCIe 3.0 x16 while the NVIDIA part is PCIe 4.0 x16.

Clock speeds are not directly comparable because the R9 M390X has no listed base or boost clock, only a memory clock of 1250 MHz. The RTX 3060 Mobile has a 900 MHz base clock and a 1425 MHz boost clock. The API support differs as noted above, with NVIDIA offering DirectX 12 Ultimate and Vulkan 1.4 versus AMD's DirectX 12 (12_0) and Vulkan 1.2.170.

Architecture Differences

The R9 M390X is built on GCN 3.0 architecture, using the Amethyst chip. It belongs to the Gem System generation (R9 M300 series) and was released in 2015. GCN 3.0 was AMD's third iteration of its Graphics Core Next design, which emphasized compute throughput and asynchronous compute capabilities. The architecture supports FP16 at a 1:1 ratio with FP32, meaning both are 2.961 TFLOPS.

The RTX 3060 Mobile uses the Ampere architecture on the GA106 chip. It belongs to the GeForce 30 Mobile generation and was released in 2021. Ampere introduced dedicated RT cores for ray tracing and tensor cores for AI workloads, which the R9 M390X does not have. The FP16 performance is also 1:1 with FP32 at 10.94 TFLOPS. The Ampere architecture brings hardware-accelerated ray tracing, DLSS support, and other modern features that are absent from GCN 3.0.

The R9 M390X's predecessor is the Solar System series, and its successor is Polaris Mobile. The RTX 3060 Mobile's predecessor is the GeForce 20 Mobile series, and no successor is listed. Both are end-of-life products, but the architectural gap is roughly six years of GPU development.

Where Each One Wins

The R9 M390X wins in the aggregate database ranking. Its average benchmark score of 20,662 places it at the 66th percentile, ahead of the RTX 3060 Mobile's 18,159 at the 62nd percentile. This is likely due to the R9 M390X having strong results in Geekbench Metal (24,373) and a solid Vulkan score (16,951). For users running macOS or Metal-based applications, the R9 M390X may be the better option, though no Metal benchmark exists for the RTX 3060 Mobile in the database.

The RTX 3060 Mobile wins decisively in the head-to-head Vulkan test, with 80,344 versus 16,951, a 78.9% advantage. It also dominates in raw FP32 compute (10.94 TFLOPS versus 2.961 TFLOPS), memory bandwidth (336.0 GB/s versus 160.0 GB/s), and pixel fill rate (68.40 GPixel/s versus 23.14 GPixel/s). The 6 GB VRAM capacity gives it more headroom for modern games, while the RT and tensor cores enable ray tracing and AI-accelerated features.

For gaming, the RTX 3060 Mobile is the clear winner. Its Passmark DirectX 11 score of 110 and DirectX 12 score of 58, while not directly comparable to the R9 M390X (which has no Passmark data), show reasonable performance across modern APIs. The 3DMark Steel Nomad DX12 score of 1,821 further confirms its capability in current-generation titles.

For legacy or lightweight workloads, the R9 M390X is adequate. Its 2.961 TFLOPS of FP32 is enough for older games, basic video playback, and casual use. The 4 GB VRAM is sufficient for 1080p gaming at medium settings in older titles. Its 75 W TDP makes it a low-power option for thin laptops, though the RTX 3060 Mobile is only slightly higher at 80 W.

In summary, the RTX 3060 Mobile is the better GPU for anyone running modern software. The R9 M390X is only relevant for niche scenarios involving legacy systems, specific MXM form factor requirements, or Metal-only workloads.