AMD Radeon R9 M390X vs NVIDIA GeForce RTX 2080 Comparison

Head-to-Head Benchmarks

The database records only one shared benchmark between the NVIDIA GeForce RTX 2080 and the AMD Radeon R9 M390X, but the result is decisive. In the Geekbench Vulkan test, the RTX 2080 scores 107,797, while the R9 M390X manages 16,951. That is a delta of 535.9% in favor of the NVIDIA card, a gap so wide it places the two products in entirely different performance tiers. The RTX 2080 does not merely outperform its older rival; it outperforms it by a factor of more than six.

The wider benchmark database reinforces this picture. The RTX 2080 posts an average benchmark score of 22,895 across all recorded tests, placing it at the 68th percentile of all GPUs tracked. Its nearest rivals in the database include the Intel Arc B580 at 23,021 (0.5% behind the RTX 2080), the AMD Radeon RX 580 2048SP at 23,061 (0.7% behind), and the NVIDIA GeForce RTX 3080 at 23,172 (1.2% behind). The RTX 2080 also edges out the RTX 4060 Mobile, which scores 22,729, a 0.7% deficit. These are tight margins, suggesting the RTX 2080 sits in a crowded performance band where a few percentage points separate closely matched cards.

The R9 M390X, by contrast, averages 20,662 across its recorded benchmarks, good for the 66th percentile of all GPUs. Its nearest rivals cluster around the same figure: the AMD Radeon RX 5600 XT at 20,713 (0.2% ahead of the R9 M390X), the Intel Arc A750 at 20,582 (0.4% behind), the Intel Arc B570 at 20,556 (0.5% behind), and the NVIDIA GeForce RTX 3070 Mobile at 20,534 (0.6% behind). The percentile gap between the two cards is small, just 2 points, but the raw average score difference is 2,233 points, roughly 10.8% separating the two averages. In the single directly comparable workload, however, the RTX 2080's advantage is not incremental; it is overwhelming.

The RTX 2080 also has a much richer benchmark record. The database lists ten separate tests for it, spanning 3DMark Steel Nomad DX12 (1,752), Geekbench OpenCL (91,313), Geekbench Vulkan (107,797), and a series of Passmark tests covering DirectX 9 through DirectX 12, G2D, G3D, and GPU compute. Its Passmark G3D score is 18,720, and its compute score is 7,872. The R9 M390X has only two recorded benchmarks, Geekbench Metal (24,373) and Geekbench Vulkan (16,951). That limited record makes direct comparison difficult outside the Vulkan result, but the Vulkan data alone establishes the performance hierarchy.

Architecture Differences

The two GPUs come from different eras of GPU design. The RTX 2080 uses the TU104 chip built on TSMC's 12 nm process, while the R9 M390X uses the Amethyst chip on TSMC's 28 nm process. The process node difference is significant: 12 nm versus 28 nm. The RTX 2080 packs 13,600 million transistors onto a 545 mm² die, yielding a transistor density of 25.0 million per mm². The R9 M390X contains 5,000 million transistors on a 366 mm² die, a density of 13.7 million per mm². The RTX 2080 therefore integrates nearly three times the transistor count on a die that is less than 50% larger.

The architectures themselves reflect different priorities. The RTX 2080 is built on NVIDIA's Turing architecture, part of the GeForce 20 generation. Turing introduced dedicated hardware features that the R9 M390X, based on AMD's GCN 3.0 architecture, simply does not have. The RTX 2080 includes 46 ray tracing cores and 368 tensor cores, hardware units designed for real-time ray tracing and AI-accelerated workloads. The R9 M390X has no ray tracing cores and no tensor cores; its GCN 3.0 design predates those features.

Compute resources differ substantially. The RTX 2080 has 2,944 shading units, 184 texture mapping units, and 64 raster operation units. The R9 M390X has 2,048 shading units, 128 TMUs, and 32 ROPs. The RTX 2080 also delivers dramatically higher fill rates: 109.4 GPixel/s pixel rate and 314.6 GTexel/s texture rate, compared to 23.14 GPixel/s and 92.54 GTexel/s for the R9 M390X. Floating-point throughput tells a similar story. The RTX 2080 reaches 10.07 TFLOPS FP32 and 20.14 TFLOPS FP16 with a 2:1 ratio. The R9 M390X manages 2.961 TFLOPS FP32 and the same 2.961 TFLOPS FP16, a 1:1 ratio that indicates no dedicated half-precision acceleration.

Memory architecture also separates the two. The RTX 2080 uses 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s of bandwidth. The R9 M390X uses 4 GB of GDDR5 on a 256-bit bus, with 160.0 GB/s of bandwidth. The RTX 2080's memory clock is listed as 1750 MHz with 14 Gbps effective data rate; the R9 M390X runs at 1250 MHz with 5 Gbps effective. The RTX 2080 has nearly three times the memory bandwidth and twice the capacity.

API support follows the generational split. The RTX 2080 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. The RTX 2080's DirectX 12 Ultimate feature set includes capabilities the older card cannot expose.

The Verdict

The data supports a clear conclusion: the NVIDIA GeForce RTX 2080 is the superior GPU in every measurable category. Its average benchmark score is 2,233 points higher than the R9 M390X's, and in the one directly comparable workload, Geekbench Vulkan, it leads by 535.9%. It has more shading units, more TMUs, double the ROPs, more than triple the memory bandwidth, and roughly 3.4 times the FP32 throughput. It also adds hardware ray tracing and tensor cores, features the R9 M390X lacks entirely.

The RTX 2080 is for anyone who needs modern feature support and high-end compute performance. Its DirectX 12 Ultimate support, Vulkan 1.4, and dedicated RT and tensor hardware make it viable for contemporary workloads. The R9 M390X, with its GCN 3.0 architecture, DirectX 12 (12_0), and Vulkan 1.2.170, is a legacy part. Its 66th percentile standing shows it still outperforms a large portion of the database, and its nearest rivals include capable cards like the RX 5600 XT and Arc A750, but it cannot compete with the RTX 2080 in any shared workload.

Pick the RTX 2080 if you want the higher average score, the larger memory pool, the faster memory subsystem, and the modern API feature set. Pick the R9 M390X only if the workload is specifically tuned for its GCN architecture or if the system requires an MXM module form factor and a 75 W power envelope, since the RTX 2080 is a dual-slot card with a 215 W TDP and requires a 550 W power supply.

Specification Differences

The two cards differ in nearly every specification field. The RTX 2080 uses the TU104 chip on a 12 nm process; the R9 M390X uses Amethyst on 28 nm. Transistor count is 13,600 million versus 5,000 million, and die size is 545 mm² versus 366 mm². Transistor density is 25.0M per mm² versus 13.7M per mm².

The RTX 2080 has a base clock of 1515 MHz and a boost clock of 1710 MHz; the R9 M390X has no recorded base or boost clock. Memory clocks are 1750 MHz with 14 Gbps effective for the RTX 2080, versus 1250 MHz with 5 Gbps effective for the R9 M390X. Memory capacity is 8 GB GDDR6 versus 4 GB GDDR5, both on 256-bit buses, but bandwidth is 448.0 GB/s versus 160.0 GB/s.

Shader resources: 2,944 shading units, 184 TMUs, and 64 ROPs for the RTX 2080; 2,048 shading units, 128 TMUs, and 32 ROPs for the R9 M390X. The RTX 2080 has 46 RT cores and 368 tensor cores; the R9 M390X has none. Pixel rate is 109.4 GPixel/s versus 23.14 GPixel/s. Texture rate is 314.6 GTexel/s versus 92.54 GTexel/s. FP32 is 10.07 TFLOPS versus 2.961 TFLOPS. FP16 is 20.14 TFLOPS (2:1) versus 2.961 TFLOPS (1:1).

Power and physical specs also differ. The RTX 2080 has a 215 W TDP, dual-slot width, 1x 6-pin plus 1x 8-pin power connectors, and a suggested 550 W PSU. The R9 M390X has a 75 W TDP, MXM Module slot width, no power connectors, and no suggested PSU. The RTX 2080 measures 267 mm in length, 116 mm in height, and 35 mm in width; the R9 M390X has no recorded dimensions. Display outputs are 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C for the RTX 2080, versus "Portable Device Dependent" for the R9 M390X.

Release dates are 2018-09-19 for the RTX 2080 and 2015-05-04 for the R9 M390X. Both are end-of-life. The RTX 2080's launch MSRP is 699 USD. The R9 M390X has no recorded launch MSRP.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce RTX 2080 averages 22,895 across all recorded tests, while the AMD Radeon R9 M390X averages 20,662.

Q: How large is the performance gap in the shared Vulkan benchmark?

A: In Geekbench Vulkan, the RTX 2080 scores 107,797 versus 16,951 for the R9 M390X, a difference of 535.9%.

Q: Does the R9 M390X support ray tracing?

A: No. The R9 M390X has no ray tracing cores. The RTX 2080 includes 46 RT cores.

Q: What are the memory specifications of each card?

A: The RTX 2080 has 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth. The R9 M390X has 4 GB of GDDR5 on a 256-bit bus with 160.0 GB/s bandwidth.

Q: Which card has higher FP32 throughput?

A: The RTX 2080 delivers 10.07 TFLOPS FP32, compared to 2.961 TFLOPS for the R9 M390X.

Q: What API levels does each card support?

A: The RTX 2080 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.