AMD Radeon R9 M395X vs NVIDIA GeForce RTX 2080 Comparison

The AMD Radeon R9 M395X and NVIDIA GeForce RTX 2080 represent two very different eras of GPU design, and the recorded data reflects a wide performance gulf despite both cards being classified as end-of-life products. The R9 M395X is a mobile-focused MXM module from 2015 built on GCN 3.0, while the RTX 2080 is a dual-slot desktop card from 2018 built on Turing. Their average benchmark scores, 25891 for the AMD part and 22895 for the NVIDIA part, are surprisingly close given the generational gap, but that figure is heavily influenced by the limited set of tests recorded for the AMD card. The only direct head-to-head test in the database, Geekbench OpenCL, shows the RTX 2080 scoring 91313 against the R9 M395X’s 17829, a difference of 80.5 percent. The data suggests that the RTX 2080 is in a different performance class, but the R9 M395X still holds relevance in certain legacy or compute contexts.

The Verdict

The data points to a clear winner for nearly all modern workloads: the NVIDIA GeForce RTX 2080. In the sole head-to-head benchmark, Geekbench OpenCL, the RTX 2080 delivers 91313 points versus the R9 M395X’s 17829, a gap of 80.5 percent. That is not a marginal lead; it is a dominant margin that reflects fundamental architectural and specification differences. Anyone choosing between these two for gaming, ray tracing, or high-throughput compute should select the RTX 2080 without hesitation, based solely on the recorded metrics.

The R9 M395X, however, is not without its own niche. Its average benchmark score of 25891 places it in the 71st percentile of all GPUs in the database, which is actually higher than the RTX 2080’s 68th percentile. This is a curious inversion: the older, slower card ranks higher in percentile despite losing the head-to-head test. The explanation lies in the benchmark set. The R9 M395X has only two recorded tests, Geekbench Metal at 33953 and Geekbench OpenCL at 17829, and both are compute-oriented. The RTX 2080 has ten recorded tests, including lower-scoring DirectX 9 and DirectX 10 results, which drag its average down. The percentile figures are therefore not directly comparable across different test suites, and the head-to-head result is the more reliable indicator of relative performance.

For users who need a low-power mobile GPU with 8 GB of GDDR5 memory and a 75 W TDP, the R9 M395X may still serve a purpose in legacy laptops or specialized portable systems. Its MXM module form factor and lack of power connectors make it a drop-in upgrade for certain notebooks. But for anyone building or upgrading a desktop system, the RTX 2080 is the only sensible choice from this comparison. The data shows no scenario where the R9 M395X wins a recorded benchmark, and its compute performance is an order of magnitude lower in the shared test.

Architecture Differences

The architectural gap between these two GPUs is vast. The AMD Radeon R9 M395X uses the Amethyst chip based on GCN 3.0 architecture, fabricated on a 28 nm process at TSMC. It packs 5,000 million transistors into a 366 mm² die, yielding a transistor density of 13.7 million per square millimeter. The NVIDIA GeForce RTX 2080 uses the TU104 chip based on Turing architecture, built on a 12 nm process, also at TSMC. It contains 13,600 million transistors on a 545 mm² die, with a density of 25.0 million per square millimeter. That means the RTX 2080 crams roughly 2.7 times as many transistors into a die that is only about 1.5 times larger.

The R9 M395X has 2048 shading units, 128 texture mapping units, and 32 render output units. It has no dedicated ray tracing cores and no tensor cores, as those concepts did not exist in GCN 3.0. The RTX 2080, by contrast, has 2944 shading units, 184 TMUs, 64 ROPs, 46 ray tracing cores, and 368 tensor cores. The Turing architecture also supports DirectX 12 Ultimate (12_2), while the older AMD card only reaches DirectX 12 (12_0). Vulkan support differs as well: the RTX 2080 supports Vulkan 1.4, whereas the R9 M395X is limited to Vulkan 1.2.170. Both cards support OpenGL 4.6.

The memory subsystems reflect the generational leap. The R9 M395X uses 8 GB of GDDR5 on a 256-bit bus, delivering 160.0 GB/s of bandwidth. The RTX 2080 also has 8 GB, but uses GDDR6 on the same 256-bit bus, achieving 448.0 GB/s. That is 2.8 times the bandwidth, which directly impacts texture-heavy workloads and high-resolution rendering. The RTX 2080’s memory clock is listed at 1750 MHz with an effective data rate of 14 Gbps, versus 1250 MHz and 5 Gbps effective for the R9 M395X. These raw specifications explain why the RTX 2080 dominates in compute and graphics tests, even before considering its AI and ray tracing hardware.

Head-to-Head Benchmarks

The database records only one direct comparison between these two GPUs: Geekbench OpenCL. The results are stark. The NVIDIA GeForce RTX 2080 scores 91313, while the AMD Radeon R9 M395X scores 17829. The delta is reported as negative 80.5 percent for the AMD card, meaning it trails by that margin. In absolute terms, the RTX 2080 is roughly 5.1 times faster in this particular compute workload. OpenCL is a general-purpose compute API, and the result reflects the RTX 2080’s far higher FP32 throughput of 10.07 TFLOPS versus 2.961 TFLOPS for the R9 M395X. The RTX 2080 also has a major advantage in FP16 performance, delivering 20.14 TFLOPS with a 2:1 ratio, while the AMD card manages only 2.961 TFLOPS with a 1:1 ratio. For any workload that can use half-precision math, the NVIDIA card is in another league.

Beyond the shared test, the RTX 2080’s other recorded benchmarks provide context for its strengths. Its PassMark G3D score is 18720, and its PassMark GPU compute score is 7872. The R9 M395X has no corresponding PassMark entries, so those cannot be compared directly. The RTX 2080 also shows a Geekbench Vulkan score of 107797, which is its highest recorded result, and a Geekbench OpenCL score of 91313. The R9 M395X’s best result is its Geekbench Metal score of 33953, a test the RTX 2080 does not have recorded. Metal is Apple’s graphics API, and the R9 M395X appears to have been tested in a macOS or similar environment, while the RTX 2080 was not. This test-suite asymmetry is important when interpreting the average benchmark scores, because the two cards are not measured across the same set of workloads.

The RTX 2080’s 3DMark Steel Nomad DX12 score of 1752 is its only modern gaming-oriented result. The R9 M395X has no 3DMark entry in the database. For gaming performance, the data is limited to the OpenCL head-to-head and the RTX 2080’s DirectX PassMark results, which include 223 in DirectX 9, 158 in DirectX 11, 136 in DirectX 10, and 72 in DirectX 12. These PassMark scores are not necessarily comparable across API versions, but they do indicate that the RTX 2080 handles legacy DirectX workloads adequately. The absence of any gaming benchmark for the R9 M395X means that its gaming capability must be inferred from its compute scores, which are far below the NVIDIA card.

Specification Differences

The recorded specifications reveal a long list of differences between the two GPUs. The R9 M395X has a 28 nm process node, while the RTX 2080 uses 12 nm. Transistor counts are 5,000 million versus 13,600 million, and die sizes are 366 mm² versus 545 mm². Transistor density is 13.7 million per square millimeter for AMD and 25.0 million for NVIDIA. The R9 M395X has no base or boost clock listed, while the RTX 2080 has a base clock of 1515 MHz and a boost clock of 1710 MHz. Memory clocks are 1250 MHz with 5 Gbps effective for the AMD card, versus 1750 MHz with 14 Gbps effective for the NVIDIA card.

The memory type differs: GDDR5 for the R9 M395X, GDDR6 for the RTX 2080. Memory bandwidth is 160.0 GB/s versus 448.0 GB/s. Shading units are 2048 versus 2944, TMUs are 128 versus 184, and ROPs are 32 versus 64. The RTX 2080 has 46 ray tracing cores and 368 tensor cores, while the R9 M395X has none. Pixel rates are 23.14 GPixel/s for AMD and 109.4 GPixel/s for NVIDIA. Texture rates are 92.54 GTexel/s versus 314.6 GTexel/s. FP32 performance is 2.961 TFLOPS versus 10.07 TFLOPS, and FP16 is 2.961 TFLOPS with a 1:1 ratio versus 20.14 TFLOPS with a 2:1 ratio.

Power consumption differs dramatically: the R9 M395X has a TDP of 75 W, while the RTX 2080 draws 215 W. The AMD card is an MXM module with no power connectors, while the NVIDIA card is a dual-slot design requiring one 6-pin and one 8-pin connector, with a suggested PSU of 550 W. The RTX 2080 has physical dimensions of 267 mm in length, 116 mm in height, and 35 mm in width. The R9 M395X has no dimensions recorded. Display outputs are “portable device dependent” for the AMD card, while the RTX 2080 offers one HDMI 2.0, three DisplayPort 1.4a, and one USB Type-C. Both cards use PCIe 3.0 x16. DirectX support is 12 (12_0) for AMD and 12 Ultimate (12_2) for NVIDIA. OpenGL is 4.6 for both, and Vulkan is 1.2.170 versus 1.4.

Release dates are 2015-05-04 for the R9 M395X and 2018-09-19 for the RTX 2080. The R9 M395X’s predecessor is listed as “Solar System” and its successor as “Polaris Mobile.” The RTX 2080’s predecessor is “GeForce 10” and its successor is “GeForce 30.” The RTX 2080 has a launch MSRP of 699 USD, while the R9 M395X has none recorded.

FAQ

Q: Which GPU is faster in the only benchmark where both were tested?

A: The NVIDIA GeForce RTX 2080 scores 91313 in Geekbench OpenCL, while the AMD Radeon R9 M395X scores 17829. The RTX 2080 leads by 80.5 percent.

Q: Why does the R9 M395X have a higher percentile rank than the RTX 2080?

A: The R9 M395X sits in the 71st percentile of all GPUs, while the RTX 2080 is in the 68th. This is because the AMD card has only two recorded benchmarks, both compute-heavy, whereas the RTX 2080 has ten recorded tests including lower-scoring legacy DirectX results that lower its average.

Q: Do both cards have the same amount of memory?

A: Yes, both have 8 GB, but the memory types differ. The R9 M395X uses GDDR5 with 160.0 GB/s bandwidth, while the RTX 2080 uses GDDR6 with 448.0 GB/s bandwidth.

Q: Does the R9 M395X support ray tracing?

A: No. The R9 M395X has no ray tracing cores, while the RTX 2080 has 46 ray tracing cores and 368 tensor cores.

Q: What is the power consumption difference?

A: The R9 M395X has a TDP of 75 W and requires no power connectors, while the RTX 2080 has a TDP of 215 W and needs one 6-pin and one 8-pin connector, with a suggested PSU of 550 W.

Q: Which card supports a newer version of DirectX?

A: The RTX 2080 supports DirectX 12 Ultimate (12_2), while the R9 M395X is limited to DirectX 12 (12_0). The RTX 2080 also supports Vulkan 1.4 versus Vulkan 1.2.170 for the AMD card.

Where Each One Wins

The RTX 2080 wins in every measurable compute scenario. Its Geekbench OpenCL score of 91313 dwarfs the R9 M395X’s 17829, and its FP32 throughput of 10.07 TFLOPS is more than three times the AMD card’s 2.961 TFLOPS. In FP16 workloads, the advantage grows even larger: 20.14 TFLOPS versus 2.961 TFLOPS. The RTX 2080’s 448.0 GB/s of memory bandwidth, 109.4 GPixel/s pixel rate, and 314.6 GTexel/s texture rate make it the clear choice for modern gaming, ray tracing, AI inference with its 368 tensor cores, and high-resolution rendering. Its 46 ray tracing cores enable hardware-accelerated ray tracing, a feature entirely absent from the R9 M395X. The RTX 2080 also has a higher shading unit count, 2944 versus 2048, and more TMUs and ROPs, which translate to better raw geometry and fill-rate performance.

The R9 M395X wins in exactly one category: power efficiency. Its 75 W TDP is less than a third of the RTX 2080’s 215 W, and it requires no external power connectors. For portable or low-power systems, the MXM module form factor is an advantage, as it allows the card to be installed in laptops or compact devices where a dual-slot desktop card cannot fit. The R9 M395X’s Geekbench Metal score of 33953 suggests it performs reasonably in Apple’s Metal API, which may be relevant for users in macOS environments. Its 8 GB of GDDR5 memory is equal in capacity to the RTX 2080, so memory size is not a differentiator. The R9 M395X also has a higher percentile rank, 71st versus 68th, but that is an artifact of the different benchmark sets rather than a genuine performance advantage.

For users who prioritize raw performance, modern API support, ray tracing, or high-bandwidth memory, the RTX 2080 is the only option. For users who need a low-power, connector-free mobile GPU with sufficient memory capacity for legacy tasks, the R9 M395X remains a viable niche product. The data does not support any other conclusion.