AMD Radeon R9 M290X vs NVIDIA GeForce RTX 2070 SUPER Comparison

AMD Radeon R9 M290X vs NVIDIA GeForce RTX 2070 SUPER

The database shows two graphics cards from different eras: the AMD Radeon R9 M290X, a mobile-focused part from early 2014 built on 28 nm GCN 1.0, and the NVIDIA GeForce RTX 2070 SUPER, a desktop Turing card from mid-2019 on 12 nm. Their recorded benchmark averages differ significantly: the R9 M290X posts an average score of 23,276 across available tests, while the RTX 2070 SUPER averages 20,282. However, this comparison is skewed by the fact that the AMD card has only two benchmark entries (Geekbench Metal and OpenCL), whereas the NVIDIA card has ten entries spanning DirectX, Vulkan, and compute workloads. The single head-to-head test, Geekbench OpenCL, shows the RTX 2070 SUPER scoring 83,358 versus the R9 M290X’s 22,028, a delta of 73.6% in favor of NVIDIA. The RTX 2070 SUPER also sits in the 65th percentile of all GPUs, while the R9 M290X is in the 68th percentile, meaning the AMD part’s average is boosted by its narrow test set.

FAQ

Q: Which card has higher raw compute performance in the shared benchmark?

A: The NVIDIA GeForce RTX 2070 SUPER wins the only common test, Geekbench OpenCL, with a score of 83,358 against the AMD Radeon R9 M290X’s 22,028. That is a 73.6% advantage for NVIDIA.

Q: How do their memory subsystems compare?

A: The R9 M290X has 4 GB of GDDR5 on a 256-bit bus, delivering 153.6 GB/s. The RTX 2070 SUPER has 8 GB of GDDR6 on the same 256-bit bus, delivering 448.0 GB/s. The NVIDIA card offers double the capacity and nearly triple the bandwidth.

Q: Which card supports hardware ray tracing?

A: Only the RTX 2070 SUPER has RT cores (40 of them) and tensor cores (320). The R9 M290X has neither, as its GCN 1.0 architecture predates dedicated ray tracing hardware.

Q: What is the power draw difference?

A: The R9 M290X is rated at 100 W TDP, while the RTX 2070 SUPER is rated at 215 W TDP. The NVIDIA card also requires a 550 W suggested power supply and uses 1x 6-pin plus 1x 8-pin connectors, whereas the AMD card uses no external power connectors and comes as an MXM module.

Q: Are there any DirectX feature level differences?

A: Yes. The RTX 2070 SUPER supports DirectX 12 Ultimate (12_2), while the R9 M290X supports DirectX 12 (11_1). The NVIDIA card also supports Vulkan 1.4, compared to Vulkan 1.2.170 on the AMD card.

Q: Which card is physically larger?

A: The RTX 2070 SUPER is a dual-slot card measuring 267 mm in length, 116 mm in height, and 35 mm in width. The R9 M290X is an MXM module with dimensions listed as portable-device dependent, so no fixed length or height is recorded.

Where Each One Wins

The R9 M290X’s strengths are confined to its narrow benchmark set. Its Geekbench Metal score of 24,524 and OpenCL score of 22,028 are the only recorded results, and both are far below the RTX 2070 SUPER’s OpenCL score. In the database’s percentile ranking, the R9 M290X sits at 68th percentile, slightly above the RTX 2070 SUPER’s 65th percentile, but this is misleading because the AMD card’s average (23,276) comes from just two tests, while NVIDIA’s average (20,282) includes many lower-scoring legacy DirectX tests. The R9 M290X wins no head-to-head tests against the RTX 2070 SUPER in the database.

The RTX 2070 SUPER wins every category that matters for modern workloads. Its Geekbench Vulkan score of 90,637 and OpenCL score of 83,358 dwarf the AMD card’s compute results. In PassMark’s G3D test, the RTX 2070 SUPER scores 18,169, and in GPU Compute it scores 7,557. Even its DirectX 9 score of 223 and DirectX 10 score of 132, while low in absolute terms, reflect a driver and architecture that handles legacy APIs better than the R9 M290X would, based on the AMD card’s lack of DirectX 12 Ultimate support. For any workload that uses Vulkan, OpenCL, or modern DirectX 12, the RTX 2070 SUPER is the clear winner.

Architecture Differences

The R9 M290X uses the Neptune chip with GCN 1.0 architecture, manufactured on TSMC’s 28 nm process. It contains 2,800 million transistors on a 212 mm² die, giving a transistor density of 13.2 million per mm². The RTX 2070 SUPER uses the TU104 chip with Turing architecture, on TSMC’s 12 nm process, packing 13,600 million transistors into a 545 mm² die, for a density of 25.0 million per mm². The NVIDIA chip has nearly five times the transistor count and almost twice the density.

The R9 M290X has 1,280 shading units, 80 texture mapping units, and 32 render output units. The RTX 2070 SUPER doubles those counts: 2,560 shading units, 160 TMUs, and 64 ROPs. Beyond raw counts, the RTX 2070 SUPER adds 40 RT cores for ray tracing and 320 tensor cores for AI acceleration, neither of which exists on the GCN 1.0 part. Clock speeds also differ: the AMD card runs at 850 MHz base and 900 MHz boost, while the NVIDIA card runs at 1,605 MHz base and 1,770 MHz boost. The NVIDIA card’s higher clocks and wider execution resources produce pixel rates of 113.3 GPixel/s and texture rates of 283.2 GTexel/s, versus 28.80 GPixel/s and 72.00 GTexel/s on the AMD card.

Memory architecture diverges as well. The R9 M290X uses 4 GB GDDR5 at 1,200 MHz (4.8 Gbps effective), while the RTX 2070 SUPER uses 8 GB GDDR6 at 1,750 MHz (14 Gbps effective). Both use 256-bit buses, but the NVIDIA card achieves 448.0 GB/s bandwidth versus 153.6 GB/s. The RTX 2070 SUPER also supports FP16 at 18.12 TFLOPS (2:1 ratio), while the R9 M290X has no recorded FP16 capability. Its FP32 output is 2.304 TFLOPS, compared to 9.062 TFLOPS on the NVIDIA card.

Specification Differences

The table below highlights only the fields where the two cards differ according to the database.

| Specification | AMD Radeon R9 M290X | NVIDIA GeForce RTX 2070 SUPER |

|---|---|---|

| Process node | 28 nm | 12 nm |

| Transistors | 2,800 million | 13,600 million |

| Die size | 212 mm² | 545 mm² |

| Transistor density | 13.2M / mm² | 25.0M / mm² |

| Base clock | 850 MHz | 1,605 MHz |

| Boost clock | 900 MHz | 1,770 MHz |

| Memory clock | 1,200 MHz, 4.8 Gbps effective | 1,750 MHz, 14 Gbps effective |

| Memory size | 4 GB | 8 GB |

| Memory type | GDDR5 | GDDR6 |

| Memory bandwidth | 153.6 GB/s | 448.0 GB/s |

| Shading units | 1,280 | 2,560 |

| TMUs | 80 | 160 |

| ROPs | 32 | 64 |

| RT cores | None | 40 |

| Tensor cores | None | 320 |

| Pixel rate | 28.80 GPixel/s | 113.3 GPixel/s |

| Texture rate | 72.00 GTexel/s | 283.2 GTexel/s |

| FP32 | 2.304 TFLOPS | 9.062 TFLOPS |

| FP16 | Not recorded | 18.12 TFLOPS (2:1) |

| TDP | 100 W | 215 W |

| Slot width | MXM Module | Dual-slot |

| Power connectors | None | 1x 6-pin + 1x 8-pin |

| Suggested PSU | Not recorded | 550 W |

| Display outputs | Portable Device Dependent | 1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C |

| DirectX support | 12 (11_1) | 12 Ultimate (12_2) |

| Vulkan support | 1.2.170 | 1.4 |

| Release date | 2014-01-08 | 2019-07-08 |

| Dimensions | Not recorded | 267 mm, 116 mm, 35 mm |

The RTX 2070 SUPER also has a launch MSRP of 499 USD, while the R9 M290X has no recorded launch price.

Head-to-Head Benchmarks

The database contains only one direct comparison between these two cards: Geekbench OpenCL. The R9 M290X scores 22,028, while the RTX 2070 SUPER scores 83,358. The delta is 73.6% in NVIDIA’s favor. This is the single largest performance gap in the entire dataset, and it reflects the architectural chasm between a 2014 mobile GCN part and a 2019 desktop Turing card.

Beyond that single test, the RTX 2070 SUPER’s other recorded benchmarks provide context. Its Geekbench Vulkan score of 90,637 is even higher than its OpenCL result, suggesting that the card is particularly strong in modern cross-platform APIs. Its PassMark G3D score of 18,169 and GPU Compute score of 7,557 further underscore its compute capabilities. The R9 M290X has no Vulkan score recorded, and its Metal score of 24,524 (which is higher than its OpenCL score of 22,028) indicates that it performs better under Apple’s Metal API than under OpenCL, but still at a level far below the NVIDIA card’s compute output.

The RTX 2070 SUPER’s nearest rivals in the database are the NVIDIA Quadro M4000M (average score 20,480, delta -1%), the RTX 3070 Mobile (20,534, -1.2%), the Intel Arc B570 (20,556, -1.3%), and the Intel Arc A750 (20,582, -1.5%). This shows that the RTX 2070 SUPER sits in a competitive cluster among mid-to-high-end cards, with no rival more than 1.5% away in average score. The R9 M290X’s nearest rivals are the RX 6600M (23,273, 0%), Pro Vega 16 (23,250, 0.1%), P106-100 (23,249, 0.1%), and AI PRO R9700 (23,315, -0.2%). The AMD card’s cluster is also tight, but the absolute scores are lower than the NVIDIA card’s cluster when considering the RTX 2070 SUPER’s OpenCL result.

The Verdict

The data is unambiguous: for any modern workload, the NVIDIA GeForce RTX 2070 SUPER is the superior card. It offers 3.9 times the FP32 throughput (9.062 TFLOPS vs 2.304 TFLOPS), nearly three times the memory bandwidth (448.0 GB/s vs 153.6 GB/s), double the memory capacity, and dedicated RT and tensor cores. Its 73.6% lead in the only head-to-head benchmark is consistent with the rest of its results: Vulkan score of 90,637, OpenCL score of 83,358, and PassMark G3D score of 18,169 are all far beyond what the R9 M290X could produce based on its recorded 22,028 OpenCL score.

The R9 M290X does hold a slight percentile edge (68th vs 65th), but that is an artifact of its limited benchmark set. With only two tests, its average of 23,276 is not comparable to the RTX 2070 SUPER’s average of 20,282 across ten diverse tests. The RTX 2070 SUPER’s DirectX 9, 10, 11, and 12 scores (223, 132, 151, and 67 respectively) drag down its average because they are legacy API tests that run poorly on modern architectures, but those scores are irrelevant to actual gaming or compute workloads.

For a PC builder, the choice is clear: the RTX 2070 SUPER is the only sensible option for gaming, rendering, or compute tasks. The R9 M290X, with its 100 W TDP and MXM form factor, was designed for laptops and has no place in a modern desktop. Its GCN 1.0 architecture lacks hardware ray tracing, tensor cores, and modern DirectX 12 Ultimate support. The RTX 2070 SUPER, despite its higher 215 W TDP and larger physical footprint, delivers a complete feature set that the older AMD part cannot match. The database records no scenario where the R9 M290X wins, and the numerical gap is so large that no driver or workload optimization could close it.