AMD Radeon R9 M290X vs NVIDIA GeForce RTX 3080 Comparison

The Verdict

The data presents a stark generational contrast. The AMD Radeon R9 M290X, a 2014 mobile part built on GCN 1.0, is fundamentally outclassed by the NVIDIA GeForce RTX 3080, a 2020 desktop flagship based on the Ampere architecture. In the single head-to-head benchmark available, the RTX 3080 delivers a decisive victory, scoring 152,423 in Geekbench OpenCL compared to the R9 M290X’s 22,028 — a 85.5% lead for NVIDIA. This is not a close contest; it is a reflection of two products separated by over six years of architectural progress.

For users considering the R9 M290X, the data suggests it belongs in legacy or low-intensity workloads. Its average benchmark score of 23,276 places it in the 68th percentile of all GPUs, essentially tied with the AMD Radeon RX 6600M (23,273) and the NVIDIA P106-100 (23,249). The R9 M290X is a capable performer for its era, but its 2.304 TFLOPS FP32 throughput and 4 GB GDDR5 memory are insufficient for modern high-end gaming or compute tasks. The RTX 3080, by contrast, sits in the same 68th percentile with an average score of 23,172, but this is misleading — the average is pulled down by a wide range of benchmarks, including low DirectX 9 and 10 scores. In compute-heavy OpenCL, the RTX 3080 is in a different league entirely.

The verdict is unambiguous: the RTX 3080 is the superior choice for anyone needing raw performance, modern API support, or high-resolution gaming. The R9 M290X is only relevant for users with specific legacy compatibility needs or those who cannot upgrade their system’s power delivery and cooling infrastructure. The RTX 3080’s launch MSRP is 699 USD, and its 320 W TDP demands a 700 W suggested PSU, making it a heavy investment. The R9 M290X, with a 100 W TDP and no external power connectors, is a far more modest proposition, but its performance ceiling is correspondingly low.

---

Architecture Differences

The architectural gap between these two GPUs is immense, reflecting two distinct design philosophies. The AMD Radeon R9 M290X uses the Neptune chip built on GCN 1.0 architecture, fabricated on a 28 nm process at TSMC. It packs 2,800 million transistors into a 212 mm² die, yielding a transistor density of 13.2M / mm². This is a mobile-first design with a 100 W TDP, delivered as an MXM Module with no power connectors. Its compute core consists of 1,280 shading units, 80 texture mapping units (TMUs), and 32 raster output units (ROPs). Clock speeds are modest, with a 850 MHz base and 900 MHz boost, pushing 2.304 TFLOPS of FP32 performance. Memory is 4 GB of GDDR5 on a 256-bit bus, providing 153.6 GB/s of bandwidth. The R9 M290X supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170, but lacks any dedicated ray tracing or tensor cores.

The NVIDIA GeForce RTX 3080 is built on the GA102 chip using Ampere architecture, fabricated on Samsung’s 8 nm process. This is a monster chip with 28,300 million transistors on a 628 mm² die, achieving a density of 45.1M / mm² — more than three times the R9 M290X’s density. The RTX 3080 features 8,704 shading units, 272 TMUs, and 96 ROPs, along with 68 dedicated RT cores and 272 tensor cores. Clock speeds are significantly higher: 1440 MHz base and 1710 MHz boost, yielding 29.77 TFLOPS of FP32 compute (and equal 29.77 TFLOPS FP16 via 1:1 ratio). Memory is 10 GB of GDDR6X on a 320-bit bus, delivering 760.3 GB/s — nearly five times the bandwidth of the older AMD part. The RTX 3080 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, and requires a 700 W suggested PSU with a 1x 12-pin power connector.

The architectural differences are not just about numbers — they reflect different capabilities. The RTX 3080’s RT and tensor cores enable hardware-accelerated ray tracing and AI-based features, while the R9 M290X has no such hardware. The process node shrink from 28 nm to 8 nm allows for a massive increase in transistor count and density, directly enabling the RTX 3080’s superior compute throughput and memory bandwidth.

---

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and the result is decisive. The NVIDIA GeForce RTX 3080 scores 152,423, while the AMD Radeon R9 M290X scores 22,028. This represents an 85.5% advantage for NVIDIA, meaning the RTX 3080 is roughly 7 times faster in this compute workload. This is not a marginal win; it is a complete obliteration of the older architecture.

To contextualize this, consider the R9 M290X’s nearest rivals in the overall benchmark database. The R9 M290X’s average score of 23,276 is within 0.1% of the AMD Radeon RX 6600M (23,273) and the NVIDIA P106-100 (23,249), and 0.2% ahead of the AMD Radeon AI PRO R9700 (23,315). This shows that the R9 M290X, despite its age, is not a slouch in general-purpose tasks — it trades blows with much newer mid-range parts. However, the RTX 3080’s average score of 23,172 is actually slightly lower than the R9 M290X’s average, which is counterintuitive given the OpenCL result.

This discrepancy is explained by the RTX 3080’s benchmark profile. Its scores range from 100 in Passmark DirectX 12 to 25,086 in Passmark G3D, with a 14,397 in Passmark GPU Compute and 1,054 in Passmark G2D. The low DirectX 9 (258), DirectX 10 (170), and DirectX 11 (207) scores are likely artifacts of driver or API overhead, dragging down the average. In contrast, the R9 M290X only has two benchmark entries: 24,524 in Geekbench Metal and 22,028 in Geekbench OpenCL, both of which are consistent. This highlights a key insight: the RTX 3080’s average score understates its true capability in modern workloads, whereas the R9 M290X’s average is more representative of its overall performance.

The head-to-head data shows only one test, but it is the most relevant one for general compute. The RTX 3080’s 85.5% lead in OpenCL is a clear indicator of its architectural superiority, driven by its higher core count, clock speeds, and memory bandwidth. For users prioritizing compute performance, the RTX 3080 is the unequivocal winner.

---

Specification Differences

The two GPUs differ in nearly every measurable specification. The most striking contrast is in the process node: the AMD R9 M290X uses 28 nm (TSMC), while the NVIDIA RTX 3080 uses 8 nm (Samsung). This enables the RTX 3080 to pack 28,300 million transistors versus 2,800 million for the R9 M290X, on a 628 mm² die versus 212 mm². Transistor density is 45.1M / mm² for NVIDIA versus 13.2M / mm² for AMD.

Clock speeds differ substantially: the R9 M290X runs at 850 MHz base and 900 MHz boost, while the RTX 3080 runs at 1440 MHz base and 1710 MHz boost. Memory clocks are also higher on the RTX 3080: 1188 MHz (19 Gbps effective) versus 1200 MHz (4.8 Gbps effective) for the AMD part. This contributes to a massive bandwidth gap: 760.3 GB/s versus 153.6 GB/s.

Compute resources are vastly different: the RTX 3080 has 8,704 shading units, 272 TMUs, and 96 ROPs, compared to 1,280, 80, and 32 for the R9 M290X. The RTX 3080 also adds 68 RT cores and 272 tensor cores, which the AMD part lacks entirely. This leads to 29.77 TFLOPS FP32 versus 2.304 TFLOPS, and the RTX 3080 offers 29.77 TFLOPS FP16, while the R9 M290X has no listed FP16 capability.

Memory configurations differ: 10 GB GDDR6X on a 320-bit bus versus 4 GB GDDR5 on a 256-bit bus. Power requirements are also stark: the RTX 3080 has a 320 W TDP and requires a 700 W suggested PSU with a 1x 12-pin connector, while the R9 M290X has a 100 W TDP, uses an MXM Module form factor, and needs no external power connectors.

API support shows the generational gap: the RTX 3080 supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the R9 M290X is limited to DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6. The RTX 3080 also has a PCIe 4.0 x16 interface versus PCIe 3.0 x16 for the AMD part. Finally, the RTX 3080 has a physical footprint of 285 mm (11.2 inches) length, 112 mm (4.4 inches) height, and 40 mm (1.6 inches) width, while the R9 M290X’s dimensions are listed as portable device dependent.

---

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The NVIDIA GeForce RTX 3080 is decisively faster, scoring 152,423 in Geekbench OpenCL compared to the AMD Radeon R9 M290X’s 22,028 — an 85.5% lead.

Q: Do these GPUs support the same DirectX version?

A: No. The RTX 3080 supports DirectX 12 Ultimate (12_2), while the R9 M290X only supports DirectX 12 (11_1). The RTX 3080 also has a newer Vulkan version (1.4 versus 1.2.170).

Q: How much memory does each GPU have, and what type?

A: The RTX 3080 has 10 GB of GDDR6X on a 320-bit bus, while the R9 M290X has 4 GB of GDDR5 on a 256-bit bus. This results in 760.3 GB/s bandwidth for NVIDIA versus 153.6 GB/s for AMD.

Q: What is the power consumption difference?

A: The RTX 3080 has a 320 W TDP and requires a 700 W suggested PSU, while the R9 M290X has a 100 W TDP and uses no external power connectors (MXM Module form factor).

Q: Are these GPUs still in production?

A: Both are listed as End-of-life. The R9 M290X was released on 2014-01-08, while the RTX 3080 was released on 2020-08-31.

Q: How do their average benchmark scores compare to their closest rivals?

A: The R9 M290X’s average score of 23,276 is within 0.1% of the AMD Radeon RX 6600M and NVIDIA P106-100. The RTX 3080’s average score of 23,172 is 0.4% lower than the AMD Radeon R9 M290X, and 0.3% lower than the NVIDIA P106-100, though this is skewed by its low DirectX 9/10/11 scores.