AMD Radeon R9 M290X vs NVIDIA GeForce GTX 980 Ti Comparison

Where Each One Wins

The recorded benchmark data splits the two contenders almost perfectly down the middle, with each card claiming one decisive victory. The AMD Radeon R9 M290X takes the Metal API test, scoring 24,524 against the NVIDIA GeForce GTX 980 Ti's 19,520, a margin of 20.4% in AMD's favor. This is a significant lead in a modern, low-level graphics API commonly used in Apple ecosystem applications and games. For workloads that leverage Metal, the R9 M290X demonstrates a clear advantage that cannot be ignored.

Conversely, the NVIDIA GeForce GTX 980 Ti dominates the OpenCL compute benchmark with a score of 43,513, compared to the R9 M290X's 22,028. The delta here is a massive 97.5%, meaning the GTX 980 Ti delivers nearly double the performance of its rival in this particular test. OpenCL is a cross-platform, vendor-neutral compute standard used heavily in professional applications, scientific simulations, and content creation tools. For any user whose primary workload is OpenCL-based, the GTX 980 Ti is the only rational choice based on this data.

The overall average benchmark scores reinforce this split. The GTX 980 Ti holds an average score of 28,020, placing it in the 73rd percentile of all GPUs in the database. The R9 M290X, with an average score of 23,276, sits in the 68th percentile. While the GTX 980 Ti has the higher overall average, the R9 M290X's commanding win in Metal shows that "overall" averages can obscure specific API-level strengths. The data suggests a clear use-case bifurcation: Metal-centric tasks belong to AMD, while OpenCL-heavy workloads belong to NVIDIA.

Architecture Differences

The architectural gap between these two GPUs is substantial, reflecting their different design philosophies and release timelines. The NVIDIA GeForce GTX 980 Ti is built on the GM200 chip using the Maxwell 2.0 architecture, manufactured on a 28 nm process at TSMC. It integrates 8,000 million transistors on a die size of 601 mm², yielding a transistor density of 13.3M per mm². The AMD Radeon R9 M290X, in contrast, uses the Neptune chip based on the older GCN 1.0 architecture, also on a 28 nm TSMC process, but with only 2,800 million transistors on a much smaller 212 mm² die, resulting in a density of 13.2M per mm².

Core configurations differ dramatically. The GTX 980 Ti fields 2,816 shading units, 176 texture mapping units, and 96 ROPs. The R9 M290X offers 1,280 shading units, 80 TMUs, and just 32 ROPs. This 2.2x advantage in shader count and 6x advantage in ROP count for NVIDIA explains much of its raw throughput advantage in compute-heavy tasks. Clock speeds also favor NVIDIA: the GTX 980 Ti runs at a base of 1000 MHz with a boost up to 1076 MHz, while the R9 M290X operates at a lower 850 MHz base and 900 MHz boost. Memory subsystems are equally disparate. The GTX 980 Ti uses 6 GB of GDDR5 on a 384-bit bus, delivering 336.6 GB/s of bandwidth. The R9 M290X has 4 GB of GDDR5 on a narrower 256-bit bus, resulting in 153.6 GB/s, less than half the bandwidth.

The resulting pixel and texture rates are telling. The GTX 980 Ti achieves 103.3 GPixel/s and 189.4 GTexel/s, while the R9 M290X manages only 28.80 GPixel/s and 72.00 GTexel/s. FP32 compute output is 6.060 TFLOPS for NVIDIA versus 2.304 TFLOPS for AMD. Power envelopes also differ: the GTX 980 Ti has a 250 W TDP requiring dual-slot cooling and both 6-pin and 8-pin power connectors, plus a suggested 600 W PSU. The R9 M290X, designed as an MXM module for portable systems, has a 100 W TDP and requires no external power connectors.

API support shows minor divergences. Both support DirectX 12, but NVIDIA lists 12_1 while AMD lists 11_1. Both support OpenGL 4.6. Vulkan support exists on both, but NVIDIA lists version 1.4 versus AMD's 1.2.170. The GTX 980 Ti also offers a richer display output set with 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2, whereas the R9 M290X's outputs are described as "Portable Device Dependent," reflecting its mobile-oriented MXM form factor.

Head-to-Head Benchmarks

The two benchmarks where both GPUs have recorded scores provide the clearest comparison. In Geekbench Metal, the AMD Radeon R9 M290X scores 24,524 against the NVIDIA GeForce GTX 980 Ti's 19,520. The delta percentage of -20.4% (from NVIDIA's perspective) means AMD leads by roughly one-fifth. This is not a marginal win; it is a decisive advantage in a specific API that has become increasingly relevant for gaming and GPU compute on macOS and iOS platforms. The R9 M290X's GCN 1.0 architecture, despite being older and having fewer shaders, evidently maps more efficiently onto Metal's execution model.

In Geekbench OpenCL, the picture reverses dramatically. The GTX 980 Ti scores 43,513, while the R9 M290X scores 22,028. The delta of 97.5% indicates NVIDIA nearly doubles AMD's output. This aligns perfectly with the raw compute specifications: 6.060 TFLOPS versus 2.304 TFLOPS. The GTX 980 Ti's Maxwell 2.0 architecture, with its far higher shader count, clock speeds, and memory bandwidth, translates directly into OpenCL performance. The result is not close, and no architectural nuance can close a gap of this magnitude.

The aggregate data also contextualizes these results. The GTX 980 Ti's average benchmark score of 28,020 puts it just ahead of the AMD Radeon Pro W5500X (27,973, a 0.2% delta) and the AMD Radeon RX 7800M (27,883, a 0.5% delta), while trailing the AMD FirePro S7150 (28,117, a -0.3% delta) by a hair. The R9 M290X's average of 23,276 is effectively tied with the AMD Radeon RX 6600M (23,273, 0% delta) and the NVIDIA P106-100 (23,249, 0.1% delta). These nearest-rival comparisons show that the GTX 980 Ti punches at a much higher performance class overall, but the R9 M290X's Metal result proves it can still win API-specific battles.

The Verdict

The data presents a clear if bifurcated verdict. The NVIDIA GeForce GTX 980 Ti is the superior GPU for raw compute throughput, as evidenced by its 97.5% lead in OpenCL and its higher average benchmark score of 28,020 versus 23,276. It also holds a significant advantage in memory bandwidth (336.6 GB/s versus 153.6 GB/s), pixel rate (103.3 GPixel/s versus 28.80 GPixel/s), and texture rate (189.4 GTexel/s versus 72.00 GTexel/s). For professional workloads, scientific computing, or any application that relies on OpenCL, the GTX 980 Ti is the only defensible choice from these two.

However, the AMD Radeon R9 M290X wins the Metal benchmark by 20.4%, a substantial margin. For users whose software stack is built around Metal, the R9 M290X offers better recorded performance despite its smaller silicon and lower power draw. It also fits into a completely different physical profile: the MXM module format and 100 W TDP versus the GTX 980 Ti's dual-slot, 250 W desktop card design. The R9 M290X is for portable systems; the GTX 980 Ti is for full-size desktops.

The database's percentile rankings reinforce the overall class difference. The GTX 980 Ti sits at the 73rd percentile of all GPUs, while the R9 M290X sits at the 68th. In absolute terms, NVIDIA is the faster card. But "faster" is not monolithic. The R9 M290X's Metal victory means it wins in its specific ecosystem. The final choice hinges on the API and form factor requirements of the target system. If the workload is OpenCL on a desktop, choose the GTX 980 Ti. If the workload is Metal in a portable chassis, the R9 M290X is the data-backed winner.

FAQ

Q: Which GPU has the higher overall average benchmark score?

A: The NVIDIA GeForce GTX 980 Ti, with an average score of 28,020, while the AMD Radeon R9 M290X averages 23,276.

Q: How large is the GTX 980 Ti's lead in OpenCL performance?

A: The GTX 980 Ti scores 43,513 in Geekbench OpenCL versus the R9 M290X's 22,028, a delta of 97.5%.

Q: Does the AMD Radeon R9 M290X win any benchmark?

A: Yes, it wins the Geekbench Metal test with a score of 24,524, beating the GTX 980 Ti's 19,520 by 20.4%.

Q: What are the memory bandwidth figures for each card?

A: The GTX 980 Ti has a bandwidth of 336.6 GB/s from 6 GB of GDDR5 on a 384-bit bus, while the R9 M290X has 153.6 GB/s from 4 GB of GDDR5 on a 256-bit bus.

Q: What is the percentile ranking for each GPU?

A: The GTX 980 Ti ranks in the 73rd percentile of all GPUs, while the R9 M290X ranks in the 68th percentile.

Q: How do the power requirements differ?

A: The GTX 980 Ti has a TDP of 250 W, requires dual-slot cooling, and needs both a 6-pin and an 8-pin power connector. The R9 M290X has a TDP of 100 W, uses an MXM module form factor, and requires no external power connectors.