NVIDIA GeForce GTX 980M

The NVIDIA GeForce GTX 980M is a mobile GPU from the GeForce 900M generation, built on the Maxwell 2.0 architecture with a 28 nm process at TSMC. It carries 5,200 million transistors on a 398 mm² die, resulting in a transistor density of 13.1M per mm². Its benchmark profile places it at the 27th percentile of all GPUs, with an average benchmark score of 4934. The data shows a part that is competitive with entry-level desktop cards from its era, but it is clearly outclassed by modern mobile silicon. This analysis draws exclusively from the provided fact pack to detail its positioning, features, and performance characteristics.

Who Should Consider It

The GTX 980M is suited for users who prioritize consistent 1080p gaming at medium to high settings in titles that rely on DirectX 11 or older APIs. Its PassMark DirectX 11 score of 57 and DirectX 9 score of 125 indicate strong legacy performance, making it a viable option for older game libraries or esports titles that do not demand modern rendering features. However, for DirectX 12 workloads, the score drops sharply to 31, suggesting that newer, more demanding games will require reduced resolutions or lower detail presets to maintain playable frame rates.

At 1440p or higher, the memory subsystem—8 GB of GDDR5 on a 256-bit bus with 160.4 GB/s bandwidth—provides enough capacity for large textures, but the raw compute throughput of 3.462 TFLOPS FP32 limits the GPU's ability to drive high pixel counts in modern titles. The 3DMark Steel Nomad DX12 score of 649 reinforces this conclusion: it is a low score that indicates the GPU struggles with contemporary DirectX 12 rendering workloads. The Geekbench OpenCL score of 23832 and Vulkan score of 13964 show moderate compute capability, but these numbers are not competitive with current-generation parts.

The 27th percentile ranking against all GPUs confirms that this is not a card for high-refresh-rate or high-resolution gaming in 2025. Users should consider it only if they have a portable device with an MXM-B (3.0) slot and require a drop-in replacement that matches the original performance class. The GTX 980M is end-of-life, so it is a candidate for refurbished systems rather than new purchases.

Ray Tracing and Feature Set

The GTX 980M does not include dedicated ray tracing cores, as the fact pack lists no RT core count. Similarly, it has no tensor cores, which means it lacks hardware acceleration for AI-based features like DLSS. This places it firmly in the pre-ray-tracing era of GPU design. The architecture is Maxwell 2.0, which supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The DirectX 12_1 feature level includes conservative rasterization and other optional features, but it does not support mesh shaders or variable rate shading found in later API revisions.

For gaming, this means the GTX 980M relies entirely on traditional rasterization. The 1536 shading units, 96 texture mapping units, and 64 raster output pipelines deliver a pixel rate of 72.13 GPixel/s and a texture rate of 108.2 GTexel/s. These figures are adequate for the era but pale against modern parts. The Vulkan 1.4 support is notable because it allows the GPU to run Vulkan-based games, but performance will be limited by the underlying hardware. The lack of RT and tensor cores means no ray-traced effects or AI upscaling are possible, so users must rely on native resolution rendering and standard anti-aliasing techniques.

Power and Cooling

The fact pack does not specify a TDP for the GTX 980M, so no thermal design power figure can be stated. What is known is that the card uses an MXM Module slot width and has no power connectors listed, meaning it draws power directly from the MXM-B (3.0) interface. This is typical for mobile GPUs, where the laptop's power delivery system handles the load. There is no suggested PSU rating, which is expected for a mobile part that does not require an external power supply.

Cooling requirements are qualitative: because the TDP is unspecified, the data cannot confirm whether a standard laptop cooler is sufficient. However, given the 28 nm process and the performance class, the GPU likely generates moderate heat that must be managed by the laptop's thermal solution. Users should ensure their portable device has adequate ventilation, especially under sustained load in GPU-intensive tasks. The absence of a power connector also means that upgrading from a lower-tier GPU to the 980M may require a compatible MXM-B slot with sufficient power delivery, but the fact pack does not provide specifics on power draw.

How It Compares

NVIDIA Quadro K3100M: The GTX 980M and Quadro K3100M are nearly identical in average benchmark score, with the 980M trailing by just 0.1% (4934 vs 4937). This places them in the same performance tier, despite the Quadro being a professional workstation part. The delta is within noise, so users should choose based on driver support and feature requirements rather than raw speed.

NVIDIA GeForce 930M: The GTX 980M leads the 930M by 0.2% (4934 vs 4927), a negligible margin. The 930M is a lower-tier mobile GPU, but the benchmark data shows that the 980M does not offer a meaningful advantage in average score. This suggests that in real-world workloads, the two are effectively interchangeable, which is surprising given the generational gap and memory differences.

AMD FirePro W5130M: The 980M is 0.6% ahead of the FirePro W5130M (4934 vs 4904). This is a marginal lead that falls within typical benchmark variance. The FirePro is a mobile workstation GPU, and the data indicates no clear winner between the two for general compute or graphics tasks.

NVIDIA GeForce GTS 450: The 980M outperforms the desktop GTS 450 by 0.8% (4934 vs 4893). The GTS 450 is a much older desktop part, yet the mobile 980M only edges it out by less than 1%. This highlights how mobile GPUs from the same era were often comparable to lower-end desktop cards, but it also shows that the 980M's performance advantage is minimal in aggregate benchmarks.

Memory Subsystem

The GTX 980M features 8 GB of GDDR5 memory on a 256-bit bus, with a bandwidth of 160.4 GB/s. The memory clock is listed at 1253 MHz, which translates to 5 Gbps effective. This configuration is generous for a mobile GPU of its generation, as 8 GB was uncommon in 2014 and remains sufficient for modern games at 1080p with high-resolution textures. The 256-bit bus width provides a balanced memory interface that avoids the bottlenecks seen in narrower-bus designs.

For high resolutions, the 160.4 GB/s bandwidth is a limiting factor. At 1440p or 4K, the GPU must move more pixel data, and the bandwidth may become saturated before the compute units are fully utilized. The 8 GB capacity is not the issue—it is the throughput. In contrast, a modern GPU with similar capacity but higher bandwidth would handle 1440p more comfortably. The pixel rate of 72.13 GPixel/s also constrains fill-rate-bound scenes, which further degrades performance at high resolutions. Users should treat the 980M as a 1080p card, with 1440p possible only in less demanding titles or with reduced settings.

FAQ

Q: Does the GTX 980M support DirectX 12?

A: Yes, it supports DirectX 12 (12_1), but its performance in DX12 workloads is significantly lower than in DX11, as shown by the PassMark scores of 31 for DX12 versus 57 for DX11.

Q: Can the GTX 980M handle ray tracing?

A: No, the GPU has no RT cores and no tensor cores, so it cannot accelerate ray-traced effects or AI-based features like DLSS.

Q: What is the memory bandwidth of the GTX 980M?

A: The memory bandwidth is 160.4 GB/s, provided by 8 GB of GDDR5 on a 256-bit bus with an effective memory clock of 5 Gbps.

Q: How does the GTX 980M compare to the Quadro K3100M?

A: The GTX 980M trails the Quadro K3100M by 0.1% in average benchmark score (4934 vs 4937), making them effectively equal in performance.

Q: Is the GTX 980M still in production?

A: No, its production status is end-of-life, and it was released on October 6, 2014.

Q: What APIs does the GTX 980M support?

A: It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

Benchmark Performance

The GTX 980M's average benchmark score of 4934 places it at the 27th percentile of all GPUs, which indicates that it sits below the median and is outperformed by the majority of modern parts. The 3DMark Steel Nomad DX12 score of 649 is particularly telling: it is a low score that shows the GPU's weakness in contemporary DirectX 12 rendering. In contrast, its PassMark DirectX 11 score of 57 is more respectable for its era, and the DirectX 9 score of 125 shows that legacy titles run comparatively well.

The Geekbench OpenCL score of 23832 and Vulkan score of 13964 provide a mixed picture. The OpenCL score is roughly 70% higher than the Vulkan score, suggesting that the GPU's compute performance is better optimized for OpenCL workloads than for Vulkan. However, both scores are modest by modern standards. The PassMark GPU Compute score of 2816 further indicates limited compute throughput, which aligns with the FP32 figure of 3.462 TFLOPS.

Relative to its nearest rivals, the 980M's performance is tightly clustered. The deltas are -0.1% versus the Quadro K3100M, +0.2% versus the GeForce 930M, +0.6% versus the FirePro W5130M, and +0.8% versus the GTS 450. These are all sub-1% differences, meaning the 980M offers no meaningful performance advantage over any of these cards in aggregate benchmarks. The PassMark G3D score of 7338 and G2D score of 490 further contextualize the GPU: the G3D score is moderate, but the G2D score is low, indicating weak 2D acceleration.

The pixel rate of 72.13 GPixel/s and texture rate of 108.2 GTexel/s are derived from the 64 ROPs and 96 TMUs, respectively. These rates are adequate for 1080p gaming but will bottleneck at higher resolutions. The 27th percentile ranking is the clearest verdict: this is a legacy GPU that is best suited for older games and light productivity, not for modern AAA titles at high settings.