NVIDIA GeForce GTX 760M vs NVIDIA GeForce GTX 980M Comparison

# NVIDIA GeForce GTX 980M vs NVIDIA GeForce GTX 760M

The data presents a clear generational clash between two NVIDIA mobile GPUs from different eras. The GTX 980M, built on Maxwell 2.0 architecture with a GM204 chip, faces the older Kepler-based GTX 760M with its GK106S processor. While both cards currently sit in the 31st percentile of all GPUs according to average benchmark scores, their performance profiles diverge dramatically in compute-oriented workloads. The head-to-head benchmark data shows the GTX 980M dominating across every shared test, yet the GTX 760M still holds its own in the broader competitive landscape, sitting within 1.3% of the newer card's average score. This paradox—similar aggregate positioning but vastly different raw compute results—invites a closer examination of what each GPU actually does best.

FAQ

Q: How much faster is the GTX 980M in OpenCL compute performance?

A: The GTX 980M scores 23,832 in Geekbench OpenCL versus the GTX 760M's 5,604, representing a delta of 325.3%—more than four times the older card's compute throughput.

Q: What is the average benchmark score difference between these two GPUs?

A: The GTX 980M averages 5,308 points across all benchmarks, while the GTX 760M averages 5,236 points. The GTX 760M's nearest rival list shows the GTX 980M at a -1.3% delta from its score, indicating the difference is surprisingly small in aggregate terms.

Q: Which GPU has more shading units and texture mapping units?

A: The GTX 980M features 1,536 shading units and 96 TMUs, compared to the GTX 760M's 768 shading units and 64 TMUs. The newer card also doubles the ROP count at 64 versus 16.

Q: How do their memory subsystems compare?

A: The GTX 980M offers 8 GB of GDDR5 on a 256-bit bus with 160.4 GB/s bandwidth, while the GTX 760M provides 2 GB of GDDR5 on a 128-bit bus with 64.13 GB/s bandwidth. The 980M's memory clock runs at 1253 MHz (5 Gbps effective) versus 1002 MHz (4 Gbps effective) on the 760M.

Q: What are the transistor and die size differences?

A: The GTX 980M's GM204 chip packs 5,200 million transistors on a 398 mm² die, while the GTX 760M's GK106S contains 2,540 million transistors on a 221 mm² die. Both use a 28 nm TSMC process, but the 980M achieves a higher density of 13.1M transistors per mm² versus 11.5M.

Q: Which GPU supports newer API versions?

A: The GTX 980M supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 760M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175—meaning the 980M has a higher DirectX feature level and a newer Vulkan version.

Where Each One Wins

The GTX 980M wins decisively in every head-to-head benchmark that exists in the data. In Geekbench OpenCL, it scores 23,832 against 5,604, a 325.3% advantage. In Geekbench Vulkan, the 980M posts 17,703 versus 4,868, a 263.7% lead. These results point to a GPU that is fundamentally better suited for compute-heavy workloads, whether they run through OpenCL or Vulkan compute APIs.

The GTX 760M's wins are more subtle and contextual. Its average benchmark score of 5,236 places it within striking distance of the 980M's 5,308, and its nearest rival list includes the Quadro 4000M at 5,211 (0.5% behind) and the Radeon R7 M260X at 5,161 (1.5% behind). This suggests the 760M remains competitive in the broader laptop GPU landscape, particularly for legacy DirectX workloads where its PassMark scores are not directly compared here. The 760M also holds a lower TDP of 55 W, which could imply better efficiency in thin-and-light chassis, though the data does not provide direct power consumption comparisons for the 980M.

Architecture Differences

The GTX 980M employs Maxwell 2.0 architecture on the GM204 chip, while the GTX 760M uses the older Kepler architecture on the GK106S. Both are fabricated on TSMC's 28 nm process, but the transistor counts tell a story of rapid advancement: the 980M packs 5,200 million transistors versus 2,540 million on the 760M. The die sizes reflect this—398 mm² for GM204 against 221 mm² for GK106S—yielding transistor densities of 13.1M per mm² and 11.5M per mm², respectively.

The Maxwell 2.0 architecture brings a higher DirectX feature level (12_1 versus 11_0) and a more recent Vulkan implementation (1.4 versus 1.2.175). Both GPUs support OpenGL 4.6, indicating parity in that API. The 980M's memory controller doubles the bus width to 256-bit from 128-bit, and its memory clock runs faster at 1253 MHz versus 1002 MHz. The shading engine has been substantially reorganized: 1,536 shading units on the 980M versus 768 on the 760M, with TMUs increasing from 64 to 96 and ROPs jumping from 16 to 64. These architectural changes translate directly into higher pixel and texture rates—72.13 GPixel/s and 108.2 GTexel/s on the 980M versus 11.50 GPixel/s and 46.02 GTexel/s on the 760M.

Specification Differences

The GTX 980M and GTX 760M differ across nearly every measurable specification. Clock speeds show a significant gap: the 980M runs at 1038 MHz base and 1127 MHz boost, while the 760M runs at 628 MHz base and 719 MHz boost. Memory capacity quadruples from 2 GB to 8 GB, and bandwidth more than doubles from 64.13 GB/s to 160.4 GB/s. The 980M's FP32 compute output reaches 3.462 TFLOPS, while the 760M produces 1,104.4 GFLOPS.

The bus interface differs as well—the 980M uses MXM-B (3.0), while the 760M uses PCIe 3.0 x16. The 760M lists a TDP of 55 W, whereas the 980M has no TDP specified in the data. Both use MXM Module slot width with no power connectors, and both have portable-device-dependent display outputs. The 980M belongs to the GeForce 900M generation and was released in October 2014, while the 760M belongs to the GeForce 700M generation from May 2013. Their predecessor/successor chains are also distinct: the 980M follows GeForce 800M and leads to GeForce 10 Mobile, while the 760M follows GeForce 600M and leads to GeForce 800M.

Head-to-Head Benchmarks

The head-to-head data contains two tests, and the GTX 980M wins both with overwhelming margins. The first is Geekbench OpenCL, where the 980M scores 23,832 against the 760M's 5,604. This 325.3% delta represents the largest performance gap in the dataset, indicating that the 980M's additional shading units, higher clocks, and wider memory bus translate into massive OpenCL compute advantages. The 980M's 3.462 TFLOPS FP32 throughput, combined with 160.4 GB/s of memory bandwidth, allows it to sustain far more parallel work than the 760M's 1,104.4 GFLOPS and 64.13 GB/s.

The second test is Geekbench Vulkan, where the 980M scores 17,703 versus 4,868 on the 760M—a 263.7% delta. This slightly smaller gap still represents a more than 3.5x performance lead. The 980M's Vulkan 1.4 support versus the 760M's Vulkan 1.2.175 may contribute to better driver-level optimization in modern workloads. Notably, the 980M's Vulkan score is lower than its OpenCL score, while the 760M shows a similar pattern (4,868 Vulkan versus 5,604 OpenCL), suggesting both GPUs perform relatively better in OpenCL than Vulkan.

Beyond the head-to-head tests, the 980M's broader benchmark suite reveals strengths in legacy DirectX workloads: it scores 125 in PassMark DirectX 9, 57 in DirectX 11, 35 in DirectX 10, and 31 in DirectX 12. Its PassMark G3D score of 7,338 and G2D score of 490, plus a GPU compute score of 2,816, round out a picture of balanced performance. The 760M lacks these additional data points, so direct comparisons beyond OpenCL and Vulkan are impossible from the available information.

The Verdict

The data makes a definitive case for the GTX 980M as the superior GPU in raw performance terms. It wins both head-to-head benchmarks by margins of 263.7% and 325.3%, offers 8 GB of memory versus 2 GB, and delivers more than triple the FP32 compute throughput. Anyone running OpenCL or Vulkan compute workloads should choose the 980M without hesitation—the numbers show a GPU that is in a different performance class entirely.

However, the aggregate benchmark scores tell a more nuanced story. The 980M averages 5,308 across all tests, while the 760M averages 5,236. The 760M's nearest rival list places the 980M at only -1.3% delta, meaning the older card's overall performance profile is remarkably close in the broader GPU landscape. This suggests that for typical gaming or general-purpose workloads—where average scores weigh multiple test types—the 760M remains a viable option, especially considering its 55 W TDP may enable thinner laptop designs.

The GTX 760M also benefits from a different connectivity path: its PCIe 3.0 x16 interface is more conventional than the 980M's MXM-B (3.0), potentially making it easier to integrate into existing laptop platforms. For users prioritizing compute performance, modern API support, and memory capacity, the GTX 980M is the clear choice. For those seeking a proven, lower-power GPU that still holds its own in average benchmarks, the 760M's data profile shows it remains competitive despite its age. The verdict from the numbers is unambiguous: the 980M wins on performance, but the 760M's aggregate scores reveal a surprisingly resilient legacy part.