AMD Radeon 890M vs NVIDIA GeForce GTX 960M Comparison

The benchmark data is unequivocal: the AMD Radeon 890M is the superior mobile graphics processor by a massive margin, decisively outperforming the NVIDIA GeForce GTX 960M in every head-to-head test recorded. The NVIDIA GTX 960M, an end-of-life part from 2015, is outclassed in raw compute, modern API support, and efficiency by the 2024 AMD RDNA 3.5 integrated processor. The 890M is the clear choice for any user prioritizing performance, while the GTX 960M offers no advantage in the data.

The Verdict

The verdict is simple: choose the AMD Radeon 890M. The data shows a generational chasm, not a close contest. In the Geekbench OpenCL test, the 890M scores 37,254, which is 70.4% higher than the GTX 960M’s 11,045. In the Vulkan test, the 890M’s 40,808 score is 79.8% higher than the GTX 960M’s 8,245. The 890M wins both recorded head-to-head benchmarks with overwhelming deltas. This is a complete victory for the newer AMD part.

The GTX 960M’s position in the current market is untenable. Its average benchmark score of 9,645 places it in the 46th percentile of all GPUs, while its nearest rivals are the NVIDIA Quadro K5000, AMD Radeon Pro WX 2100, NVIDIA Quadro P4000, and NVIDIA Tesla C2070, with scores ranging from 9,637 to 9,716. These deltas are incredibly small, between -0.7% and 0.1%, indicating that the GTX 960M is squarely in a cluster of older, mid-range workstation parts. The 890M, despite having a slightly lower overall percentile rank of 45, is surrounded by a different class of rivals, including the NVIDIA GeForce GTX 960 and GTX 850M, and achieves its average score of 9,210 through a different, more modern benchmark suite.

For users, the choice is unambiguous. The Radeon 890M is for anyone who wants the best possible performance in a thin-and-light laptop, delivering desktop-class compute in a 15 W integrated package. The GTX 960M, with its 75 W TDP and older Maxwell architecture, is a relic. It should only be considered if legacy driver support or a specific, older software requirement demands it, but even then, its performance is inferior in every measurable way. The 890M is the only rational choice based on the data.

FAQ

Q: Which GPU is faster in the Geekbench OpenCL benchmark?

A: The AMD Radeon 890M is dramatically faster, scoring 37,254 compared to the NVIDIA GTX 960M’s 11,045. This represents a 70.4% performance advantage for the 890M.

Q: How do the two GPUs compare in Vulkan performance?

A: The AMD Radeon 890M wins decisively with a score of 40,808, which is 79.8% higher than the NVIDIA GTX 960M’s score of 8,245.

Q: What are the TDP (power draw) differences?

A: The AMD Radeon 890M has a TDP of 15 W, while the NVIDIA GeForce GTX 960M has a TDP of 75 W. The 890M delivers significantly more performance while consuming far less power.

Q: Which GPU has a higher transistor density?

A: The AMD Radeon 890M has a far higher transistor density at 145.9M / mm², compared to the NVIDIA GTX 960M’s 12.6M / mm². This is due to its modern 4 nm process node versus the GTX 960M’s 28 nm node.

Q: Do these GPUs support the same DirectX version?

A: No. The AMD Radeon 890M supports DirectX 12 Ultimate (12_2), while the NVIDIA GeForce GTX 960M supports an older DirectX 12 (11_0) feature level. This gives the 890M access to newer rendering features.

Q: What is the memory configuration of the AMD Radeon 890M?

A: The AMD Radeon 890M uses system shared memory, meaning its memory size, type, bus width, and bandwidth are all system-dependent. In contrast, the GTX 960M has a dedicated 4 GB of GDDR5 memory on a 128-bit bus.

Architecture Differences

The two GPUs represent fundamentally different eras and design philosophies. The NVIDIA GeForce GTX 960M is built on the Maxwell architecture using the GM107 chip, fabricated on a 28 nm process at TSMC. It packs 1,870 million transistors into a die size of 148 mm². In contrast, the AMD Radeon 890M is based on the modern RDNA 3.5 architecture, using the Strix Point chip, also fabricated by TSMC but on a much more advanced 4 nm node. This allows AMD to pack a massive 34,000 million transistors into a 233 mm² die.

The transistor density difference is stark: the 890M achieves 145.9M transistors per mm², while the GTX 960M manages only 12.6M per mm². This density advantage is a core reason for the 890M's efficiency and performance. The 890M also features 16 dedicated ray tracing cores, a feature entirely absent from the older GTX 960M. This architectural addition enables hardware-accelerated ray tracing, a key feature of modern games and applications.

The API support further illustrates the generational gap. The 890M supports DirectX 12 Ultimate (12_2), the latest version with features like variable rate shading and mesh shaders. The GTX 960M is limited to DirectX 12 (11_0) and lacks these modern capabilities. Furthermore, the 890M is a true integrated graphics processor (IGP) with a 15 W TDP, while the GTX 960M is a discrete MXM module with a 75 W TDP, requiring a dedicated slot. The 890M's memory architecture is also fundamentally different, relying on system shared memory, while the GTX 960M has its own dedicated VRAM.

Specification Differences

Beyond the core architecture, the specifications diverge completely. The AMD Radeon 890M has 1,024 shading units, 64 texture mapping units (TMUs), and 32 render output units (ROPs). The NVIDIA GeForce GTX 960M has 640 shading units, 40 TMUs, and 16 ROPs. The 890M’s clock speeds are also in a different league, with a boost clock of 2900 MHz compared to the GTX 960M’s 1176 MHz boost.

This hardware translates to vastly different compute and fill rates. The 890M achieves 5.939 TFLOPS of FP32 performance, while the GTX 960M manages only 1.505 TFLOPS. The pixel rate is 92.80 GPixel/s for the 890M versus 18.82 GPixel/s for the GTX 960M, and the texture rate is 185.6 GTexel/s versus 47.04 GTexel/s. The 890M also supports FP16 compute at a 1:1 ratio, a feature the GTX 960M lacks.

Memory is another point of distinction. The GTX 960M uses 4 GB of GDDR5 on a 128-bit bus, providing 80.19 GB/s of bandwidth. The 890M uses system shared memory, making its bandwidth and capacity dependent on the host system’s RAM. The GTX 960M uses an MXM-B (3.0) interface, while the 890M uses PCIe 4.0 x8. The GTX 960M is listed as end-of-life, released in 2015, while the 890M is an active product released in 2024.

Head-to-Head Benchmarks

The head-to-head results are a one-sided affair, with the AMD Radeon 890M winning both recorded tests. The first test, Geekbench OpenCL, shows the 890M scoring 37,254 against the GTX 960M’s 11,045. This gives the 890M a 70.4% lead, a performance gap that translates to dramatically faster compute workloads, from video encoding to physics simulations. The second test, Geekbench Vulkan, shows an even larger disparity. Here, the 890M scores 40,808, while the GTX 960M scores 8,245, a 79.8% difference. This indicates that in modern Vulkan-based games and applications, the 890M is nearly five times faster.

These results are not marginal wins; they are a complete rout. The GTX 960M’s scores are closer to its nearest rivals, which include the NVIDIA Quadro K5000 (9,637) and the AMD Radeon Pro WX 2100 (9,653), all within a 0.7% range. This context shows that the GTX 960M is a competent, if old, mid-range performer. The 890M, on the other hand, is in a class of its own compared to the GTX 960M, with its average score of 9,210 being pulled down by older DirectX 9 and 10 tests where it scores 97 and 33 respectively, but its modern API scores are exceptional.

Where Each One Wins

Based on the benchmark data, the AMD Radeon 890M wins in every measurable category. It is the clear victor in modern compute and graphics workloads, as evidenced by its massive leads in both the Geekbench OpenCL and Vulkan tests. Its 5.939 TFLOPS FP32 performance and support for DirectX 12 Ultimate make it the superior choice for modern gaming, content creation, and any GPU-accelerated application. Its low 15 W TDP also makes it ideal for thin-and-light laptops where battery life and thermals are critical, while still delivering high-end performance.

The NVIDIA GeForce GTX 960M has no benchmark wins in the data. Its only potential advantages are speculative and not supported by the head-to-head results. While it has dedicated GDDR5 memory, which can sometimes be beneficial in specific legacy scenarios, the 890M's sheer compute power overcomes this in all recorded tests. The GTX 960M's place is firmly in the past, as an end-of-life product from the Maxwell era. Its lower scores and older API support mean it is not recommended for any modern workload where the 890M is available. The data is clear: the 890M is the definitive winner, with the GTX 960M offering no performance advantage whatsoever.