AMD Radeon 880M vs NVIDIA GeForce GTX 960M Comparison
AMD Radeon 880M
GeForce GTX 960M
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon 880M vs NVIDIA GeForce GTX 960M
# NVIDIA GeForce GTX 960M vs AMD Radeon 880M
The database comparison places two mobile graphics processors from different generations against each other. The NVIDIA GeForce GTX 960M is an end-of-life discrete part from the Maxwell era, while the AMD Radeon 880M is an active integrated graphics processor built on the newer RDNA 3.5 architecture. The benchmark data shows a clear generational split, with the AMD part winning both head-to-head tests decisively.
Head-to-Head Benchmarks
The recorded head-to-head tests are unambiguous. In the Geekbench OpenCL test, the NVIDIA GTX 960M scores 11,045 points, while the AMD Radeon 880M scores 31,285 points. This is a 64.7% advantage for the AMD part, a massive lead that reflects not just architectural improvements but also the difference between a dedicated mobile GPU and a modern integrated GPU. The OpenCL test is heavily compute-bound, and the Radeon's 12 RT cores and higher FP32 throughput, listed in the database as 4.454 TFLOPS, give it a substantial edge in parallel workloads.
The Vulkan test shows an even larger gap. The GTX 960M records 8,245 points, but the Radeon 880M reaches 40,006 points. That is a 79.4% advantage for the AMD part. Vulkan is a lower-level API that can expose GPU compute resources more directly, and the Radeon architecture is clearly better suited for this kind of load. The NVIDIA part trails by a wide margin in both scenarios, and the data indicates the gap is not marginal but fundamental.
Architecture Differences
The two chips are from different design philosophies. The NVIDIA GTX 960M uses the GM107 chip, built on a 28 nm process at TSMC. It packs 1,870 million transistors into a 148 mm² die, resulting in a transistor density of 12.6 million per mm². The architecture is Maxwell, a design that prioritized power efficiency for the 900M series of notebook GPUs. The chip has 640 shading units, 40 texture mapping units, and 16 ROPs. Its memory is 4 GB of GDDR5 on a 128-bit bus, with a bandwidth of 80.19 GB/s. The base clock runs at 1097 MHz, with a boost of 1176 MHz, and the memory clock is 1253 MHz (5 Gbps effective).
The AMD Radeon 880M, on the other hand, is a Strix Point integrated GPU with an RDNA 3.5 architecture. It is built on a 4nm process at TSMC. This is a much more advanced chip with 34,000 million transistors on a 233 mm² die. The density is much higher, at 145.9 million per mm². Data shows the GPU has 768 shading units, 48 texture mapping units, and 16 ROPs. It has 12 ray tracing cores, which the NVIDIA chip lacks entirely. The AMD part uses system shared memory, so the bandwidth is system-dependent, but the architecture supports a substantial data throughput. The clock is a 400 MHz base with a 2900 MHz boost.
The architectural differences are not just a matter of counts. The RDNA 3.5 design incorporates newer instruction sets and scheduling methods. The ray tracing cores, for example, enable a different class of rendering and compute workloads. The Maxwell chip on the NVIDIA side is simpler and older, and its limited feature set is reflected in the lower benchmark results.
FAQ
Q: Which GPU has a better score in the Geekbench Vulkan test?
A: The AMD Radeon 8800M wins with a score of 40,006 vs 8,245 for the NVIDIA GeForce GTX 960M. This corresponds to a 79.4% difference in the database.
Q: Are the benchmark differences between the two parts consistent?
A: No. The AMD gap is larger in the Vulkan test (79.4%) than in the OpenCL test (64.7%). This discrepancy suggests that the NVIDIA drivers handle OpenCL better than Vulkan, but the AMD architecture is still far superior in both tests. The deltas indicate the AMD part has a bigger performance gain on Vulkan.
Q: Does the AMD part have a more advanced manufacturing process?
A: Yes. The AMD Radeon 880M is built on a 4nm process at TSMC, while the NVIDIA GeForce GTX 960M uses a 28nm process. The AMD part has a much higher transistor density, 145.9 million per mm², compared to 12.6 per mm² for the NVIDIA. This is indicative of a much newer design.
Specification Differences
The table below shows the main differences between the two processors, with only the fields in which the two differ.
| Specification | NVIDIA GeForce GTX 960M | AMD Radeon 880M |
|---|---|---|
| Chip | GM107 | Strix Point |
| Architecture | Maxwell | RDNA 3.5 |
| Process Node | 28 nm | 4 nm |
| Transistors | 1,870 million | 34,000 million |
| Die Size | 148 mm² | 233 mm² |
| Transistor Density | 12.6M / mm² | 145.9M / mm² |
| Shading Units | 640 | 768 |
| TMUs | 40 | 48 |
| ROPs | 16 | 16 |
| Ray Tracing Cores | None | 12 |
| Core Clock | 1097 MHz | 400 MHz |
| Boost Clock | 1176 MHz | 2900 MHz |
| Memory Size | 4 GB GDDR5 | System Shared |
| Memory Bus | 128 bit | System Shared |
| Memory Bandwidth | 80.19 GB/s | System Dependent |
| Bus Interface | MXM-B (3.0) | PCIe 4.0 x8 |
The AMD part is a mobile integrated GPU, so a video memory is borrowed from the system RAM. The NVIDIA part has its own 4 GB of GDDR5 dedicated memory on a 128-bit bus. This distinction is fundamental for the memory bandwidth: the AMD part is limited by system memory speed and allocation, while the NVIDIA part has a fixed bandwidth of 80.19 GB/s.
The feature sets also differ. The AMD part includes DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 support. The NVIDIA part supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. The AMD has a newer feature set for DirectX.
Where Each One Wins
The benchmarks from the database only contain two tests, and the AMD part wins both. It is a clear sweep in the recorded data, but the split between the two tests suggests the AMD part is particularly strong in Vulkan workloads, where the 79.4% advantage is larger than the 64.7% advantage in OpenCL. This suggests the RDNA 3.5 architecture is more efficient in translating the lower-level Vulkan API into its execution resources.
The NVIDIA GTX 960M can only be considered a winner in scenarios where the dedicated GDDR5 memory is decisive, for example in a laptop with a weak system memory configuration where the AMD part would be at a disadvantage due to the shared memory. In the benchmark data, however, the NVIDIA part does not win either test.
For users who need the highest performance in modern APIs, the AMD Radeon 880M is the data-driven choice. It wins in both head-to-head tests, and the margin is large. The NVIDIA part can be a good option for a legacy system with the MXM module, where the user wants a dedicated memory pool.
The Verdict
The database is clear: the AMD Radeon 880M is the superior part in the recorded tests. It wins the Geekbench OpenCL test with a 64.7% lead and the Geekbench Vulkan test with a 79.4% lead. The architecture's modern design, with more shading units, texture units, and dedicated ray tracing cores, provides a substantial performance advantage. The NVIDIA GTX 960M is an end-of-life processor from 2015, and its older Maxwell architecture is not competitive with the newer RDNA 3.5 design in these workloads.
The choice depends on the use case. For a laptop with the MXM module slot, a dedicated VRAM, and the need to run older DX11 workloads, the GTX 960M could be an option, but the data does not support it. The AMD Radeon 880M is the recommended choice for any user of the two, based on the average benchmark scores. The recorded data suggests that the AMD part is the one to pick for modern gaming and compute tasks.
The numbers are consistent across the board. The average score of 8,436 for the AMD part versus 9,645 for the NVIDIA part is not a close race. The winner, in the data, is the AMD Radeon 880M.