AMD Radeon 890M vs NVIDIA GeForce GTX 950A Comparison

# NVIDIA GeForce GTX 950A vs AMD Radeon 890M

The database record for these two mobile graphics processors shows a clear split. The NVIDIA GeForce GTX 950A belongs to the Maxwell generation, built for notebooks with an MXM module form factor. The AMD Radeon 890M is a current integrated graphics processor from the Strix Point family, using the RDNA 3.5 architecture. The two parts target different segments of the laptop market, and the benchmark data reflects that positioning.

Where Each One Wins

A breakdown of the recorded benchmark wins shows the AMD Radeon 890M takes a single head-to-head victory in the available comparison tests. The NVIDIA GeForce GTX 950A does not record a win in the direct comparison. The decisive test is the Geekbench OpenCL workload, where the AMD part scores 37,254 against NVIDIA's 10,273. This represents a 72.4% advantage for the AMD Radeon 890M. The NVIDIA part is not without its own strong points; its average score across all recorded benchmarks sits at 10,273, which is a solid result for a mobile part from the Maxwell generation. However, the AMD silicon records a much higher average benchmark score of 9,210 across the suite, showing a generational leap in raw compute capability.

The single head-to-head result in the database is the Geekbench OpenCL test. In this test, the AMD Radeon 890M delivers 37,254 points, while the NVIDIA GeForce GTX 950A reaches 10,272. The delta of -72.4% means the NVIDIA part trails by a very wide margin. The data indicates the 890M is in a different performance class for this type of workload.

Architecture Differences

The architectural gap between the two parts is substantial. The NVIDIA GeForce GTX 950A uses the GM107 chip, built on the Maxwell architecture. It is manufactured on a 28 nm process at TSMC, with 1,870 million transistors on a 148 mm² die. That results in a transistor density of 12.6 million per square millimeter. The chip has 640 shading units, 40 texture mapping units, and 16 ROPs. Its base clock is 993 MHz, with a boost clock of 1124 MHz. The memory subsystem is 2 GB of DDR3 on a 128-bit bus, delivering a bandwidth of 32.03 GB/s. The effective memory clock is listed at 1001 MHz, with the note of 2 Gbps effective throughput.

The AMD Radeon 890M, in contrast, uses the Strix Point chip based on RDNA 3.5. The manufacturing process is 4 nm at TSMC, with 34,000 million transistors on a 233 mm² die. The density reaches 145.9 million transistors per square millimeter. The 890M has 1024 shading units, 64 texture mapping units, and 32 ROPs. It also includes 16 ray tracing cores. Base clock is 400 MHz, with a boost up to 2900 MHz. Memory is system shared, so bandwidth and bus width are system dependent.

The compute figures show the gap: the AMD part reaches 5.939 TFLOPS of FP32 performance, while the NVIDIA part's FP32 figure is 1,438.7 GFLOPS. The AMD also reaches 5.939 TFLOPS in FP16, at a 1:1 ratio. The NVIDIA part does not list a FP16 figure in the pack. This is a major architectural gap: the 890M is a modern integrated GPU with FP16 support, while the GTX 950A is a discrete part from 2015 with a focus on 28nm efficiency.

FAQ

Q: Which part has the higher memory bandwidth?

A: The AMD Radeon 890M uses system shared memory, so its bandwidth is system dependent. The NVIDIA GeForce GTX 950A has a dedicated 2 GB DDR3 memory pool on a 128-bit bus, with a bandwidth of 32.03 GB/s.

Q: Are both parts compatible with DirectX 12?

A: Yes. The NVIDIA GeForce GTX 950A supports DirectX 12 (11_0) based on its feature level. The AMD Radeon 890M supports DirectX 12 Ultimate.

Q: Which part has more shading units?

A: AMD Radeon 890M has 1024 shading units, while the NVIDIA GeForce GTX 950A has 640 units.

Q: Does the AMD 890M include ray tracing cores?

A: Yes, the AMD Radeon 890M includes 16 ray tracing cores. The NVIDIA GTX 950A has no dedicated ray tracing core in the data.

Q: What is the transistor count?

A: The AMD Radeon 890M has 34,000 million transistors. The NVIDIA GeForce GTX 950A has 1,870 million transistors.

Specification Differences

The table below lists only the fields where the two parts differ.

| Field | NVIDIA GeForce GTX 950A | AMD Radeon 890M |

|---|---|---|

| Chip | GM107 | Strix Point |

| Architecture | Maxwell | RDNA 3.5 |

| Process node | 28 nm | 4 nm |

| Foundry | TSMC | TSMC |

| Transistors | 1,870 million | 34,000 million |

| Die size | 148 mm² | 233 mm² |

| Transistor density | 12.6M / mm² | 145.9M / mm² |

| Base clock | 993 MHz | 400 MHz |

| Boost clock | 1124 MHz | 2900 MHz |

| Memory | 2 GB DDR3 | System Shared |

| Memory bus width | 128 bit | System Shared |

| Memory bandwidth | 32.03 GB/s | System Dependent |

| Shading units | 640 | 1024 |

| TMUs | 40 | 64 |

| ROPs | 16 | 32 |

| RT cores | None | 16 |

| Pixel rate | 17.98 GPixel/s | 92.80 GPixel/s |

| Texture rate | 44.96 GTexel/s | 185.6 GTexel/s |

| FP32 | 1,438.7 GFLOPS | 5.939 TFLOPS |

| FP16 | Not listed | 5.939 TFLOPS (1:1) |

| TDP | 75 W | 15 W |

| Bus interface | MXM-B (3.0) | PCIe 4.0 x8 |

| Power connectors | None | None |

| DirectX | 12 (11_0) | 12 Ultimate |

| OpenGL | 4.6 | 4.6 |

| Vulkan | 1.4 | 1.4 |

| Release date | March 12, 2015 | July 14, 2024 |

| Production status | End-of-life | Active |

Head-to-Head Benchmarks

The largest recorded win for the AMD Radeon 890M comes in the Geekbench OpenCL test. The AMD part scores 37,254, the NVIDIA scores 10,272. That is a 72.4% delta in favor of AMD. This is expected: the 890M is a modern integrated GPU with FP16 support and a much higher shading unit count.

The NVIDIA GeForce GTX 950A has no head-to-head win in the database. Its highest recorded result is 10,273 in the same Geekbench OpenCL test. The two parts are close in the other tests, but the AMD part is clearly dominant in this particular workload. The numbers show the GTX 950A is a legacy mobile part, while the Radeon 890M is a current-generation integrated GPU.

The Verdict

Strictly from the recorded data, the AMD Radeon 890M is the stronger part in the measured compute workloads. It wins the only head-to-head test in the database with a 72.4% margin. The database also shows the 890M sits at the 45th percentile among all GPUs, while the GTX 950A sits at the 48th percentile. The AMD part has a higher average benchmark score, 9,210 across the suite, versus 10,273 for the NVIDIA part.

The NVIDIA GeForce GTX 950A is a specific product: a Maxwell mobile GPU with a fixed 2GB memory configuration, designed for the MXM module. It will be a good fit for older laptop systems that want a known quantity with a fixed 32.03 GB/s memory bandwidth and a fixed 1,438.7 GFLOPS FP32 range. It has a production status of end-of-life.

The AMD Radeon 890M is a different animal. It is an active, system-integrated GPU with a very high FP32 throughput of 5.939 TFLOPS, 16 ray tracing cores, and a much newer feature set: DirectX 12 Ultimate, Vulkan 1.4, and OpenGL 4.6. The data shows it is the part to choose if the task can use FP16 and ray tracing or if the workload is OpenCL heavy. The NVIDIA part remains a reasonable choice for a legacy MXM slot, but for raw compute the AMD Radeon 890M dominates the benchmark record.