AMD Radeon Vega 8 vs NVIDIA GeForce GTX 880M Comparison

# AMD Radeon Vega 8 vs NVIDIA GeForce GTX 880M

The comparison between the AMD Radeon Vega 8 integrated graphics processor and the NVIDIA GeForce GTX 880M mobile discrete part is a study in contrasting design goals. The Vega 8 is a modern integrated solution built on the Raven Ridge APU, while the GTX 880M is a high-end Kepler-based mobile discrete part from the previous generation. The benchmark data in the database shows a clear overall winner, but the margin varies significantly depending on the workload.

Head-to-Head Benchmarks

The database includes two head-to-head benchmark comparisons between these parts. The results are decisive, though the magnitude of the victory changes dramatically between the tests.

In the geekbench_metal test, the AMD Radeon Vega 8 scores 10,706 points, while the NVIDIA GeForce GTX 880M scores 10,458 points. This gives the AMD part a narrow 2.4% lead. This is a close result, suggesting near parity in compute workloads that scale well with the modern feature set of the Vega architecture.

The geekbench_opencl test tells a different story. Here, the AMD Radeon Vega 8 scores 8,822 points, against 5,622 for the GTX 880M. This works out to a massive 56.9% advantage for the AMD part. OpenCL workloads can stress raw compute throughput, memory bandwidth, and driver efficiency, and the data indicates the Vega 8 has a clear edge in these areas.

Overall, the AMD Radeon Vega 8 wins both head-to-head matchups. The database records 2 wins for the AMD part and 0 wins for the NVIDIA part. The average score across all recorded benchmarks shows the AMD Radeon Vega 8 at 9,221 points, placing it in the 45th percentile of all GPUs. The NVIDIA GeForce GTX 880M, on the other hand, has an average score of 8,040 points, which places it in the 42nd percentile.

FAQ

Q: How does the AMD Radeon Vega 8 compare to the NVIDIA GeForce GTX 880M in synthetic benchmarks?

A: In the recorded head-to-head data, the AMD Radeon Vega 8 wins in both recorded tests. The largest win is in the geekbench_opencl test, where it leads by 56.9% (8,822 versus 5,622). In the geekbench_metal test, the AMD part is 2.4% faster (10,706 versus 10,458).

Q: What is the performance difference relative to the nearest rivals of each GPU?

A: The AMD Radeon Vega 8 has an average score of 9,221, which is 0.1% higher than the AMD Radeon 890M (9,210) and 0.6% higher than the GeForce GTX 960 (9,273). It is also 0.8% higher than the GeForce GTX 465 (9,294) and 0.9% higher than the GeForce GTX 850M (9,302). The GeForce GTX 880M averages 8,040 points, which is 0% away from the NVIDIA Quadro P5000 (8,039) and 0.2% below the GeForce GTX 650 Ti (8,053).

Q: Which GPU has the higher raw compute throughput, the Vega 8 or the GTX 880M?

A: Based on the recorded data, the AMD Radeon Vega 8 shows a significant lead in FP32 compute with 1,126.4 GFLOPS, while the NVIDIA GeForce GTX 880M is rated at 3.050 TFLOPS. The Vega 8 also supports FP16 at 2.253 TFLOPS, a feature not present on the older NVIDIA part.

Architecture Differences

The two chips come from different architectural schools. The AMD Radeon Vega 8 is built on the Graphics Core Next (GCN) 5.0 architecture with a Raven Ridge integrated graphics processor. This is AMD's 5th generation GCN design, built on a 14nm process at GlobalFoundries. The die size is 210 mm² and it holds 4,940 million transistors, which gives a transistor density of 23.5 million per square millimeter.

In contrast, the NVIDIA GeForce GTX 880M is a Kepler architecture part with the GK104 chip. This is a 28nm part fabricated at TSMC. The die size is 294 mm² with 3,540 million transistors. This results in a density of 12.0 million transistors per square millimeter, significantly lower than the AMD chip.

The AMD chip has 512 shading units, 32 texture mapping units, and 8 raster output units. The NVIDIA also has a higher count of 1536 shading units, 128 texture mapping units, and 32 ROPs. Both chips support DirectX 12 (11_0 for NVIDIA), OpenGL 4.6, and Vulkan 1.3 (1.2.175 for NVIDIA). The AMD has a pixel rate of 8.800 GPixel/s and a texture rate of 35.20 GTexel/s. The GeForce achieves 31.78 GPixel/s pixel rate and 127.1 GTexel/s texture rate.

Specification Differences

The differences between these two products go beyond the specification sheet. The AMD Radeon Vega 8 is an integrated graphics processor (IGP) with a slot width of IGP. Its memory is system shared, meaning it uses the main system RAM. Consequently, the memory bus width, bandwidth, and type are all system-dependent. The power connector is listed as none. The maximum board power consumption is 25 W.

In contrast, the NVIDIA GeForce GTX 880M is a discrete MXM module. It has 8 GB of dedicated GDDR5 memory on a 256-bit bus. The memory operates at 1250 MHz with an effective bandwidth of 160.0 GB/s. The module uses the MXM-B (3.0) bus interface. It is a portable device dependent on the output. Its power draw is 122W and also uses no power connectors. A suggested power supply is not listed.

Where each one wins

Based on the benchmark wins, there are two different usage segments where each product excels.

The AMD Radeon Vega 8 claims wins in both recorded tests, but the size of the lead points to its strengths. The first, Geekbench Metal, is a close race. The AMD part is just 2.4% ahead. This indicates that in a modern API workload, the two are in the same performance class. The AMD's newer architecture likely helps to offset a lower shading unit and a narrower memory bus.

The Geekbench OpenCL test is the exact opposite. The AMD Radeon Vega 8 is 56.9% ahead. This is a huge gap. This test tends to capture raw parallel throughput and driver efficiency. The AMD GCN 5.0 architecture is clearly better suited to this type of workload. The NVIDIA GeForce GTX 880M, despite having 3 times more shading units, does not translate that advantage into this API. Its older Kepler architecture and its mobile power envelope hold it back.

The NVIDIA GeForce GTX 880M still claims a relevant advantage in the traditional rasterization pipeline, where its specification gives it an edge. With 1536 shading units, 128 texture units and 32 ROPs, its specification sheet is much more robust. The AMD part with 512 shaders, 32 TMUs and 8 ROPs is a more modest configuration. For a GPU user engaged in heavy geometric rasterization, the NVIDIA part is often the more attractive option. The AMD Vega 8, on the other hand, makes more sense for compute-oriented tasks and OpenCL-style workloads where its higher FP32 rate and memory sharing can be used.

The Verdict

The data points to a clear, if nuanced, conclusion. The AMD Radeon Vega 8 is the better synthetic compute in the database. It wins both head-to-head matches and does so with a particularly strong lead in OpenCL. Its average score of 9,221 is also higher than the 8,040 average of the GTX 880M, and it sits at a higher percentile relative to all GPUs. For users who are primarily concerned with compute workloads, OpenCL, or shader-heavy game, the AMD Radeon Vega 8 is the more capable part.

The NVIDIA GeForce GTX 880M remains the better choice for a specific set of tasks. Its larger set of shading units, 3x more, and its dedicated GDDR5 memory with high bandwidth, make it a stronger candidate for traditional game rendering and high resolution rasterization workloads. The AMD is a modern, power-efficient integrated GPU, but it is a best fit in compute, not the more traditional GPU rendering space. Pick the Radeon Vega 8 for compute and OpenCL. For game rasterization and graphics heavy workloads, the GTX 880M remains the more balanced part.