AMD Radeon 880M vs NVIDIA GeForce GTX 870M Comparison

Where Each One Wins

The recorded data splits these two mobile graphics solutions into very different usage profiles. The NVIDIA GeForce GTX 870M, a Kepler-based discrete part from 2014, and the AMD Radeon 880M, a 2024 RDNA 3.5 integrated processor, share only a single direct benchmark comparison in the database. That comparison, Geekbench OpenCL, goes decisively to the AMD part. The AMD Radeon 880M scores 31,285 versus 12,630 for the GTX 870M, a delta of 59.6% in favor of the newer chip. This is the only head-to-head test recorded, so the AMD GPU takes the sole win in that category.

Beyond that single direct test, the broader benchmark suite reveals where each part excels. The AMD Radeon 880M holds a significant advantage in compute-oriented workloads. Its OpenCL score of 31,285 dwarfs the GTX 870M's 12,630, and its Vulkan score reaches 40,006. The Geekbench Metal score for the NVIDIA part is 7,288, which indicates some Apple ecosystem compute capability, but the AMD part's raw throughput in OpenCL and Vulkan suggests it is better suited to general-purpose GPU compute tasks, especially those that leverage modern API features.

The NVIDIA GeForce GTX 870M, despite its age, still posts competitive numbers in the database's aggregate scoring. Its average benchmark score is 9,959, placing it in the 48th percentile of all GPUs. The AMD Radeon 880M averages 8,436, landing in the 43rd percentile. This means that when averaging across all recorded tests, the older NVIDIA discrete GPU actually edges out the newer integrated AMD solution by about 15% in average score. However, that average is heavily influenced by the types of tests each part was subjected to. The GTX 870M's nearest rivals in average score include the AMD Radeon Pro 5300M at 10,013 (0.5% ahead), the NVIDIA Quadro K5100M at 10,043 (0.8% ahead), and the AMD Radeon R9 M375 at 10,070 (1.1% ahead). It also sits 1.1% ahead of the NVIDIA Quadro 6000, which scores 9,846.

The AMD Radeon 880M's nearest rivals tell a different story. It sits nearly level with the NVIDIA GeForce GTX 675MX (8,427, a 0.1% delta), the NVIDIA GeForce MX330 (8,458, 0.3% behind), and the AMD Radeon HD 8870M (8,462, 0.3% behind). It is 1.3% ahead of the AMD Radeon R9 M375X, which scores 8,325. This clustering suggests the 880M is competitive with older mid-range discrete mobile GPUs in aggregate performance, while the GTX 870M sits in a slightly higher aggregate tier.

The use-case split is clear. For compute-heavy workloads, particularly those that scale well with modern shader counts and API features, the AMD Radeon 880M is the stronger choice. For legacy DirectX 11-era gaming or older applications where raw pixel throughput and established driver maturity matter, the GTX 870M's 27.08 GPixel/s pixel rate and its higher average benchmark percentile may hold appeal. The AMD part counters with a 46.40 GPixel/s pixel rate and a 139.2 GTexel/s texture rate, both substantially higher than the NVIDIA's 108.3 GTexel/s. The data suggests that the 880M is the more future-proof part for modern APIs, while the 870M remains a serviceable older discrete option.

Architecture Differences

The architectural gap between these two GPUs is generational and fundamental. The NVIDIA GeForce GTX 870M uses the GK104 chip, built on Kepler architecture, manufactured on a 28 nm process at TSMC. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0 million per square millimeter. The AMD Radeon 880M uses the Strix Point chip, based on RDNA 3.5, built on a 4 nm process, also at TSMC. It contains 34,000 million transistors on a 233 mm² die, achieving a transistor density of 145.9 million per square millimeter. That is more than twelve times the density of the Kepler part, a direct reflection of the decade-long process node advancement.

The memory subsystems are entirely different. The GTX 870M has a dedicated 3 GB of GDDR5 memory on a 192-bit bus, delivering 120.0 GB/s of bandwidth. The 880M uses system shared memory, with its size, type, bus width, and bandwidth all listed as system dependent. This is the classic distinction between a discrete GPU with its own VRAM and an integrated GPU that borrows from the host system's memory pool. The GTX 870M's memory clock is 1250 MHz, or 5 Gbps effective, while the 880M's memory clock is simply labeled "System Shared."

The compute configurations differ sharply. The GTX 870M has 1,344 shading units, 112 texture mapping units, and 24 ROPs. The 880M has 768 shading units, 48 TMUs, and 16 ROPs. Despite having fewer of each, the AMD part achieves higher fill rates due to its much higher clock speeds. The 880M's base clock is 400 MHz, but it boosts to 2,900 MHz, compared to the GTX 870M's 941 MHz base and 967 MHz boost. The pixel rate for the 880M is 46.40 GPixel/s versus 27.08 GPixel/s for the GTX 870M, and the texture rate is 139.2 GTexel/s versus 108.3 GTexel/s. The AMD part also has 12 ray tracing cores, which the NVIDIA Kepler part lacks entirely, and it supports DirectX 12 Ultimate (12_2) versus the GTX 870M's DirectX 12 (11_0).

The FP32 throughput is notably higher on the AMD side: 4.454 TFLOPS versus 2.599 TFLOPS. The AMD part also offers FP16 at 4.454 TFLOPS (1:1), while the GTX 870M has no recorded FP16 capability. Power consumption is a major differentiator. The GTX 870M has a TDP of 100 W and is an MXM module, requiring no external power connectors. The 880M has a TDP of 15 W and is an IGP, also with no power connectors. The bus interface differs as well: the GTX 870M uses MXM-B (3.0), while the 880M uses PCIe 4.0 x8. The 880M supports Vulkan 1.4 and OpenGL 4.6, and the GTX 870M also supports OpenGL 4.6 but only Vulkan 1.2.175.

The production statuses reflect their lifecycles. The GTX 870M is end-of-life, released on March 11, 2014, with its predecessor being the GeForce 700M and its successor the GeForce 900M. The 880M is active, released on July 14, 2024, succeeding the Navi II IGP with no successor listed. The 880M's API support for ray tracing and DirectX 12 Ultimate is a clear architectural leap over the Kepler design, which predates hardware ray tracing entirely.

Head-to-Head Benchmarks

The only direct benchmark comparing these two GPUs in the database is Geekbench OpenCL. In this test, the AMD Radeon 880M scores 31,285, while the NVIDIA GeForce GTX 870M scores 12,630. The delta percentage is 59.6% in favor of the AMD part, meaning the 880M outperforms the 870M by nearly two-thirds in this compute workload. This is a substantial margin, and it highlights the generational improvement in compute efficiency and raw throughput. The 880M's FP32 count of 4.454 TFLOPS versus 2.599 TFLOPS for the GTX 870M explains much of this gap, as OpenCL workloads often scale with shader throughput.

Looking at the broader benchmark landscape, the GTX 870M's Geekbench Metal score of 7,288 is the only other compute-oriented figure recorded for that part. The AMD 880M has a much wider range of recorded tests. Its Vulkan score of 40,006 is particularly strong, and its Passmark scores show a mixed profile: G3D at 7,615, GPU Compute at 3,719, G2D at 969, DirectX 9 at 97, DirectX 11 at 73, DirectX 12 at 32, and DirectX 10 at 31. These Passmark figures are low in absolute terms, but they likely reflect the integrated nature of the GPU and the specific test conditions, not necessarily real-world gaming performance. The 3DMark Steel Nomad DX12 score of 535 is a modern benchmark result that the GTX 870M has no equivalent for, since Kepler does not support DirectX 12 Ultimate features.

The average benchmark scores put the GTX 870M ahead at 9,959 versus 8,436 for the 880M, a 15.3% advantage for the NVIDIA part. However, this average is derived from only two tests for the NVIDIA part (Geekbench Metal and OpenCL) versus ten tests for the AMD part. The single head-to-head test favors AMD heavily, but the aggregate data suggests that the GTX 870M still holds its own in certain legacy workloads. The percentile ranking reinforces this: the GTX 870M sits at the 48th percentile of all GPUs, while the 880M sits at the 43rd. This means the GTX 870M outperforms a slightly larger share of the GPU population in the database's overall scoring, despite being a decade older and having far less advanced hardware.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 870M has an average benchmark score of 9,959, which is 15.3% higher than the AMD Radeon 880M's average of 8,436. The GTX 870M also ranks higher in percentile, sitting at 48 versus 43 for the 880M.

Q: What is the result of the direct OpenCL comparison?

A: The AMD Radeon 880M scores 31,285 in Geekbench OpenCL, while the NVIDIA GeForce GTX 870M scores 12,630. This gives the AMD part a 59.6% advantage in that workload.

Q: Does the GTX 870M support ray tracing?

A: No. The GTX 870M has no ray tracing cores and supports DirectX 12 (11_0). The AMD Radeon 880M has 12 ray tracing cores and supports DirectX 12 Ultimate (12_2).

Q: How do their power requirements compare?

A: The GTX 870M has a TDP of 100 W, while the AMD Radeon 880M has a TDP of 15 W. Neither requires external power connectors, but the GTX 870M is an MXM module, whereas the 880M is an IGP.

Q: What are the memory configurations?

A: The GTX 870M has 3 GB of GDDR5 on a 192-bit bus with 120.0 GB/s bandwidth. The 880M uses system shared memory, with all memory parameters listed as system dependent.

Q: Which GPU has a higher pixel rate?

A: The AMD Radeon 880M has a pixel rate of 46.40 GPixel/s, compared to 27.08 GPixel/s for the GTX 870M. The 880M also has a higher texture rate at 139.2 GTexel/s versus 108.3 GTexel/s.

Specification Differences

The two GPUs differ across nearly every specification field recorded in the database. The chip and architecture are entirely distinct: the NVIDIA GeForce GTX 870M uses the GK104 chip on Kepler architecture, while the AMD Radeon 880M uses the Strix Point chip on RDNA 3.5. The process node is 28 nm for the NVIDIA part and 4 nm for the AMD part, both fabricated by TSMC. Transistor counts differ massively: 3,540 million for the GTX 870M versus 34,000 million for the 880M. Die size is 294 mm² for the NVIDIA part and 233 mm² for the AMD part, with transistor densities of 12.0M per mm² and 145.9M per mm², respectively.

Clock speeds are dramatically different. The GTX 870M runs at a base of 941 MHz and a boost of 967 MHz, with a memory clock of 1250 MHz (5 Gbps effective). The 880M has a base clock of 400 MHz and a boost of 2,900 MHz, with memory clock listed as system shared. The memory configurations are also fundamentally different: the GTX 870M has 3 GB of GDDR5 on a 192-bit bus with 120.0 GB/s bandwidth, while the 880M has system shared memory with system dependent bandwidth and bus width.

The compute unit counts differ: the GTX 870M has 1,344 shading units, 112 TMUs, and 24 ROPs, while the 880M has 768 shading units, 48 TMUs, and 16 ROPs. The 880M has 12 ray tracing cores; the GTX 870M has none. Fill rates favor the AMD part: pixel rate is 46.40 GPixel/s versus 27.08 GPixel/s, and texture rate is 139.2 GTexel/s versus 108.3 GTexel/s. FP32 throughput is 4.454 TFLOPS for the 880M versus 2.599 TFLOPS for the GTX 870M. The 880M also has FP16 at 4.454 TFLOPS (1:1), while no FP16 figure exists for the GTX 870M.

Power and form factor differ substantially. The GTX 870M has a TDP of 100 W and uses an MXM Module slot, while the 880M has a TDP of 15 W and is an IGP. Neither requires power connectors. The bus interface is MXM-B (3.0) for the NVIDIA part and PCIe 4.0 x8 for the AMD part. Display outputs are listed as portable device dependent for both. API support differs: the GTX 870M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175, while the 880M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Production status and release dates are far apart. The GTX 870M is end-of-life, released on March 11, 2014, with a predecessor of GeForce 700M and a successor of GeForce 900M. The 880M is active, released on July 14, 2024, with a predecessor of Navi II IGP and no successor listed. No launch MSRP is recorded for either part. The GTX 870M's benchmarks include Geekbench Metal at 7,288 and Geekbench OpenCL at 12,630. The 880M's benchmarks include 3DMark Steel Nomad DX12 at 535, Geekbench OpenCL at 31,285, Geekbench Vulkan at 40,006, and a series of Passmark scores ranging from 31 (DirectX 10) to 969 (G2D).