AMD Radeon 880M vs NVIDIA GeForce GTX 675MX Comparison

The AMD Radeon 880M and NVIDIA GeForce GTX 675MX land within 0.1% of each other in average benchmark score (8436 vs 8427), placing both at the 43rd percentile of all GPUs. That statistical dead heat, however, masks a generational chasm in how each card achieves its results. The data shows a single head-to-head comparison in Geekbench OpenCL, where the 880M posts a 191.8% advantage over the GTX 675MX (31285 vs 10723). This is the only direct benchmark overlap in the FACT PACK, making it the definitive data point for their compute performance relationship.

Head-to-Head Benchmarks

The sole shared benchmark is Geekbench OpenCL, and the result is not close. The AMD Radeon 880M scores 31285, nearly triple the GTX 675MX’s 10723. That translates to a 191.8% delta — the 880M delivers roughly three times the OpenCL compute throughput of the older NVIDIA part. For context, the GTX 675MX’s nearest rivals in the FACT PACK include the 880M itself (deltaPct -0.1) and the NVIDIA GeForce MX330 (deltaPct -0.4), meaning the 675MX is effectively mid-pack among its peers. The 880M, by contrast, sits 0.1% above the 675MX in average score, which is a statistical tie, yet its OpenCL output is in a different performance tier entirely.

The 880M also holds a decisive advantage in raw shading resources. It features 768 shading units, 48 texture mapping units, and 16 ROPs, while the GTX 675MX counters with 960 shading units, 80 TMUs, and 32 ROPs. Despite having fewer execution units, the 880M achieves a far higher FP32 throughput of 4.454 TFLOPS versus the 675MX’s 1,255.7 GFLOPS. Pixel rate tells a similar story: 46.40 GPixel/s for the 880M against 13.08 GPixel/s for the 675MX. Texture rate is 139.2 GTexel/s versus 52.32 GTexel/s. In every throughput metric listed, the newer architecture wins decisively, even with a 20% reduction in shader count and a 40% reduction in TMUs.

The GTX 675MX does not win any head-to-head benchmarks in the FACT PACK; the wins tally is 1 for the 880M and 0 for the 675MX. The 675MX’s only listed benchmarks — Geekbench Metal (6131) and Geekbench OpenCL (10723) — show it performing in line with its 43rd percentile ranking, but the 880M’s OpenCL score of 31285 is more than double any figure the 675MX produces. This is a one-sided comparison in compute workloads, though the 675MX’s dedicated 2GB of GDDR5 memory with a 256-bit bus and 115.2 GB/s bandwidth could mitigate some real-world gaming gaps that raw compute benchmarks do not capture.

Architecture Differences

The architectural divide is stark. The 880M uses RDNA 3.5 on a 4 nm TSMC process, while the GTX 675MX uses Kepler on a 28 nm TSMC node. Transistor counts reflect the process gap: 34,000 million transistors in the 880M’s Strix Point chip versus 3,540 million in the 675MX’s GK104. Die size is smaller for the newer part (233 mm² versus 294 mm²), yet transistor density is 145.9M per mm² versus just 12.0M per mm² — a 12x density advantage. The 880M is an integrated graphics processor (IGP) on a 15 W TDP, whereas the 675MX is a discrete MXM module rated at 100 W.

The 880M includes 12 ray tracing cores, a feature the Kepler-based 675MX lacks entirely. The 880M supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6, while the 675MX peaks at DirectX 12 (11_0), Vulkan 1.2.175, and OpenGL 4.6. The 880M’s memory is system-shared with no dedicated VRAM, relying on the host system’s RAM and PCIe 4.0 x8 interface. The 675MX has 2GB of GDDR5 on a 256-bit bus with 115.2 GB/s bandwidth. Clock behavior differs fundamentally: the 880M boosts to 2900 MHz from a 400 MHz base, while the 675MX lists no base or boost clock, only a 900 MHz memory clock (3.6 Gbps effective).

The 880M is built on the Navi III IGP generation (Strix Point Mobile), succeeding Navi II IGP, and remains in active production. The 675MX belongs to the GeForce 600M series, succeeded the GeForce 500M, and was itself succeeded by GeForce 700M; it is now end-of-life. The 675MX launched in September 2012, while the 880M arrived in July 2024 — a nearly 12-year gap that explains the feature and efficiency disparity. The 880M’s FP16 throughput matches its FP32 at 4.454 TFLOPS (1:1), while the 675MX has no listed FP16 capability.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon 880M scores 8436 on average, which is 0.1% higher than the NVIDIA GeForce GTX 675MX’s 8427. Both sit at the 43rd percentile of all GPUs.

Q: How large is the performance gap in OpenCL compute?

A: In the Geekbench OpenCL test, the 880M scores 31285 versus the 675MX’s 10723, a 191.8% advantage for the AMD part.

Q: Does the GTX 675MX have any benchmark where it wins?

A: No. The head-to-head tally shows 1 win for the AMD Radeon 880M and 0 wins for the NVIDIA GeForce GTX 675MX.

Q: Why does the 880M have fewer shading units yet higher performance?

A: The 880M has 768 shading units versus 960 on the 675MX, but its RDNA 3.5 architecture on a 4 nm node boosts FP32 to 4.454 TFLOPS versus 1,255.7 GFLOPS. Higher clocks (2900 MHz boost) and newer instruction efficiency overcome the unit deficit.

Q: What are the memory configurations?

A: The 880M uses system-shared memory with bandwidth dependent on the host system. The 675MX has 2GB of GDDR5 on a 256-bit bus with 115.2 GB/s bandwidth.

Q: Which GPU supports ray tracing?

A: Only the AMD Radeon 880M, which includes 12 ray tracing cores. The NVIDIA GeForce GTX 675MX has no ray tracing hardware.

Q: What are the power requirements?

A: The 880M is an IGP rated at 15 W, while the 675MX is a 100 W MXM module. Neither requires external power connectors.

Specification Differences

| Specification | AMD Radeon 880M | NVIDIA GeForce GTX 675MX |

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

| Architecture | RDNA 3.5 | Kepler |

| Process Node | 4 nm | 28 nm |

| Transistors | 34,000 million | 3,540 million |

| Die Size | 233 mm² | 294 mm² |

| Transistor Density | 145.9M / mm² | 12.0M / mm² |

| Shading Units | 768 | 960 |

| TMUs | 48 | 80 |

| ROPs | 16 | 32 |

| RT Cores | 12 | None |

| Pixel Rate | 46.40 GPixel/s | 13.08 GPixel/s |

| Texture Rate | 139.2 GTexel/s | 52.32 GTexel/s |

| FP32 | 4.454 TFLOPS | 1,255.7 GFLOPS |

| FP16 | 4.454 TFLOPS (1:1) | None listed |

| TDP | 15 W | 100 W |

| Slot Width | IGP | MXM Module |

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

| Memory Size | System Shared | 2 GB |

| Memory Type | System Shared | GDDR5 |

| Memory Bus | System Shared | 256 bit |

| Memory Bandwidth | System Dependent | 115.2 GB/s |

| DirectX | 12 Ultimate (12_2) | 12 (11_0) |

| Vulkan | 1.4 | 1.2.175 |

| Production Status | Active | End-of-life |

| Release Date | 2024-07-14 | 2012-09-30 |

| Predecessor | Navi II IGP | GeForce 500M |

| Successor | None | GeForce 700M |

Where Each One Wins

The AMD Radeon 880M wins decisively in compute throughput. Its OpenCL score of 31285 versus 10723 represents a 191.8% lead, making it the clear choice for GPU-accelerated workloads such as rendering, machine learning inference, or any OpenCL-based application. The 880M also wins on efficiency: 15 W TDP versus 100 W, with higher pixel rate (46.40 GPixel/s), texture rate (139.2 GTexel/s), and FP32 (4.454 TFLOPS). Its 12 ray tracing cores and DirectX 12 Ultimate support make it the only option for ray-traced effects and modern API features like mesh shaders. The 4 nm process and 2900 MHz boost clock indicate superior architectural efficiency per watt.

The NVIDIA GeForce GTX 675MX wins on dedicated memory and legacy compatibility. Its 2GB of GDDR5 with 115.2 GB/s bandwidth is a fixed resource that does not compete with the host system for RAM, which can be an advantage in memory-constrained laptops or when system RAM is slow. The 675MX’s 256-bit bus is significantly wider than any system-shared configuration, potentially offering more consistent memory performance in scenarios where the host RAM bandwidth is limited. Its MXM form factor means it is a replaceable module, whereas the 880M is soldered as an IGP. The 675MX also has more shading units (960) and TMUs (80), which could benefit older games optimized for Kepler-era architectures that do not scale well with newer RDNA designs.

For gaming, the data is inconclusive beyond compute. The 880M’s higher pixel rate and texture rate suggest better rasterization throughput, but the 675MX’s dedicated VRAM provides a predictable memory pipeline. The 880M supports Vulkan 1.4 and DirectX 12 Ultimate, enabling modern titles with ray tracing; the 675MX is capped at Vulkan 1.2.175 and DirectX 12 (11_0), excluding it from next-generation graphics features. The 880M’s active production status means ongoing driver support, while the 675MX is end-of-life. For new workloads, the 880M is the superior choice; for legacy software that relies on Kepler-specific optimizations or requires discrete VRAM, the 675MX retains niche utility.