NVIDIA GeForce GTX 850M vs NVIDIA GeForce GTX 880M Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is Geekbench OpenCL, and the result strongly favors the GTX 850M. The 850M scores 9,821 points, while the 880M scores 5,622 points. That is a 74.7% advantage for the smaller, lower-power chip. This is a striking outcome because the 880M carries far more hardware on paper, yet the recorded OpenCL workload clearly responds better to the 850M's architecture and driver execution.

Looking at the broader database averages, the 850M sits at 9,302 points, which places it in the 46th percentile of all GPUs. The 880M averages 8,040 points, putting it in the 42nd percentile. The relative positions are close, only four percentile points apart, but the average benchmark score gap of 1,262 points is meaningful. The 850M's nearest rivals include the GTX 465 (9,294, delta 0.1%), the Radeon R7 M380 (9,313, delta -0.1%), the GTX 960 (9,273, delta 0.3%), and the Vega 8 (9,221, delta 0.9%). These are all within a single percentage point, meaning the 850M sits at the center of a tightly packed cluster of mobile and desktop parts.

The 880M's nearest rivals tell a different story. The Quadro P5000 matches it almost exactly at 8,039 points (delta 0%), while the GTX 650 Ti (8,053, delta -0.2%), GTX 650 Ti Boost (8,067, delta -0.3%), and GRID K2 (8,080, delta -0.5%) are all slightly ahead in score, though the negative deltas mean the 880M trails each of those parts by a fraction of a point. The 880M is effectively at parity with a Quadro workstation GPU and slightly behind some midrange desktop cards.

The single head-to-head metric, with one win for the 850M and zero for the 880M, mirrors what the scores indicate: the 850M is the stronger performer in the database, even though its technical specifications suggest it should be the lesser chip. Benchmark results indicate that raw hardware counts alone, such as shading units and memory bandwidth, do not automatically translate into higher scores in every workload. The OpenCL test, in particular, shows a massive swing toward the Maxwell-based part.

Architecture Differences

The GTX 850M uses the GM107 chip, built on the Maxwell architecture, fabricated by TSMC on a 28 nm process. The die contains 1,870 million transistors spread over 148 mm², giving a transistor density of 12.6 million per mm². In contrast, the GTX 880M uses the GK104 chip, built on the Kepler architecture, also fabricated by TSMC at 28 nm. The GK104 die is much larger at 294 mm², with 3,540 million transistors, for a density of 12.0 million per mm². The 880M has roughly 89% more transistors on a die almost exactly twice the area of the 850M's silicon.

The 850M has 640 shading units, 40 texture mapping units, and 16 ROPs. The 880M has 1,536 shading units, 128 TMUs, and 32 ROPs. That is 2.4 times the shading units, 3.2 times the TMUs, and double the ROPs compared to the 850M. The pixel rate for the 850M is 14.43 GPixel/s, while the 880M reaches 31. The texture rate is 36 GT/s versus 127. The 880M offers roughly 1.7 times the pixel throughput and about 3.5 times the texture throughput. FP32 performance is 1,154.6 GFLOPS for the 850M, while the 880M reaches 3,050 TFLOPS. The 880M has no base clock listed in the database for the 850M, but the 880M runs at a base clock of 954 MHz and boosts to 993 MHz.

Memory configurations diverge sharply. The 850M uses 2 GB of DDR3 memory on a 128-bit bus, with 32.03 GB/s bandwidth. The 880M uses 8 GB of GDDR5 memory on a 256-bit bus, at 1,250 MHz, 5 Gbps effective, delivering 160 GB/s bandwidth. Memory bandwidth is 5 times higher on the 880M. The 850M has no base clock listed in the database. The 880M memory clock is not listed for the 850M beyond 1001 MHz, 2 Gbps effective.

Both chips support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 4.1. The 880M supports Vulkan 1.2.175, while the 850M lists Vulkan 1. The 850M lists no power connectors, TDP of 45 W for the 850M, 122 W for the 880M. The 850M uses an IGP slot width, 850M uses no power connectors, 880M uses MXM Module. Both have end-of-life production status, both released on 2014-03-11, both predecessor to GeForce 700M, successor to GeForce 900M.

FAQ

Q: Which GPU has higher average benchmark score?

A: GTX 850M averages 9,302 points, GTX 880M averages 8,040 points.

Q: What is the OpenCL score for each?

A: GTX 850M scores 9,821, GTX 880M scores 5,622, a 74.7% gap.

Q: How does memory compare?

A: GTX 850M has 2 GB DDR3 over 128-bit, 32.03 GB/s bandwidth; GTX 880M has 8 GB GDDR5 over 256-bit, 160.0 GB/s bandwidth.

Q: What about power draw?

A: GTX 850M is 45 W TDP, GTX 880M is 122 W TDP.

Q: Which chip is larger?

A: GTX 880M's GK104 die is 294 mm² with 3,540 million transistors; GTX 850M's GM107 is 148 mm² with 1,870 million transistors.

Q: Are they the same generation?

A: Both are GeForce 800M, both released on 2014-03-11, both end-of-life, both succeeded by GeForce 900M.

Specification Differences

| Specification | GTX 850M | GTX 880M |

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

| Chip | GM107 | GK104 |

| Architecture | Maxwell | Kepler |

| Process Node | 28 nm | 28 nm |

| Transistors | 1,870 million | 3,540 million |

| Die Size | 148 mm² | 294 mm² |

| Transistor Density | 12.6M / mm² | 12.0M / mm² |

| Base Clock | Not listed | 954 MHz |

| Boost Clock | Not listed | 993 MHz |

| Memory Clock | 1001 MHz, 2 Gbps | 1250 MHz, 5 Gbps |

| Memory Size | 2 GB | 8 GB |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus Width | 128 bit | 256 bit |

| Memory Bandwidth | 32.03 GB/s | 160.0 GB/s |

| Shading Units | 640 | 1536 |

| TMUs | 40 | 128 |

| ROPs | 16 | 32 |

| Pixel Rate | 14.43 GPixel/s | 31.78 GPixel/s |

| Texture Rate | 36.08 GTexel/s | 127.1 GTexel/s |

| FP32 | 1,154.6 GFLOPS | 3.050 TFLOPS |

| TDP | 45 W | 122 W |

| Slot Width | IGP | MXM Module |

| Power Connectors | None | None |

| Bus Interface | PCIe 3.0 x16 | MXM-B (3.0) |

| Vulkan | 1.4 | 1.2.175 |

The Verdict

The data points in opposite directions depending on the metric. On paper, the GTX 880M is the far more powerful GPU. It has 2.4 times the shading units, 3.2 times the TMUs, double the ROPs, 5 times the memory bandwidth, and 2.6 times the FP32 throughput. Its TDP of 122 W reflects that hardware. The GTX 850M, by contrast, is a 45 W part with a much smaller die and modest memory configuration.

However, the recorded database scores tell a different story. The 850M wins the only head-to-head benchmark by 74.7%, and its average benchmark score is 1,262 points higher than the 880M's. The 850M also ranks higher in percentile, 46th versus 42nd. If the goal is to pick a GPU based purely on measured performance in the database, the 850M is the better choice.

The 880M's closest rivals are desktop and workstation parts, including the Quadro P5000 and GTX 650 Ti series, and it slightly trails all of them. The 850M's rivals include the GTX 960 and Vega 8, and it slightly leads or matches them. This suggests that in real-world database workloads, the 850M behaves more like a midrange desktop GPU, while the 880M, despite its massive memory and core count, does not convert that hardware into higher scores.

For a laptop buyer looking at these two options, the 850M offers better measured performance with a fraction of the power draw. The 880M offers more memory capacity and bandwidth, which could matter for specific workloads, but the benchmark data does not show that advantage materializing. The verdict from the database is clear: the GTX 850M outperforms the GTX 880M in recorded tests, and it does so at 45 W instead of 122 W.

Where Each One Wins

The GTX 850M wins in every recorded benchmark category. It has the higher OpenCL score, higher average benchmark score, and higher percentile placement. It also wins on transistor density, 12.6M / mm² versus 12.0M / mm², meaning it packs more transistors per area on a smaller die. For any workload represented by the database's Geekbench tests, the 850M is the faster GPU.

The GTX 880M is not without advantages, but those advantages are in specifications rather than measured performance. It has 8 GB of GDDR5 memory versus 2 GB of DDR3, which gives it 5 times the memory bandwidth. It has 3.2 times the texture fill rate and 2.6 times the FP32 compute. Its MXM form factor and 122 W TDP indicate it is designed for larger, more power-hungry notebooks, while the 850M's IGP slot width and 45 W TDP suit thinner systems.

If the use case involves massive frame buffers, such as driving very high-resolution displays or storing large textures locally, the 880M's 8 GB GDDR5 pool is the clear spec-sheet winner. If the use case involves general-purpose compute or any workload that resembles the OpenCL benchmark in the database, the 850M is the stronger part. The 880M also supports Vulkan 1.2.175, while the 850M lists Vulkan 1.4, so the 850M has the newer Vulkan API version.

In terms of nearest rivals, the 850M competes with the GTX 960 and Vega 8, both of which it slightly beats or matches, while the 880M sits just below the GTX 650 Ti and GTX 650 Ti Boost. This means the 850M belongs to a faster performance tier in the database, despite having less memory and fewer cores. Users who prioritize measured performance and efficiency should pick the 850M. Users who need the 880M's memory capacity and raw fill rates for specific applications should consider it, but they must accept lower benchmark scores and higher power consumption.