AMD Radeon HD 8950M vs NVIDIA GeForce GTX 680 Comparison

The benchmark data is unambiguous: the NVIDIA GeForce GTX 680 decisively outperforms the AMD Radeon HD 8950M in every shared test. The GTX 680 leads by 22.8% in OpenCL and by a substantial 36% in Vulkan, making it the clear performance winner across the board. While the HD 8950M holds its own in the lower tier of the GPU landscape, it cannot match the raw compute and graphics throughput of NVIDIA’s desktop-class part.

Head-to-Head Benchmarks

The GeForce GTX 680 wins every head-to-head comparison available, with no contest from the Radeon HD 8950M. In the Geekbench OpenCL test, the GTX 680 scores 16906 against the HD 8950M’s 13772, a delta of 22.8% in NVIDIA’s favor. This is not a marginal gap; it represents a substantial lead in general-purpose compute performance, likely driven by the GTX 680’s larger shader array and higher memory bandwidth. The Vulkan test tells an even more lopsided story: the GTX 680 hits 17646, while the HD 8950M manages only 12979, a 36% difference. That 36% margin is the single largest win for either card in any metric, underscoring how much faster the NVIDIA part is in modern graphics API workloads.

Looking at aggregate scores reinforces this trend. The GTX 680’s average benchmark score is 14150, which places it in the 55th percentile of all GPUs. The HD 8950M’s average is 13376, putting it just one percentile lower at 54th. Despite the proximity in percentile rank, the raw score difference of 774 points (about 5.8%) is meaningful in real-world terms. The GTX 680 also has a stronger showing in its nearest rival comparisons: it sits within 1.4% of the AMD Radeon Vega 11 and Intel Iris Xe MAX Graphics, while the HD 8950M’s nearest rival, the AMD Radeon RX 5500M, is a hair ahead at 0.2% — a much tighter grouping that shows the mobile AMD card is closer to its own peers than the GTX 680 is to its desktop-class competitors.

Where Each One Wins

The data offers no scenario where the AMD Radeon HD 8950M takes a victory. In both OpenCL and Vulkan, the GTX 680 is the outright winner, so any use case that relies on those APIs — from compute acceleration to modern gaming — will favor NVIDIA. The GTX 680’s 22.8% OpenCL advantage makes it the better choice for tasks like physics simulation, video encoding, or any workload that leverages general-purpose GPU compute. Its 36% Vulkan lead is even more pronounced, suggesting that games or applications built on newer graphics APIs will run noticeably smoother on the NVIDIA card.

For the HD 8950M, the only place it “wins” is in efficiency relative to its form factor. It is a mobile part with a 100 W TDP, compared to the GTX 680’s 195 W, and it uses an MXM module slot rather than a dual-slot desktop cooler. That makes it a more practical option for laptops or compact systems where power and space are constrained. However, that is a physical design advantage, not a performance one. Benchmark results indicate that for any task where raw frame rates or compute throughput matter, the GTX 680 is the only rational pick. The HD 8950M’s lower pixel rate (17.20 GPixel/s vs. 33.86 GPixel/s) and texture rate (51.60 GTexel/s vs. 135.4 GTexel/s) further reinforce that it is simply outclassed in every performance dimension measured.

Architecture Differences

The two GPUs come from different design philosophies and eras. The NVIDIA GeForce GTX 680 is built on the Kepler architecture with the GK104 chip, manufactured on a 28 nm process at TSMC. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0M per mm². The AMD Radeon HD 8950M uses the GCN 2.0 architecture with the Saturn chip, also on TSMC’s 28 nm node, but with only 2,080 million transistors on a much smaller 160 mm² die, achieving a higher density of 13.0M per mm². That density advantage for AMD is notable, but it does not translate into performance wins here.

The core configurations diverge sharply. The GTX 680 has 1536 shading units, 128 texture mapping units, and 32 ROPs. The HD 8950M is less than half of that in most respects: 768 shaders, 48 TMUs, and just 16 ROPs. This explains the massive throughput gaps. The GTX 680’s pixel rate is 33.86 GPixel/s versus 17.20 GPixel/s for AMD, and its texture rate of 135.4 GTexel/s is more than double the HD 8950M’s 51.60 GTexel/s. Floating-point performance follows the same pattern: the GTX 680 delivers 3.250 TFLOPS of FP32 compute, while the HD 8950M manages just 1.651 TFLOPS.

Memory subsystems also differ. Both cards have 2 GB of GDDR5, but the GTX 680 uses a 256-bit bus with 192.3 GB/s of bandwidth, while the HD 8950M is constrained to a 128-bit bus and only 88.00 GB/s. That 104.3 GB/s bandwidth gap is critical for high-resolution textures and compute workloads. Clock speeds tell a similar story: the GTX 680 runs at a 1006 MHz base and 1058 MHz boost, with memory at 1502 MHz (6 Gbps effective). The HD 8950M does not list a base or boost clock, but its memory runs at 1375 MHz (5.5 Gbps effective), which is slower on both fronts. The GTX 680 also supports PCIe 3.0 x16 with dual DVI, HDMI 1.4a, and DisplayPort 1.2 outputs, whereas the HD 8950M’s display outputs are listed as “Portable Device Dependent,” reflecting its mobile nature.

The Verdict

From the data alone, the choice is clear: the NVIDIA GeForce GTX 680 is the superior GPU in every measured benchmark. It wins both head-to-head tests, has a higher average score (14150 vs. 13376), and sits at a higher percentile rank (55th vs. 54th). For any user prioritizing performance in OpenCL or Vulkan workloads, the GTX 680 is the only defensible option. Its 36% Vulkan lead is particularly decisive, indicating that modern games or applications using that API will see dramatically better frame rates on the NVIDIA card.

The AMD Radeon HD 8950M is not without purpose, but that purpose is not performance. It is a mobile part with a 100 W TDP and an MXM form factor, making it suitable for laptops or small systems where the GTX 680’s 195 W TDP and dual-slot cooler would be impractical. If you need a discrete GPU in a portable chassis, the HD 8950M is the workable choice. But if you have the space and power budget for a desktop card, the GTX 680 outperforms it in every shared test, with no exceptions. The HD 8950M’s nearest rivals — the Radeon RX 5500M and FirePro M6100 — are within 0.2% of its score, showing it is competitive with its own class, but that class is far below the GTX 680’s level.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 680, with an average score of 14150 compared to the AMD Radeon HD 8950M’s 13376, a difference of about 5.8%.

Q: How much faster is the GTX 680 in Vulkan performance?

A: The GTX 680 scores 17646 in Geekbench Vulkan, while the HD 8950M scores 12979, giving NVIDIA a 36% lead in that test.

Q: What is the memory bandwidth of each card?

A: The GTX 680 has 192.3 GB/s of bandwidth over a 256-bit bus, while the HD 8950M has 88.00 GB/s over a 128-bit bus.

Q: Which GPU has more shading units?

A: The GTX 680 has 1536 shading units, while the HD 8950M has 768, exactly half as many.

Q: Are both GPUs built on the same manufacturing process?

A: Yes, both use a 28 nm process at TSMC, but the GTX 680 has a larger die (294 mm²) and more transistors (3,540 million) than the HD 8950M (160 mm², 2,080 million).

Q: Which card is more suitable for a laptop?

A: The AMD Radeon HD 8950M, due to its 100 W TDP and MXM module form factor, compared to the GTX 680’s 195 W TDP and dual-slot desktop design.

Specification Differences

| Specification | NVIDIA GeForce GTX 680 | AMD Radeon HD 8950M |

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

| Architecture | Kepler | GCN 2.0 |

| Chip | GK104 | Saturn |

| Process Node | 28 nm | 28 nm |

| Transistors | 3,540 million | 2,080 million |

| Die Size | 294 mm² | 160 mm² |

| Transistor Density | 12.0M / mm² | 13.0M / mm² |

| Base Clock | 1006 MHz | Not listed |

| Boost Clock | 1058 MHz | Not listed |

| Memory Clock | 1502 MHz (6 Gbps effective) | 1375 MHz (5.5 Gbps effective) |

| Memory Bus Width | 256 bit | 128 bit |

| Memory Bandwidth | 192.3 GB/s | 88.00 GB/s |

| Shading Units | 1536 | 768 |

| TMUs | 128 | 48 |

| ROPs | 32 | 16 |

| Pixel Rate | 33.86 GPixel/s | 17.20 GPixel/s |

| Texture Rate | 135.4 GTexel/s | 51.60 GTexel/s |

| FP32 Performance | 3.250 TFLOPS | 1.651 TFLOPS |

| TDP | 195 W | 100 W |

| Slot Width | Dual-slot | MXM Module |

| Power Connectors | 2x 6-pin | Not listed |

| Suggested PSU | 450 W | Not listed |

| Display Outputs | 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 | Portable Device Dependent |

| DirectX Support | 12 (11_0) | 12 (12_0) |

| Vulkan Version | 1.2.175 | 1.2.170 |

| OpenGL Version | 4.6 | 4.6 |

| Release Date | 2012-03-21 | 2013-05-13 |

| Launch MSRP | 499 USD | Not listed |

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