NVIDIA GeForce GTX 1080 Ti vs NVIDIA GeForce GTX 950 Comparison

Where Each One Wins

The benchmark data splits cleanly along generational lines. The NVIDIA GeForce GTX 1080 Ti wins every recorded head-to-head comparison, and the margins are not subtle. Across the three shared tests, the GTX 1080 Ti takes all three wins. The GTX 950 records zero wins in the direct comparison.

For compute-oriented workloads, the GTX 1080 Ti is in a different class entirely. Its Geekbench OpenCL score of 67929 dwarfs the GTX 950's 15662, a 333.7% advantage. That is not a marginal improvement, it is a generational leap that makes the older card irrelevant for serious compute tasks. The GTX 950 was never positioned as a compute powerhouse, and the data confirms it.

For gaming-oriented API workloads, the story is similar but with a narrower gap. In Geekbench Vulkan, the GTX 1080 Ti scores 40511 against 16734 for the GTX 950, a 142.1% lead. Vulkan is where the newer architecture's efficiency and raw throughput show up most clearly. The GTX 950 can still run Vulkan titles, but it will be doing so at dramatically lower performance levels.

The GTX 1080 Ti also dominates in Apple's Metal API. Its score of 30624 is 327% higher than the GTX 950's 7172. This matters for users in macOS or Metal-based workloads, where the newer card is essentially four times faster.

Where the GTX 950 does have a niche is in its overall standing. It sits at the 53rd percentile among all GPUs in the database, which is remarkably close to the GTX 1080 Ti's 58th percentile. That sounds odd given the massive benchmark gaps, but it reflects the fact that the database includes many older and weaker cards. The GTX 950 is not a bad card, it is just completely outclassed by the flagship GTX 1080 Ti.

The GTX 1080 Ti's average benchmark score of 15548 puts it in the same neighborhood as cards like the AMD Radeon R9 M380 (15521, 0.2% ahead) and the AMD Radeon Pro W5500 (15679, 0.8% behind). The GTX 950's average of 13189 places it near the NVIDIA GeForce GTX 480 (13300, 0.8% behind) and the NVIDIA Tesla M2090 (13075, 0.9% ahead). In other words, the GTX 1080 Ti competes with modern mid-range and prosumer cards, while the GTX 950 competes with a much older generation.

Architecture Differences

The two cards come from different architectural eras. The GTX 1080 Ti uses the GP102 chip built on Pascal architecture, while the GTX 950 uses the GM206 chip on Maxwell 2.0. Both are manufactured by TSMC, but on different process nodes: the GTX 1080 Ti uses 16 nm, the GTX 950 uses 28 nm. That process difference alone explains a large portion of the performance gap.

The transistor counts reflect the scale difference. The GTX 1080 Ti packs 11,800 million transistors on a 471 mm² die, giving a transistor density of 25.1 million per mm². The GTX 950 has 2,940 million transistors on a 228 mm² die, with a density of 12.9 million per mm². The newer process node allows the GTX 1080 Ti to pack nearly four times the transistors into just over twice the die area.

Core configuration differences are stark. The GTX 1080 Ti has 3584 shading units, 224 texture mapping units, and 88 raster output pipelines. The GTX 950 has 768 shading units, 48 TMUs, and 32 ROPs. That is roughly 4.7 times more shading units, 4.7 times more TMUs, and 2.75 times more ROPs on the newer card.

Memory architecture is another major divider. The GTX 1080 Ti uses 11 GB of GDDR5X on a 352-bit bus, delivering 484.4 GB/s of bandwidth. The GTX 950 uses 2 GB of GDDR5 on a 128-bit bus, with 105.8 GB/s. The newer card has more than 4.5 times the memory bandwidth and over 5 times the memory capacity. For modern games that routinely exceed 2 GB of VRAM usage, the GTX 950's 2 GB allocation is a hard ceiling.

Clock speeds tell a more nuanced story. The GTX 1080 Ti runs at a base clock of 1481 MHz and a boost of 1582 MHz. The GTX 950 runs at 1024 MHz base and 1188 MHz boost. The newer card has higher clocks, but the real advantage comes from having far more cores running at those clocks. The GTX 1080 Ti's FP32 throughput of 11.34 TFLOPS versus the GTX 950's 1.825 TFLOPS is a 6.2x difference that no clock adjustment could bridge.

Both cards support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 1080 Ti also lists FP16 capability at 177.2 GFLOPS, while the GTX 950 has no recorded FP16 figure. Power requirements differ substantially: the GTX 1080 Ti is rated at 250 W TDP with a 600 W suggested PSU, while the GTX 950 is rated at 90 W with a 250 W suggested PSU. The GTX 1080 Ti also uses a larger physical footprint at 267 mm length versus 202 mm for the GTX 950.

Head-to-Head Benchmarks

The direct comparison tests cover three APIs, and the GTX 1080 Ti wins all three by margins that range from substantial to overwhelming.

In Geekbench OpenCL, the GTX 1080 Ti scores 67929 against the GTX 950's 15662. That is a 333.7% delta, the largest margin in the head-to-head set. OpenCL is a general-purpose compute API, and the GTX 1080 Ti's massive shading unit count and memory bandwidth give it an insurmountable advantage. For any workload that scales across parallel cores, this result is representative.

Geekbench Metal shows a 327% delta. The GTX 1080 Ti scores 30624, while the GTX 950 manages 7172. Metal is Apple's graphics API, used in macOS and iOS environments. The GTX 1080 Ti's result here is slightly less dominant than in OpenCL, but still represents roughly a 4.3x performance advantage. Users running Metal-based applications on compatible hardware should expect the GTX 1080 Ti to be in a completely different performance tier.

Geekbench Vulkan narrows the gap somewhat, with a 142.1% delta. The GTX 1080 Ti scores 40511, and the GTX 950 scores 16734. Vulkan is a low-overhead API that benefits from efficient driver utilization, and the Pascal architecture's improvements in this area show. Still, a 142.1% lead is decisive. The GTX 950's Vulkan result is respectable for its class, but it cannot compete with the GTX 1080 Ti's raw throughput.

Beyond the head-to-head, the broader benchmark suite shows the GTX 1080 Ti's strengths across older APIs. It scores 18600 in Passmark G3D, 9632 in Passmark GPU Compute, and 231 in Passmark DirectX 9. The GTX 950 has no recorded scores in these tests, so direct comparison is not possible from the database. However, the GTX 1080 Ti's 3DMark Steel Nomad DX12 score of 2231 and Passmark DirectX 11 score of 151 indicate strong modern API performance.

FAQ

Q: How much faster is the GTX 1080 Ti than the GTX 950 in OpenCL?

A: The GTX 1080 Ti scores 67929 in Geekbench OpenCL, while the GTX 950 scores 15662. That is a 333.7% advantage for the GTX 1080 Ti.

Q: Is the GTX 950 competitive in Vulkan workloads?

A: The GTX 950 scores 16734 in Geekbench Vulkan, which is a respectable result for its class. However, the GTX 1080 Ti scores 40511, a 142.1% lead, so the GTX 950 is not competitive against the newer card.

Q: What are the memory specifications of each card?

A: The GTX 1080 Ti has 11 GB of GDDR5X on a 352-bit bus with 484.4 GB/s bandwidth. The GTX 950 has 2 GB of GDDR5 on a 128-bit bus with 105.8 GB/s bandwidth.

Q: Do both cards support the same modern APIs?

A: Yes, both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 1080 Ti also has FP16 capability at 177.2 GFLOPS, while the GTX 950 has no recorded FP16 figure.

Q: How do the power requirements compare?

A: The GTX 1080 Ti has a 250 W TDP and a 600 W suggested PSU. The GTX 950 has a 90 W TDP and a 250 W suggested PSU.

Q: What is the GTX 1080 Ti's standing among all GPUs?

A: It sits at the 58th percentile among all GPUs in the database, with an average benchmark score of 15548. Its closest rivals include the AMD Radeon R9 M380 (0.2% ahead) and the AMD Radeon Pro W5500 (0.8% behind).

The Verdict

The data makes the choice straightforward. The GTX 1080 Ti wins every recorded benchmark, and the margins are decisive. For anyone needing maximum compute throughput, the 333.7% lead in OpenCL and 142.1% lead in Vulkan are the numbers that matter. The GTX 1080 Ti is the clear choice for demanding gaming, compute workloads, and high-resolution texture work where its 11 GB of VRAM and 484.4 GB/s bandwidth come into play.

The GTX 950 is not without merit, but its merits are contextual. It sits at the 53rd percentile among all GPUs, close to the GTX 1080 Ti's 58th percentile, which means it is not a weak card in absolute terms. Its 90 W TDP and 250 W suggested PSU make it far easier to integrate into an existing system without upgrading the power supply. Its 202 mm length makes it suitable for compact cases. But the performance gap is too large to recommend it for anything beyond light or legacy workloads.

For users choosing between these two, the deciding factor is workload. If the priority is modern gaming at high settings, GPU compute, or any task that benefits from more than 2 GB of VRAM, the GTX 1080 Ti is the only rational choice. If the priority is minimal power draw, small physical footprint, and modest performance for older titles or basic tasks, the GTX 950 remains functional. The database does not record a single test where the GTX 950 wins, and that is the summary that matters.

Specification Differences

| Specification | NVIDIA GeForce GTX 1080 Ti | NVIDIA GeForce GTX 950 |

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

| Architecture | Pascal | Maxwell 2.0 |

| Process Node | 16 nm | 28 nm |

| Transistors | 11,800 million | 2,940 million |

| Die Size | 471 mm² | 228 mm² |

| Transistor Density | 25.1M / mm² | 12.9M / mm² |

| Base Clock | 1481 MHz | 1024 MHz |

| Boost Clock | 1582 MHz | 1188 MHz |

| Memory Size | 11 GB | 2 GB |

| Memory Type | GDDR5X | GDDR5 |

| Memory Bus Width | 352 bit | 128 bit |

| Memory Bandwidth | 484.4 GB/s | 105.8 GB/s |

| Memory Clock | 1376 MHz, 11 Gbps effective | 1653 MHz, 6.6 Gbps effective |

| Shading Units | 3584 | 768 |

| TMUs | 224 | 48 |

| ROPs | 88 | 32 |

| Pixel Rate | 139.2 GPixel/s | 38.02 GPixel/s |

| Texture Rate | 354.4 GTexel/s | 57.02 GTexel/s |

| FP32 | 11.34 TFLOPS | 1.825 TFLOPS |

| FP16 | 177.2 GFLOPS (1:64) | Not recorded |

| TDP | 250 W | 90 W |

| Power Connectors | 1x 6-pin + 1x 8-pin | 1x 6-pin |

| Suggested PSU | 600 W | 250 W |

| Length | 267 mm (10.5 inches) | 202 mm (8 inches) |

| Height | 112 mm (4.4 inches) | Not recorded |

| Width | 40 mm (1.6 inches) | Not recorded |

| Display Outputs | 1x HDMI 2.0, 3x DisplayPort 1.4a | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2 |

| Release Date | 2017-03-09 | 2015-08-19 |

| Launch MSRP | 699 USD | 159 USD |