NVIDIA GeForce GTX 1070 vs NVIDIA GeForce GTX 950M Comparison

Opening paragraph: The NVIDIA GeForce GTX 1070 and the NVIDIA GeForce GTX 950M occupy very different corners of the GPU landscape: one is a desktop-class part built for high-refresh and high-resolution gaming, the other is a mobile chip aimed at mainstream laptops. The database’s recorded measurements show a massive performance gulf, but the gap is not uniform across every workload. This analysis walks through the head-to-head results, answers common questions from the data, and identifies where each card makes sense based strictly on benchmark scores and specifications.

Head-to-Head Benchmarks

The database lists only two shared benchmark results for these two GPUs, and the GTX 1070 wins both decisively. In Geekbench OpenCL, the GTX 1070 scores 44,700 points against the GTX 950M’s 9,745 points. That is a delta of 358.7%, meaning the GTX 1070 delivers roughly four and a half times the compute throughput in this API. The GTX 950M’s score is closer to entries like the NVIDIA GeForce 945M (8,099 points) or the NVIDIA GRID K2 (8,080 points), which confirms its position as an entry-level mobile part. The GTX 1070, by contrast, sits in a completely different league: its nearest rivals in the database are the AMD FirePro W5000 (9,803 points, delta of -0.2%), the NVIDIA Quadro M2000M (9,832 points, delta of -0.5%), the NVIDIA Tesla M10 (9,724 points, delta of 0.6%), and the NVIDIA Tesla C2070 (9,716 points, delta of 0.7%). These deltas are all within 1% of the GTX 1070’s average score, which shows that the GTX 1070’s overall average is tightly clustered with professional workstation cards, despite the raw OpenCL blowout over the GTX 950M.

In Geekbench Vulkan, the margin narrows but remains enormous. The GTX 1070 scores 22,121 points, while the GTX 950M scores 6,525 points. That is a delta of 239%, meaning the GTX 1070 is about 3.4 times faster in this test. Vulkan is a low-level API that tends to favor newer architectures, and the data reflects that: the GTX 1070’s Vulkan score is over half of its OpenCL score, while the GTX 950M’s Vulkan score is only about two-thirds of its OpenCL score. The GTX 950M’s 6,525 points in Vulkan are far below the GTX 1070’s 22,121 points, and this gap translates directly into real-world frame rate differences in Vulkan-based titles or compute workloads.

The overall win count is 2 for the GTX 1070 and 0 for the GTX 950M, but the average benchmark scores tell the broader story. The GTX 1070 has an average benchmark score of 9,780 points across all its recorded tests (including 3DMark Steel Nomad DX12 at 1,082 points, Geekbench Metal at 18,801 points, Passmark G3D at 13,498 points, and Passmark GPU Compute at 6,102 points, among others). The GTX 950M’s average benchmark score is 8,135 points, but that figure is based on only two tests (OpenCL and Vulkan). If the GTX 950M had been subjected to the same suite, its average would likely be far lower, because its OpenCL and Vulkan scores are already roughly 78% and 70% lower than the GTX 1070’s respective scores. The percentile rankings confirm the gap: the GTX 1070 sits at the 47th percentile of all GPUs in the database, while the GTX 950M sits at the 42nd percentile. That 5-point difference might sound small, but it is compressed because the database includes many weak integrated and low-end mobile parts that drag the percentile curve down.

FAQ

Q: Which GPU is faster in raw compute performance?

A: The GTX 1070 is dramatically faster. In Geekbench OpenCL, it scores 44,700 points versus the GTX 950M’s 9,745 points, a delta of 358.7%. In Geekbench Vulkan, it scores 22,121 points versus 6,525 points, a delta of 239%.

Q: Does the GTX 950M win any benchmark in the shared test set?

A: No. The database records 2 wins for the GTX 1070 and 0 wins for the GTX 950M across the two head-to-head tests (OpenCL and Vulkan). The GTX 950M has no recorded benchmark where it exceeds the GTX 1070.

Q: How does the GTX 1070 compare to its nearest rivals in average score?

A: The GTX 1070’s average benchmark score is 9,780 points. Its nearest rivals are the AMD FirePro W5000 (9,803 points, delta -0.2%), NVIDIA Quadro M2000M (9,832 points, delta -0.5%), NVIDIA Tesla M10 (9,724 points, delta 0.6%), and NVIDIA Tesla C2070 (9,716 points, delta 0.7%). All four are within a fraction of a percent of the GTX 1070.

Q: What is the GTX 950M’s closest competition in the database?

A: The GTX 950M’s average score is 8,135 points. Its nearest rivals are the AMD Radeon R9 M360 (8,129 points, delta 0.1%), NVIDIA GeForce GTX 980 (8,167 points, delta -0.4%), NVIDIA GeForce 945M (8,099 points, delta 0.4%), and NVIDIA GRID K2 (8,080 points, delta 0.7%). All are within 1% of each other.

Q: Are these two GPUs from the same architecture generation?

A: No. The GTX 1070 uses the Pascal architecture on a 16 nm process, while the GTX 950M uses the Maxwell architecture on a 28 nm process. This is a two-generation gap, and the GTX 1070’s successor is the GeForce 20 series, while the GTX 950M’s successor is the GeForce 10 Mobile series.

Q: What is the memory configuration difference?

A: The GTX 1070 has 8 GB of GDDR5 memory on a 256-bit bus, delivering 256.3 GB/s of bandwidth. The GTX 950M has 4 GB of DDR3 memory on a 128-bit bus, delivering 28.80 GB/s of bandwidth. That is a 9x difference in memory bandwidth.

Where Each One Wins

The GTX 1070 wins every shared benchmark category in the database: Geekbench OpenCL and Geekbench Vulkan. Its compute performance is so far ahead that any workload leveraging OpenCL (e.g., physics simulation, video encoding, or general-purpose GPU compute) will be several times faster on the GTX 1070. The Vulkan results also indicate that modern games using Vulkan will see a huge advantage on the GTX 1070. For the GTX 950M, there is no benchmark category in the shared set where it comes out ahead. Its only realistic strengths are implicit in its specifications: it is a low-power mobile part (75 W TDP, no power connectors, PCIe 3.0 x8 interface) that fits in laptops without requiring external power, and its portable-device-dependent display outputs make it suitable for integrated mobile systems. But in terms of recorded performance, the GTX 950M wins nothing.

The GTX 1070’s additional benchmark results (not shared with the GTX 950M) show where it excels: 3DMark Steel Nomad DX12 at 1,082 points, Geekbench Metal at 18,801 points, Passmark DirectX 10 at 82 points, DirectX 11 at 100 points, DirectX 12 at 48 points, DirectX 9 at 197 points, Passmark G2D at 846 points, Passmark G3D at 13,498 points, and Passmark GPU Compute at 6,102 points. These scores indicate a balanced part that handles legacy DirectX 9 titles, modern DirectX 12 workloads, and compute tasks without a single weak spot. The GTX 950M’s only recorded tests are OpenCL (9,745 points) and Vulkan (6,525 points), so there is no data to suggest it wins in any other category.

Specification Differences

The two GPUs differ in nearly every measurable specification. The GTX 1070 uses the GP104 chip on a 16 nm process with 7,200 million transistors on a 314 mm² die, giving a transistor density of 22.9 million per mm². The GTX 950M uses the GM107 chip on a 28 nm process with 1,870 million transistors on a 148 mm² die, giving a transistor density of 12.6 million per mm². The GTX 1070 has a base clock of 1506 MHz and a boost clock of 1683 MHz, while the GTX 950M has a base clock of 993 MHz and a boost clock of 1124 MHz. Memory clocks differ as well: the GTX 1070 runs at 2002 MHz (8 Gbps effective), while the GTX 950M runs at 900 MHz (1800 Mbps effective).

The memory subsystem is the most striking difference. The GTX 1070 has 8 GB of GDDR5 on a 256-bit bus with 256.3 GB/s bandwidth. The GTX 950M has 4 GB of DDR3 on a 128-bit bus with 28.80 GB/s bandwidth. That is a 8.9x bandwidth advantage for the GTX 1070. The compute units also differ: the GTX 1070 has 1,920 shading units, 120 texture mapping units (TMUs), and 64 render output units (ROPs), while the GTX 950M has 640 shading units, 40 TMUs, and 16 ROPs. Pixel rate is 107.7 GPixel/s for the GTX 1070 versus 17.98 GPixel/s for the GTX 950M, and texture rate is 202.0 GTexel/s versus 44.96 GTexel/s. FP32 throughput is 6.463 TFLOPS for the GTX 1070 versus 1,438.7 GFLOPS (1.44 TFLOPS) for the GTX 950M. The GTX 1070 also has an FP16 rate of 101.0 GFLOPS (1:64 ratio), while the GTX 950M has no recorded FP16 value. Power draw is 150 W for the GTX 1070 versus 75 W for the GTX 950M, and the GTX 1070 requires a dual-slot cooler with a single 8-pin power connector and a 450 W suggested PSU, while the GTX 950M is an integrated graphics package (IGP) with no power connectors and no suggested PSU. The GTX 1070 uses a PCIe 3.0 x16 interface and has display outputs of 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a, while the GTX 950M uses PCIe 3.0 x8 and has portable-device-dependent outputs.

Architecture Differences

The GTX 1070 is built on the Pascal architecture (GeForce 10-series, generation GeForce 10), while the GTX 950M is built on the Maxwell architecture (GeForce 900M series). Pascal is a newer microarchitecture with significant improvements in clock scaling, memory compression, and asynchronous compute. The process node difference is critical: 16 nm for Pascal versus 28 nm for Maxwell, which explains why the GTX 1070 packs 7,200 million transistors into a 314 mm² die (22.9 million per mm²), while the GTX 950M fits 1,870 million transistors into a 148 mm² die (12.6 million per mm²). The GTX 1070’s transistor density is 82% higher, which directly enables its higher clock speeds and larger shader count.

Both GPUs support DirectX 12, but there is a feature-level difference. The GTX 1070 supports DirectX 12 (12_1), while the GTX 950M supports DirectX 12 (11_0). This means the GTX 1070 can use DirectX 12_1 features like conservative rasterization and rasterizer ordered views, which the GTX 950M cannot. Both support OpenGL 4.6 and Vulkan 1.4, so API compatibility is otherwise identical. The GTX 1070 has no ray tracing cores or tensor cores, and neither does the GTX 950M, so both rely on traditional shader-based rendering. The GTX 1070’s predecessor is the GeForce 900 series, and its successor is the GeForce 20 series, while the GTX 950M’s predecessor is the GeForce 800M series, and its successor is the GeForce 10 Mobile series. This places the GTX 1070 one generation newer than the GTX 950M in NVIDIA’s product stack.

The Verdict

The data is unambiguous: the GTX 1070 outperforms the GTX 950M in every shared benchmark by a factor of 3.4 to 4.5. If you are building a desktop for 1080p or 1440p gaming, or if you need a GPU that can handle compute workloads in OpenCL or Vulkan, the GTX 1070 is the only choice from this pair. Its 8 GB of GDDR5 memory with 256.3 GB/s bandwidth is sufficient for modern textures, and its 6.463 TFLOPS of FP32 throughput is roughly 4.5 times the GTX 950M’s 1,438.7 GFLOPS. The GTX 1070’s average benchmark score of 9,780 points places it alongside workstation cards like the Quadro M2000M and Tesla M10, while the GTX 950M’s 8,135 points places it with low-end mobile parts like the Radeon R9 M360 and GeForce 945M.

The GTX 950M is not a bad GPU for its intended use case: it is a 75 W mobile chip with no power connectors, designed to fit in laptops where power and cooling are limited. Its 4 GB of DDR3 memory and 28.80 GB/s bandwidth are enough for older games at modest settings, and its Maxwell architecture still supports Vulkan 1.4 and OpenGL 4.6. But the benchmark data shows no scenario where it beats the GTX 1070, so anyone choosing between these two should pick the GTX 1070 unless they absolutely need the GTX 950M’s integrated form factor and low power draw. For a desktop builder, the GTX 1070’s 150 W TDP and 450 W suggested PSU are reasonable, and its dual-slot cooler with a single 8-pin connector is standard. For a laptop buyer, the GTX 950M is a legacy part that is end-of-life, as is the GTX 1070, but the GTX 1070 remains a capable performer by current database standards, while the GTX 950M is firmly in entry-level territory.