NVIDIA GeForce 945M vs NVIDIA GeForce GTX 950M Comparison

# NVIDIA GeForce GTX 950M vs NVIDIA GeForce 945M

The two mobile GPUs share the same silicon but diverge in execution, with the GTX 950M posting an average benchmark score of 8135 versus 8099 for the 945M. That 0.4% aggregate gap might look trivial, yet the single head-to-head test tells a different story: the GTX 950M leads by 20.3% in Geekbench OpenCL, scoring 9745 against 8099. Both cards sit at the 42nd percentile among all GPUs, placing them in the same broad performance tier, but the data suggests the 950M extracts more from the shared architecture. The 945M’s nearest-rival list even includes the GTX 950M at a -0.4% delta, confirming the two are direct competitors separated by a measurable margin.

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and it decisively favors the GTX 950M. The 950M scores 9745, while the 945M manages 8099 — a 20.3% advantage for the former. That is not a marginal victory; it is a substantial gap that implies the 950M delivers noticeably better compute throughput in OpenCL workloads. The 945M’s 8099 score is nearly identical to its average of 8099, since it has only that one benchmark entry, while the 950M’s 9745 OpenCL result sits well above its 8135 average — meaning its Vulkan score of 6525 drags the average down. This asymmetry matters: in OpenCL, the 950M is clearly superior, but in Vulkan, the 945M has no recorded result, so cross-API comparisons remain incomplete.

Looking at the rivals confirms the 950M’s position. Its nearest competitor, the AMD Radeon R9 M360, scores 8129 — just 0.1% behind the 950M’s 8135 average. The NVIDIA GeForce GTX 980 sits at 8167, a mere 0.4% ahead, which is remarkable given the GTX 980’s desktop-class reputation. The 945M’s rivals tell a similar story: the NVIDIA GRID K2 scores 8080 (0.2% behind), the Radeon R9 M360 is 0.4% ahead at 8129, and the GTX 650 Ti Boost trails at 8067 (0.4% behind). In essence, both mobile GPUs orbit the same performance cluster, but the 950M consistently lands on the upper edge of that cluster.

Architecture Differences

Both GPUs are built on the same GM107 chip using the Maxwell architecture, fabricated by TSMC on a 28 nm process. The transistor count is identical at 1,870 million, and the die size matches at 148 mm², yielding the same 12.6M transistors per mm² density. This is the same fundamental silicon, so architectural differences are minimal. The shading units, texture mapping units, and render output units are all equal: 640 shading units, 40 TMUs, and 16 ROPs. The memory bus is also identical at 128 bit, and both use DDR3 memory with 900 MHz clock speed and 1800 Mbps effective data rate, producing the same 28.80 GB/s bandwidth.

The divergence begins with clocks. The GTX 950M runs a base clock of 993 MHz with a boost of 1124 MHz, while the 945M operates at 928 MHz base and 1020 MHz boost. That difference translates directly into throughput metrics: the 950M achieves a pixel rate of 17.98 GPixel/s versus 16.32 GPixel/s for the 945M, and a texture rate of 44.96 GTexel/s versus 40.80 GTexel/s. FP32 compute follows the same pattern, with the 950M delivering 1,438.7 GFLOPS against 1,305.6 GFLOPS for the 945M. The 950M’s higher clocks make it roughly 10% faster in raw pixel, texture, and compute rates — a consistent advantage across all shading workloads.

Memory capacity differs as well: the 950M comes with 4 GB of DDR3, while the 945M has 2 GB. Both have the same 128-bit bus and bandwidth, so the extra capacity on the 950M does not increase throughput but does allow larger datasets to reside on the GPU. The slot form factor also differs: the 950M is listed as an IGP (integrated graphics processor) while the 945M is an MXM Module with a bus interface of MXM-B (3.0). The 950M uses PCIe 3.0 x8, whereas the 945M connects via MXM-B. Both share the same API support: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. Power draw is identical at 75 W, and neither uses external power connectors.

Where Each One Wins

The GTX 950M wins the only head-to-head benchmark available, and it wins by a wide margin. In Geekbench OpenCL, the 950M’s 9745 score beats the 945M’s 8099 by 20.3%. This advantage stems entirely from the higher clock speeds — the 950M’s boost clock of 1124 MHz versus 1020 MHz on the 945M gives it a ~10% frequency edge, which compounds across the 640 shading units. For compute-heavy tasks like OpenCL workloads, the 950M is the clear pick. Its 4 GB memory capacity also provides headroom for applications that need to store larger buffers or textures locally, even though bandwidth remains the same.

The 945M, by contrast, has no recorded wins in the head-to-head data. Its single benchmark score of 8099 in OpenCL is its only data point, and it falls short. However, the 945M is not without merit. Its lower clocks mean it likely runs cooler and quieter in the same chassis, though the FACT PACK does not include thermal or acoustic measurements to confirm this. The 945M’s MXM form factor makes it a drop-in replacement for laptops with MXM slots, whereas the 950M’s IGP designation suggests it is soldered to the motherboard. For users with an MXM-based laptop, the 945M is the only one of the two that fits that socket. The 945M also shares the same 75 W TDP, so power consumption is equal, but the 950M delivers more performance for that same power envelope.

FAQ

Q: Which GPU is faster in OpenCL benchmarks?

A: The GTX 950M scores 9745 in Geekbench OpenCL, which is 20.3% higher than the 945M’s 8099. The 950M’s higher base and boost clocks (993/1124 MHz versus 928/1020 MHz) drive this performance advantage.

Q: Do both GPUs have the same memory bandwidth?

A: Yes. Both use DDR3 memory on a 128-bit bus at 900 MHz with 1800 Mbps effective speed, resulting in identical 28.80 GB/s bandwidth. The 950M has 4 GB capacity while the 945M has 2 GB, but throughput is the same.

Q: How do these GPUs compare to the AMD Radeon R9 M360?

A: The GTX 950M scores 8135 on average, which is 0.1% ahead of the R9 M360’s 8129. The 945M averages 8099, placing it 0.4% behind the R9 M360. All three GPUs sit in the same performance tier.

Q: What is the transistor configuration of these chips?

A: Both use the GM107 chip with 1,870 million transistors on a 148 mm² die, produced on TSMC’s 28 nm process. The transistor density is 12.6M per mm² for both.

Q: Do they support the same graphics APIs?

A: Yes, both support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. There is no difference in API feature levels between the two.

Q: Are there any benchmark results where the 945M wins?

A: No. In the single head-to-head Geekbench OpenCL test, the GTX 950M wins with a 20.3% margin. The 945M has no recorded Vulkan score, so no comparison exists for that API.

The Verdict

The data points to a straightforward conclusion: the GeForce GTX 950M is the superior mobile GPU. Its 20.3% lead in OpenCL is substantial, and its higher clocks deliver proportional gains in pixel rate (17.98 vs 16.32 GPixel/s), texture rate (44.96 vs 40.80 GTexel/s), and FP32 compute (1,438.7 vs 1,305.6 GFLOPS). The 950M also doubles the memory capacity to 4 GB, which helps in memory-constrained scenarios even if bandwidth is unchanged. The average benchmark scores reinforce this: 8135 for the 950M versus 8099 for the 945M, a 0.4% aggregate difference that understates the OpenCL gap because the 950M’s Vulkan score of 6525 pulls its average down.

That said, the 945M is not obsolete. It matches the 950M in every architectural fundamental — same chip, same 640 shading units, same 40 TMUs, same 16 ROPs, same 75 W TDP. For users with an MXM slot, the 945M is the compatible choice, as the 950M’s IGP form factor suggests it is not socketed. The 945M also runs at lower clocks, which could translate to lower heat output in thin laptops, though the FACT PACK does not provide thermal data. If the choice is between these two for a new system, the GTX 950M is the clear winner on performance. If upgrading an existing MXM laptop, the 945M is the only option that fits.

Specification Differences

The following table highlights only the fields where the two GPUs differ:

| Specification | GeForce GTX 950M | GeForce 945M |

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

| Base Clock | 993 MHz | 928 MHz |

| Boost Clock | 1124 MHz | 1020 MHz |

| Memory Size | 4 GB | 2 GB |

| Pixel Rate | 17.98 GPixel/s | 16.32 GPixel/s |

| Texture Rate | 44.96 GTexel/s | 40.80 GTexel/s |

| FP32 Compute | 1,438.7 GFLOPS | 1,305.6 GFLOPS |

| Slot Width | IGP | MXM Module |

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

| Release Date | 2015-03-12 | 2015-10-26 |

All other specifications — including chip, architecture, process node, transistor count, die size, shading units, TMUs, ROPs, memory type, bus width, bandwidth, TDP, APIs, and display outputs — are identical between the two.