Intel Arc A310 vs NVIDIA GeForce GTX 950M Comparison

# Head-to-Head Benchmarks

The benchmark data presents a stark generational divide between these two mobile graphics processors. In the two shared tests—Geekbench OpenCL and Geekbench Vulkan—the Intel Arc A310 dominates the NVIDIA GeForce GTX 950M with decisive margins that leave little room for ambiguity.

Starting with Geekbench OpenCL, the Intel Arc A310 scores 30,607 points against the GTX 950M's 9,745. That translates to a 68.2% advantage for the Intel part. The magnitude of this gap is striking: the A310 delivers more than three times the raw compute throughput in this workload. Even more telling is the Vulkan result, where the A310 posts 28,964 points versus the GTX 950M's 6,525—a 77.5% lead. This suggests the architectural gulf is even wider in modern low-level graphics APIs, where the older Maxwell design struggles to keep pace.

The head-to-head tally reflects this sweep: the Intel Arc A310 wins both available benchmark comparisons, leaving the NVIDIA GeForce GTX 950M with zero victories. The average benchmark scores reinforce the trend, though with a narrower gap—the GTX 950M averages 8,135 points across its tests, while the A310 averages 7,550. This apparent contradiction deserves scrutiny: the A310's average is dragged down by its Passmark DirectX 9 score of 69, a legacy API where the newer architecture clearly underperforms. The GTX 950M's percentile ranking of 42 versus the A310's 40 also reflects this mixed picture—the Intel part wins big in modern workloads but loses ground in older ones.

# Where Each One Wins

The data paints a clear picture of two different usage scenarios. The Intel Arc A310 is the unequivocal winner in compute-heavy and modern API workloads. Its Geekbench OpenCL score of 30,607 indicates strong general-purpose compute capability, while the Vulkan score of 28,964 suggests excellent performance in contemporary game engines and applications built around low-level graphics APIs. The A310's Passmark G3D score of 5,433 further supports its strength in DirectX 11 and 12 scenarios, with DirectX 12 scoring 29 and DirectX 11 scoring 33 in the individual API tests.

The NVIDIA GeForce GTX 950M's wins, while absent from the head-to-head table, can be inferred from its benchmark profile. Its average score of 8,135 actually exceeds the A310's 7,550, indicating that in the aggregate, the older card holds its own. The GTX 950M's closest rivals—the AMD Radeon R9 M360 at 8,129 (0.1% behind) and the NVIDIA GeForce 945M at 8,099 (0.4% behind)—suggest it sits in a stable performance band. Its GeForce 900M generation was designed for DirectX 12 (11_0) support, which means it handles legacy titles and older API workloads without the penalties the A310 exhibits in DirectX 9 (scoring just 69).

For users running modern, compute-intensive applications or Vulkan-based games, the A310 is the clear choice. For those dealing with older game libraries or legacy software, the GTX 950M's more consistent performance across its supported APIs—even if lower in peak throughput—could prove more reliable. The A310's Passmark G2D score of 625 versus its DirectX 9 score of 69 indicates a particular weakness in 2D and older 3D workloads that the GTX 950M, with its Maxwell architecture's mature driver support, likely handles more gracefully.

# Architecture Differences

The architectural chasm between these two GPUs is enormous, reflecting seven years of semiconductor evolution. The NVIDIA GeForce GTX 950M uses the GM107 chip built on TSMC's 28 nm process, packing 1,870 million transistors into a 148 mm² die. This yields a transistor density of 12.6 million per square millimeter—a figure that was competitive for 2015 but now looks positively ancient. The Intel Arc A310, by contrast, employs the DG2-128 chip on TSMC's 6 nm node, cramming 7,200 million transistors into nearly the same physical footprint at 157 mm². The density difference is stark: 45.9 million transistors per square millimeter versus the GTX 950M's 12.6 million—a 3.6x improvement in packing efficiency.

The compute configurations differ in ways that matter. The GTX 950M fields 640 shading units, 40 texture mapping units, and 16 raster output pipelines. The A310 counters with 768 shading units, 32 TMUs, and 16 ROPs. While the A310 has more shaders and the same ROP count, it actually has fewer TMUs—yet its texture rate of 56.00 GTexel/s exceeds the GTX 950M's 44.96 GTexel/s thanks to much higher clock speeds. The A310 also introduces 6 ray tracing cores, a feature entirely absent from the Maxwell-based GTX 950M, enabling hardware-accelerated ray tracing that the older card simply cannot perform.

Memory architecture shows a similar generational leap. Both cards have 4 GB of VRAM, but the GTX 950M uses DDR3 on a 128-bit bus delivering 28.80 GB/s of bandwidth. The A310 uses GDDR6 on a 64-bit bus, yet achieves 124.0 GB/s—over 4.3 times the bandwidth despite half the bus width. This confirms the massive clock speed and signaling improvements in memory technology. The A310's base clock of 1750 MHz matches its boost clock, whereas the GTX 950M runs at 993 MHz base and 1124 MHz boost—the Intel part enjoys a 56% clock advantage.

Power efficiency tells an even more dramatic story. The GTX 950M carries a 75 W TDP, while the A310 sips just 30 W. That means the A310 delivers roughly 2.7 times the FP32 throughput (2.688 TFLOPS versus 1,438.7 GFLOPS) while consuming less than half the power. The A310 also supports FP16 compute at 5.376 TFLOPS (2:1 ratio), a capability the GTX 950M lacks entirely. The A310's PCIe 4.0 x8 interface doubles the bandwidth of the GTX 950M's PCIe 3.0 x8 connection, and its four mini-DisplayPort 2.0 outputs far exceed the portable-device-dependent outputs of the NVIDIA part.

# FAQ

Q: Which GPU has higher raw compute performance?

A: The Intel Arc A310 delivers 2.688 TFLOPS of FP32 compute versus the GTX 950M's 1,438.7 GFLOPS—a 87% advantage. The A310 also supports FP16 at 5.376 TFLOPS, which the GTX 950M cannot do at all.

Q: How do the memory bandwidth figures compare?

A: The A310 achieves 124.0 GB/s over a 64-bit GDDR6 bus, while the GTX 950M manages 28.80 GB/s over a 128-bit DDR3 bus. The A310 has 4.3 times the bandwidth despite using half the bus width.

Q: Is the GTX 950M better in any benchmark category?

A: The head-to-head data shows the A310 winning both shared tests. However, the GTX 950M's average benchmark score of 8,135 exceeds the A310's 7,550, and its percentile ranking of 42 versus 40 indicates more consistent performance across its supported workload range.

Q: What API features differentiate the two?

A: The A310 supports DirectX 12 Ultimate (12_2) and includes 6 ray tracing cores, while the GTX 950M only reaches DirectX 12 (11_0) with no ray tracing hardware. Both support OpenGL 4.6 and Vulkan 1.4.

Q: How does power consumption compare?

A: The A310 has a 30 W TDP with no power connectors and a suggested 200 W PSU, while the GTX 950M draws 75 W with no power connectors. The A310 delivers higher performance at 40% of the power draw.

Q: What are the transistor density differences?

A: The A310 packs 7,200 million transistors into 157 mm² at 45.9M transistors per mm² on 6 nm, versus the GTX 950M's 1,870 million transistors in 148 mm² at 12.6M per mm² on 28 nm—a 3.6x density advantage for Intel.

# The Verdict

The data supports a clear recommendation for most users: the Intel Arc A310 is the superior GPU in nearly every measurable dimension. Its 68.2% OpenCL lead and 77.5% Vulkan lead over the GTX 950M are decisive, and its architectural advantages—ray tracing cores, FP16 support, 4.3x memory bandwidth, and 87% higher FP32 throughput—make it the obvious choice for modern workloads. The A310 achieves all this at 30 W TDP versus 75 W, a remarkable efficiency gain.

However, the GTX 950M is not without merit. Its average benchmark score of 8,135 exceeds the A310's 7,550, and its 42nd percentile ranking versus the A310's 40th suggests better consistency. For users running legacy DirectX 9 applications, the A310's catastrophic score of 69 in that API test versus the GTX 950M's more balanced profile makes the older card a safer bet for vintage game libraries. The GTX 950M's nearest rival list includes the GTX 980 at only 0.4% higher average score, indicating it sits in well-established performance territory.

The verdict hinges on workload. For modern gaming, compute tasks, or any application leveraging Vulkan, DirectX 12 Ultimate, or ray tracing, the Intel Arc A310 is the only rational choice. For legacy compatibility and consistent performance across older APIs, the GTX 950M remains functional but its end-of-life status and 2015 release date mean it offers no upgrade path. The A310, despite also being end-of-life, represents a far more future-proof architecture with its successor Battlemage already announced.

# Specification Differences

| Specification | NVIDIA GeForce GTX 950M | Intel Arc A310 |

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

| Architecture | Maxwell | Xe-HPG |

| Process Node | 28 nm | 6 nm |

| Transistors | 1,870 million | 7,200 million |

| Die Size | 148 mm² | 157 mm² |

| Transistor Density | 12.6M / mm² | 45.9M / mm² |

| Base Clock | 993 MHz | 1750 MHz |

| Boost Clock | 1124 MHz | 1750 MHz |

| Memory Type | DDR3 | GDDR6 |

| Memory Bus Width | 128 bit | 64 bit |

| Memory Bandwidth | 28.80 GB/s | 124.0 GB/s |

| Shading Units | 640 | 768 |

| TMUs | 40 | 32 |

| ROPs | 16 | 16 |

| RT Cores | None | 6 |

| Pixel Rate | 17.98 GPixel/s | 28.00 GPixel/s |

| Texture Rate | 44.96 GTexel/s | 56.00 GTexel/s |

| FP32 Performance | 1,438.7 GFLOPS | 2.688 TFLOPS |

| FP16 Performance | None | 5.376 TFLOPS (2:1) |

| TDP | 75 W | 30 W |

| Slot Width | IGP | Single-slot |

| Bus Interface | PCIe 3.0 x8 | PCIe 4.0 x8 |

| Display Outputs | Portable Device Dependent | 4x mini-DisplayPort 2.0 |

| DirectX Support | 12 (11_0) | 12 Ultimate (12_2) |

| Release Date | 2015-03-12 | 2022-10-11 |