GPU Comparison

# NVIDIA GeForce GTX 1050 vs NVIDIA GeForce 920M

The NVIDIA GeForce GTX 1050 and NVIDIA GeForce 920M occupy different tiers of the mobile and desktop GPU landscape, separated by architecture, process node, and raw compute capability. The GTX 1050, built on the Pascal architecture with a 14 nm Samsung process, delivers substantially higher performance across every shared benchmark, while the 920M, a Kepler 2.0 part on a 28 nm TSMC node, targets low-power integrated-class workloads. The data shows a clear hierarchy: the GTX 1050 achieves an average benchmark score of 3629 against the 920M's 3287, a difference that widens dramatically in compute-oriented tests. The GTX 1050 wins both head-to-head comparisons, with the Geekbench OpenCL delta reaching 308.9% and the Vulkan delta at 215.7%. For any user choosing between these two, the GTX 1050 is the only reasonable pick for gaming or GPU-accelerated tasks, while the 920M's sole advantage lies in its lower power draw and compact IGP form factor for basic display output.

The Verdict

The data points to a decisive recommendation for the NVIDIA GeForce GTX 1050 in nearly all scenarios. Its average benchmark score of 3629 places it in the 21st percentile of all GPUs, while the 920M's 3287 average lands in the 20th percentile, a narrow overall gap that masks the enormous per-test differences. In Geekbench OpenCL, the GTX 1050 scores 15233 versus the 920M's 3725, a 308.9% advantage that reflects the Pascal card's 640 shading units, 40 TMUs, and 32 ROPs operating at a 1354 MHz base clock. The Vulkan test tells a similar story: 8995 against 2849, a 215.7% margin. These are not marginal wins; they represent generational leaps in compute throughput.

For users who need a discrete GPU for gaming, content creation, or any CUDA-accelerated workload, the GTX 1050 is the clear choice. Its 1.862 TFLOPS FP32 throughput and 112.1 GB/s memory bandwidth (from GDDR5 on a 128-bit bus) dwarf the 920M's 732.7 GFLOPS and 14.40 GB/s (DDR3 on a 64-bit bus). The GTX 1050's 75 W TDP is manageable for a dual-slot card with no power connectors, and its PCIe 3.0 x16 interface provides ample bandwidth. The 920M, with a 33 W TDP and IGP slot width, suits ultra-portable laptops where battery life and thermals matter more than performance, but even there, its benchmark scores suggest it struggles to deliver playable frame rates in modern titles. The 920M's nearest rivals include the Intel HD Graphics 530 (average score 3332, deltaPct -1.4%), indicating it barely edges out integrated graphics. The GTX 1050, by contrast, sits near the NVIDIA GeForce GT 735M (3616, deltaPct 0.4%) and AMD Radeon HD 6770 (3649, deltaPct -0.5%), a much higher performance tier.

FAQ

Q: Which GPU has higher raw compute performance?

A: The GTX 1050 dominates with 1.862 TFLOPS FP32 versus the 920M's 732.7 GFLOPS. This 2.5x gap in theoretical throughput is confirmed by real-world tests: Geekbench OpenCL scores of 15233 and 3725, respectively, a 308.9% difference.

Q: Do these GPUs support the same API levels?

A: No. The GTX 1050 supports DirectX 12 (12_1) and Vulkan 1.4, while the 920M is limited to DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6. The higher feature level on the GTX 1050 enables more advanced rendering techniques.

Q: How do their memory subsystems compare?

A: The GTX 1050 uses 2 GB of GDDR5 on a 128-bit bus for 112.1 GB/s bandwidth. The 920M also has 2 GB, but it is DDR3 on a 64-bit bus, yielding only 14.40 GB/s, a 7.8x bandwidth deficit that severely limits texture fetch and frame buffer performance.

Q: Which GPU is more power-efficient?

A: The 920M draws 33 W TDP versus the GTX 1050's 75 W. However, the GTX 1050 delivers roughly 4.5x the FP32 throughput per watt (24.8 GFLOPS/W vs 22.2 GFLOPS/W), making it the more efficient performer despite higher absolute power consumption.

Q: What are the nearest performance rivals for each card?

A: The GTX 1050's closest competitors are the NVIDIA GeForce GT 735M (avg score 3616, deltaPct 0.4%) and AMD Radeon HD 6770 (3649, deltaPct -0.5%). The 920M's nearest rivals include the NVIDIA GeForce GT 730M (3316, deltaPct -0.9%) and Intel HD Graphics 530 (3332, deltaPct -1.4%).

Q: Which card is better for Vulkan-based games?

A: The GTX 1050 wins decisively with a Geekbench Vulkan score of 8995 versus 2849, a 215.7% advantage. This reflects the Pascal architecture's superior driver optimization and hardware support for Vulkan 1.4 compared to the older Kepler's 1.2.175.

Architecture Differences

The GTX 1050 is built on NVIDIA's Pascal architecture using the GP107 chip, fabricated on Samsung's 14 nm process. This node packs 3,300 million transistors into a 132 mm² die, achieving a transistor density of 25.0M / mm². The 920M uses the older Kepler 2.0 architecture with the GK208B chip, manufactured on TSMC's 28 nm process. It contains 1,020 million transistors on an 87 mm² die, with a density of just 11.7M / mm², less than half the GTX 1050's transistor density.

These architectural differences manifest in compute capabilities. The GTX 1050 features 640 shading units, 40 TMUs, and 32 ROPs, yielding a pixel rate of 46.56 GPixel/s and texture rate of 58.20 GTexel/s. The 920M has 384 shading units, 32 TMUs, and only 8 ROPs, producing 7.632 GPixel/s and 30.53 GTexel/s. The ROP count difference is particularly stark, the GTX 1050 has 4x the pixel throughput, which directly impacts fill-rate-bound scenarios like high-resolution rendering and anti-aliasing.

Clock behavior also differs. The GTX 1050 runs at 1354 MHz base and 1455 MHz boost, while the 920M is locked at 954 MHz for both base and boost. Memory clocks reflect the technology gap: the GTX 1050's GDDR5 runs at 1752 MHz (7 Gbps effective), while the 920M's DDR3 operates at 900 MHz (1800 Mbps effective). The GTX 1050 supports FP16 at 29.10 GFLOPS (1:64 ratio), while the 920M has no FP16 capability listed. Both cards lack ray tracing and tensor cores, as they predate the RTX series.

Specification Differences

| Specification | GTX 1050 | GeForce 920M |

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

| Architecture | Pascal | Kepler 2.0 |

| Process Node | 14 nm (Samsung) | 28 nm (TSMC) |

| Transistors | 3,300 million | 1,020 million |

| Die Size | 132 mm² | 87 mm² |

| Base Clock | 1354 MHz | 954 MHz |

| Boost Clock | 1455 MHz | 954 MHz |

| Memory Clock | 1752 MHz (7 Gbps) | 900 MHz (1800 Mbps) |

| Memory Type | GDDR5 | DDR3 |

| Memory Bus | 128 bit | 64 bit |

| Memory Bandwidth | 112.1 GB/s | 14.40 GB/s |

| Shading Units | 640 | 384 |

| TMUs | 40 | 32 |

| ROPs | 32 | 8 |

| Pixel Rate | 46.56 GPixel/s | 7.632 GPixel/s |

| Texture Rate | 58.20 GTexel/s | 30.53 GTexel/s |

| FP32 | 1.862 TFLOPS | 732.7 GFLOPS |

| FP16 | 29.10 GFLOPS | None |

| TDP | 75 W | 33 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | None | None |

| Bus Interface | PCIe 3.0 x16 | PCIe 3.0 x8 |

| DirectX | 12 (12_1) | 12 (11_0) |

| Vulkan | 1.4 | 1.2.175 |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a | Portable Device Dependent |

| Launch MSRP | 109 USD | None |

The GTX 1050 is a dual-slot desktop card measuring 145 mm (5.7 inches) in length and 111 mm (4.4 inches) in height, requiring a 250 W suggested PSU. The 920M is an IGP (integrated graphics processor) with no dimensions listed, designed for portable devices with no external power connectors. The GTX 1050 offers full display outputs including DVI, HDMI 2.0, and DisplayPort 1.4a, while the 920M's outputs are entirely dependent on the host device.

Head-to-Head Benchmarks

The head-to-head comparison includes two tests, and the GTX 1050 wins both. In Geekbench OpenCL, the GTX 1050 scores 15233 against the 920M's 3725, yielding a delta of 308.9%. This is the larger of the two margins, reflecting the massive difference in compute units and memory bandwidth. OpenCL workloads often stress raw shader throughput and memory access patterns, where the GTX 1050's 640 shading units and 112.1 GB/s bandwidth overwhelm the 920M's 384 units and 14.40 GB/s.

The Geekbench Vulkan test shows a similar but slightly narrower gap: 8995 for the GTX 1050 versus 2849 for the 920M, a 215.7% delta. Vulkan's lower-level API reduces driver overhead, but the underlying hardware differences remain decisive. The GTX 1050's Vulkan 1.4 support versus the 920M's 1.2.175 also contributes to better feature utilization and performance in modern graphics workloads.

Beyond the head-to-head tests, the GTX 1050's PassMark scores provide context for its overall capability: 5028 in G3D, 2091 in GPU compute, 457 in G2D, and 83 in DirectX 9. The 920M lacks PassMark results in the data, so direct comparisons are limited to the two Geekbench tests. However, the GTX 1050's average benchmark score of 3629 versus the 920M's 3287 (a 10.4% difference) underscores that the GTX 1050 is consistently faster across its broader test suite, while the 920M's sparse benchmark coverage suggests it is rarely tested for serious performance workloads.

The GTX 1050's nearest rival, the NVIDIA GeForce GT 735M, scores 3616 (deltaPct 0.4%), nearly identical to the GTX 1050's average. This places the GTX 1050 in a performance tier where it competes with older mid-range desktop parts. The 920M's nearest rival, the Intel HD Graphics 530, scores 3332 (deltaPct -1.4%), meaning the 920M slightly outperforms a mainstream integrated GPU, but only marginally, and it falls behind the NVIDIA GeForce GT 640 (3210, deltaPct 2.4%). These rankings confirm that the GTX 1050 belongs to a decidedly higher class of hardware, while the 920M hovers at the boundary between low-end discrete and integrated graphics.