GPU Comparison

FAQ

Q: Which GPU wins the only direct head-to-head benchmark?

A: The NVIDIA GeForce 920MX wins the sole head-to-head test, the Geekbench OpenCL benchmark, with a score of 4,068 against the GeForce GT 735M's 3,616. The deltaPct is -11.1%, which means the 920MX outperforms the GT 735M by 11.1%.

Q: How does the GeForce GT 735M compare to its nearest rivals in average benchmark score?

A: The GT 735M has an average benchmark score of 3,616. It is 0.3% behind the NVIDIA GeForce GTX 1050 (average score 3,629), 0.6% ahead of the NVIDIA RTX 5000 Mobile Ada Generation (3,596), 0.6% ahead of the NVIDIA GeForce GT 545 (3,594), and 0.9% behind the AMD Radeon HD 6770 (3,649).

Q: What is the average benchmark score of the GeForce 920MX, and how does it rank against the GT 735M?

A: The 920MX has an average benchmark score of 3,528, which is 2.4% lower than the GT 735M's average of 3,616. Despite losing on average score, the 920MX wins the direct OpenCL head-to-head, illustrating a discrepancy between aggregate and specific test results.

Q: Are there any differences in memory capacity, bus width, or memory type?

A: No. Both GPUs feature 2 GB of DDR3 memory on a 64-bit bus, with identical bandwidth of 14.40 GB/s and the same memory clock of 900 MHz (1,800 Mbps effective).

Q: Which GPU has a higher pixel rate and texture rate?

A: The GeForce 920MX has the higher pixel rate at 7.944 GPixel/s versus the GT 735M's 5.024 GPixel/s. It also leads in texture rate with 23.83 GTexel/s compared to 20.10 GTexel/s for the GT 735M.

Q: What are the power consumption differences between the two?

A: The GeForce 920MX has a TDP of 16 W, while the GeForce GT 735M has a TDP of 33 W. The 920MX consumes less than half the power of the GT 735M.

Architecture Differences

The two GPUs belong to different NVIDIA architectures. The GeForce GT 735M is built on Kepler 2.0, using the GK208 chip, while the GeForce 920MX uses the Maxwell architecture with the GM108S chip. Both are manufactured by TSMC on the same 28 nm process node, but the die sizes differ: the GT 735M has an 87 mm² die, while the 920MX is smaller at 77 mm². Both chips contain exactly 1,020 million transistors, which results in a higher transistor density for the 920MX at 13.2M / mm² versus 11.7M / mm² for the GT 735M.

The shading units differ notably. The GT 735M has 384 shading units, while the 920MX has fewer at 256 shading units. Texture mapping units also differ: the GT 735M has 32 TMUs, while the 920MX has 24. Both have the same number of render output units (8 ROPs). Despite having fewer shading units and TMUs, the 920MX achieves higher clock speeds, 965 MHz base and 993 MHz boost versus 575 MHz base and 628 MHz boost for the GT 735M, which helps compensate for the lower core counts.

Neither GPU supports ray tracing cores or tensor cores, as both are from older generation mobile lineups. The GT 735M is part of the GeForce 700M generation, while the 920MX belongs to the GeForce 900M generation. Both support DirectX 12 (11_0), OpenGL 4.6, and share the same PCIe 3.0 x8 bus interface. The Vulkan API support differs: the GT 735M supports Vulkan 1.2.175, while the 920MX supports Vulkan 1.4.

The GT 735M has a slot width of IGP (integrated graphics processor), while the 920MX is an MXM Module. Both have no power connectors and portable-device-dependent display outputs. The GT 735M was released earlier, on 2013-03-31, while the 920MX came later on 2016-03-24. The GT 735M's predecessor is the GeForce 600M and its successor is the GeForce 800M, while the 920MX's predecessor is the GeForce 800M and its successor is GeForce 10 Mobile.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, where the GeForce 920MX scores 4,068 against the GeForce GT 735M's 3,616. This represents an 11.1% advantage for the 920MX, as indicated by the deltaPct of -11.1% (negative meaning the item A, GT 735M, is behind). This is a decisive win for the newer Maxwell-based GPU.

The victory is somewhat counterintuitive given the GT 735M's higher shading unit count (384 vs. 256) and TMU count (32 vs. 24). However, the 920MX's higher clock speeds, 965 MHz base and 993 MHz boost versus 575 MHz and 628 MHz, translate to a raw compute advantage. The FP32 performance reflects this: the 920MX delivers 508.4 GFLOPS versus the GT 735M's 482.3 GFLOPS, a 5.4% margin in theoretical floating-point throughput.

The 920MX also wins on pixel rate (7.944 GPixel/s vs. 5.024 GPixel/s, a 58.1% advantage) and texture rate (23.83 GTexel/s vs. 20.10 GTexel/s, an 18.6% advantage). These rates are derived from the combination of clock speeds and ROP/TMU counts, and they align with the OpenCL benchmark result showing the 920MX ahead.

However, the average benchmark scores tell a slightly different story. The GT 735M has an average benchmark score of 3,616, while the 920MX has an average of 3,528. This puts the GT 735M 2.4% ahead of the 920MX on average. The 920MX's nearest rivals list includes the GT 735M with a deltaPct of -2.4%, meaning the 920MX trails the GT 735M in average score. This discrepancy suggests that while the 920MX wins the OpenCL test, the GT 735M may perform better on other unlisted benchmarks that contribute to the average.

Both GPUs sit at the 21st percentile among all GPUs, indicating they are in the same performance tier overall. The GT 735M's nearest rivals include the GTX 1050 (average score 3,629, deltaPct -0.3%), RTX 5000 Mobile Ada (3,596, +0.6%), GT 545 (3,594, +0.6%), and Radeon HD 6770 (3,649, -0.9%). The 920MX's nearest rivals include the GT 545 (3,594, -1.8%), RTX 5000 Mobile Ada (3,596, -1.9%), GT 735M (3,616, -2.4%), and Quadro 2000M (3,434, +2.7%).

Specification Differences

The two GPUs differ on several key specification fields:

| Specification | GeForce GT 735M | GeForce 920MX |

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

| Architecture | Kepler 2.0 | Maxwell |

| Chip | GK208 | GM108S |

| Generation | GeForce 700M | GeForce 900M |

| Die Size | 87 mm² | 77 mm² |

| Transistor Density | 11.7M / mm² | 13.2M / mm² |

| Base Clock | 575 MHz | 965 MHz |

| Boost Clock | 628 MHz | 993 MHz |

| Shading Units | 384 | 256 |

| TMUs | 32 | 24 |

| Pixel Rate | 5.024 GPixel/s | 7.944 GPixel/s |

| Texture Rate | 20.10 GTexel/s | 23.83 GTexel/s |

| FP32 | 482.3 GFLOPS | 508.4 GFLOPS |

| TDP | 33 W | 16 W |

| Slot Width | IGP | MXM Module |

| Vulkan API | 1.2.175 | 1.4 |

| Release Date | 2013-03-31 | 2016-03-24 |

| Predecessor | GeForce 600M | GeForce 800M |

| Successor | GeForce 800M | GeForce 10 Mobile |

Fields that are identical include memory size (2 GB), memory type (DDR3), bus width (64 bit), memory bandwidth (14.40 GB/s), memory clock (900 MHz / 1,800 Mbps effective), ROPs (8), process node (28 nm), foundry (TSMC), transistor count (1,020 million), power connectors (None), bus interface (PCIe 3.0 x8), display outputs, DirectX support (12 (11_0)), OpenGL support (4.6), production status (End-of-life), and launch MSRP (both null).

The Verdict

The data presents a nuanced picture. In the only direct head-to-head benchmark, the GeForce 920MX is clearly superior, delivering 4,068 points versus 3,616 for the GT 735M, an 11.1% margin. The 920MX also achieves higher pixel rates, texture rates, and FP32 throughput despite having fewer shading units and TMUs. Its lower TDP of 16 W versus 33 W makes it substantially more power-efficient.

However, the average benchmark score tells a different story. The GT 735M's average score of 3,616 exceeds the 920MX's 3,528 by 2.4%. This suggests that across a broader set of tests, the GT 735M may hold an advantage in certain workloads, even though it loses the specific OpenCL test. Both GPUs occupy the same 21st percentile among all GPUs, confirming they are in the same performance bracket.

For users prioritizing the specific OpenCL workload, the 920MX is the clear choice. Its higher clock speeds and architectural efficiency (Maxwell over Kepler 2.0) deliver tangible gains. For users looking at the aggregate benchmark profile, the GT 735M's higher average score indicates more consistent performance across varied tests. The 920MX's much lower TDP (16 W vs. 33 W) makes it the better option for thermally constrained laptops, while the GT 735M's IGP form factor may suit integrated designs.

Where Each One Wins

NVIDIA GeForce 920MX wins on: the Geekbench OpenCL benchmark (4,068 vs. 3,616, an 11.1% lead), pixel rate (7.944 GPixel/s vs. 5.024 GPixel/s), texture rate (23.83 GTexel/s vs. 20.10 GTexel/s), FP32 throughput (508.4 GFLOPS vs. 482.3 GFLOPS), and power efficiency (TDP of 16 W vs. 33 W). It also offers a newer architecture (Maxwell), higher transistor density (13.2M / mm²), a smaller die (77 mm²), faster base and boost clocks, and newer Vulkan API support (1.4 vs. 1.2.175). The 920MX is the better choice for OpenCL compute tasks, and its lower power draw makes it suitable for thinner, cooler-running notebooks.

NVIDIA GeForce GT 735M wins on: average benchmark score (3,616 vs. 3,528, a 2.4% margin), shading unit count (384 vs. 256), TMU count (32 vs. 24), and die size (87 mm² vs. 77 mm², which may indicate more physical resources). It also has a higher relative standing against its nearest rivals, it leads the RTX 5000 Mobile Ada and GT 545 by 0.6% each, while the 920MX trails them by 1.9% and 1.8% respectively. The GT 735M's higher average suggests it may perform better on benchmarks beyond the single OpenCL test. Its IGP slot width also makes it a candidate for integrated motherboard designs, whereas the 920MX requires an MXM module slot. Users who value a broader performance profile across multiple benchmark types may prefer the GT 735M.