NVIDIA GeForce 920MX vs NVIDIA GeForce GT 755M Comparison

The NVIDIA GeForce GT 755M and the NVIDIA GeForce 920MX represent two distinct approaches to mobile graphics from different generations of NVIDIA’s lineup. The data shows a clear performance hierarchy, but the differences extend beyond raw speed into architectural philosophy and feature support. The GT 755M is an older, higher-power Kepler part, while the 920MX is a newer, more efficient Maxwell chip. This analysis breaks down the recorded benchmark results and specification differences to determine where each GPU holds an advantage.

Where Each One Wins

The recorded benchmark data points to a straightforward split: the GT 755M wins in raw compute performance, while the 920MX wins in API compatibility and power efficiency. The only direct head-to-head benchmark in the database is Geekbench OpenCL, where the GT 755M scores 4934 against the 920MX’s 4068, a 21.3% advantage. This is a significant margin and establishes the GT 755M as the faster GPU for general-purpose compute workloads that rely on OpenCL.

The 920MX, however, counters with a newer feature set. It supports Vulkan 1.4, whereas the GT 755M is limited to Vulkan 1.2.175. This means the 920MX is better positioned for modern applications that take advantage of newer Vulkan extensions and optimizations. The 920MX also has a recorded Geekbench Vulkan score of 2987, a test the GT 755M does not have a recorded result for in the database. This suggests the 920MX can handle Vulkan workloads, while the GT 755M’s Vulkan capabilities are unknown from the available data.

Power consumption is another area where the 920MX wins decisively. The GT 755M has a TDP of 50 W, while the 920MX draws only 16 W. This is a dramatic difference. For thin-and-light laptops or systems where thermal and power budgets are tight, the 920MX is the clear choice. The GT 755M’s higher TDP indicates it requires a more robust cooling solution and a larger power envelope, which is typical of a performance-oriented mobile part.

In terms of overall standing, the GT 755M sits at the 24th percentile of all GPUs in the database, with an average benchmark score of 4033. The 920MX sits at the 21st percentile, with an average score of 3528. This 505-point gap in average score reinforces the GT 755M’s lead in general performance, but the 920MX’s efficiency and modern API support make it a more versatile option for specific use cases.

Architecture Differences

The two GPUs come from different architectural generations. The GT 755M is based on the Kepler architecture, using the GK107 chip, while the 920MX uses the Maxwell architecture with the GM108S chip. Both are manufactured on a 28 nm process at TSMC, but the transistor counts and die sizes differ notably.

The GT 755M packs 1,270 million transistors on a 118 mm² die, resulting in a transistor density of 10.8M per mm². The 920MX has 1,020 million transistors on a much smaller 77 mm² die, giving it a higher transistor density of 13.2M per mm². This suggests the Maxwell architecture in the 920MX is more efficient at packing transistors, which contributes to its lower power consumption.

The compute resources are heavily skewed toward the GT 755M. It features 384 shading units, 32 texture mapping units, and 16 ROPs. The 920MX has 256 shading units, 24 TMUs, and only 8 ROPs. This gives the GT 755M a clear advantage in pixel throughput and texture processing. The pixel rate of the GT 755M is 8.160 GPixel/s versus 7.944 GPixel/s for the 920MX, and the texture rate is 32.64 GTexel/s versus 23.83 GTexel/s. The FP32 compute performance is also much higher on the GT 755M at 783.4 GFLOPS, compared to 508.4 GFLOPS on the 920MX.

Memory architecture is another major differentiator. The GT 755M uses 2 GB of GDDR5 memory on a 128-bit bus, delivering 86.40 GB/s of bandwidth. The 920MX uses 2 GB of DDR3 memory on a 64-bit bus, which yields only 14.40 GB/s of bandwidth. This is a massive 6x difference in memory bandwidth. The GT 755M’s memory clock is listed as 1350 MHz with 5.4 Gbps effective data rate, while the 920MX runs at 900 MHz with 1800 Mbps effective. The GT 755M’s GDDR5 memory and wider bus make it far better suited for bandwidth-intensive workloads like high-resolution textures or compute tasks that access large datasets.

The bus interface also differs. The GT 755M uses PCIe 3.0 x16, while the 920MX uses PCIe 3.0 x8. This halves the potential data transfer rate between the GPU and the host system for the 920MX, which could impact workloads that rely on frequent CPU-GPU data exchanges.

Both GPUs share the same DirectX support at 12 (11_0) and OpenGL 4.6. The Vulkan support differs, as noted. The GT 755M is from the GeForce 700M generation, released on 2013-06-24, while the 920MX is from the GeForce 900M generation, released on 2016-03-24. The GT 755M’s predecessor is the GeForce 600M and its successor is the GeForce 800M. The 920MX’s predecessor is the GeForce 800M and its successor is the GeForce 10 Mobile. Both are listed as end-of-life production status.

The Verdict

The data presents a clear choice based on priorities. If the primary goal is raw compute performance and memory bandwidth, the GT 755M is the superior GPU. Its 21.3% lead in Geekbench OpenCL, combined with its 6x memory bandwidth advantage and significantly higher FP32 throughput, makes it the stronger option for demanding workloads. It also has a higher average benchmark score of 4033 versus 3528, and a higher percentile ranking at 24 versus 21.

If the priority is power efficiency and modern API support, the 920MX is the better choice. Its 16 W TDP is less than a third of the GT 755M’s 50 W, making it ideal for portable devices where battery life and thermals are critical. Its Vulkan 1.4 support is a meaningful upgrade over the GT 755M’s Vulkan 1.2.175, which could matter for future software compatibility.

The GT 755M is for users who need the most performance out of a mobile GPU without worrying about power constraints. The 920MX is for users who want a capable GPU that sips power and supports the latest graphics APIs. The data does not support a single winner; it supports two different winners for two different scenarios.

FAQ

Q: Which GPU is faster in OpenCL compute performance?

A: The GT 755M is faster. In the Geekbench OpenCL test, it scores 4934 compared to the 920MX’s 4068, a 21.3% advantage.

Q: How much more power does the GT 755M consume compared to the 920MX?

A: The GT 755M has a TDP of 50 W, while the 920MX has a TDP of 16 W. The GT 755M consumes 34 W more under load.

Q: Do both GPUs support the same DirectX version?

A: Yes, both support DirectX 12 (11_0) and OpenGL 4.6. They differ only in Vulkan support, with the 920MX supporting Vulkan 1.4 and the GT 755M supporting Vulkan 1.2.175.

Q: Which GPU has higher memory bandwidth?

A: The GT 755M has significantly higher memory bandwidth. It offers 86.40 GB/s over a 128-bit GDDR5 interface, while the 920MX offers 14.40 GB/s over a 64-bit DDR3 interface.

Q: What is the average benchmark score for each GPU?

A: The GT 755M has an average benchmark score of 4033, while the 920MX has an average score of 3528. This places them at the 24th and 21st percentiles of all GPUs, respectively.

Q: Which GPU has more shading units?

A: The GT 755M has 384 shading units, while the 920MX has 256 shading units. The GT 755M also has more TMUs and ROPs, 32 TMUs and 16 ROPs versus 24 TMUs and 24 ROPs for the 920MX.

Head-to-Head Benchmarks

The only recorded head-to-head benchmark is Geekbench OpenCL, and the GT 755M wins it convincingly. The GT 755M scores 4934, and the 920MX scores 4068, giving the GT 755M a 21.3% delta. This is a substantial margin that reflects the GT 755M’s superior compute resources, including 384 shading units versus 256, and its much higher memory bandwidth of 86.40 GB/s versus 14.40 GB/s. The GT 755M’s FP32 performance of 783.4 GFLOPS is also 54% higher than the 920MX’s 508.4 GFLOPS, which directly contributes to its OpenCL lead.

The 920MX does not win any recorded head-to-head benchmark, but its Vulkan support is a notable counterpoint. The database lists a Geekbench Vulkan score of 2987 for the 920MX, while the GT 755M has no Vulkan score recorded. This does not mean the GT 755M is incapable of Vulkan, but the 920MX has a verified result in that API. The 920MX’s Vulkan 1.4 support is also a full version ahead of the GT 755M’s Vulkan 1.2.175, which could translate to better performance or compatibility in Vulkan-based applications.

The average benchmark scores tell a similar story. The GT 755M averages 4033 across its recorded tests, while the 920MX averages 3528. This 505-point gap puts the GT 755M at the 24th percentile and the 920MX at the 21st percentile. The GT 755M’s nearest rivals include the AMD Radeon RX 9060 XT 8 GB, which scores 4093, and the Intel HD Graphics 630, which scores 4075. The GT 755M trails both by 1.5% and 1%, respectively. The 920MX’s nearest rivals include the NVIDIA GeForce GT 735M at 3616 and the NVIDIA Quadro 2000M at 3434, with the 920MX trailing the former by 2.4% and leading the latter by 2.7%.

Specification Differences

The GT 755M and 920MX differ across nearly every major specification category. The GT 755M uses the Kepler architecture with the GK107 chip, while the 920MX uses Maxwell with the GM108S chip. Both are 28 nm parts from TSMC, but the GT 755M has 1,270 million transistors on a 118 mm² die, while the 920MX has 1,020 million transistors on a 77 mm². The transistor density is higher on the 920MX at 13.2M per mm² versus 10.8M per mm².

The clock speeds are similar. The GT 755M has a base clock of 980 MHz and a boost clock of 1020 MHz, while the 920MX has a base clock of 965 MHz and a boost clock of 993 MHz. The memory clocks differ significantly. The GT 755M runs at 1350 MHz with 5.4 Gbps effective data rate, while the 920MX runs at 900 MHz with 1800 Mbps effective.

Memory configuration is a major split. The GT 755M has 2 GB of GDDR5 on a 128-bit bus, while the 920MX has 2 GB of DDR3 on a 64-bit bus. This results in 86.40 GB/s of bandwidth for the GT 755M versus 14.40 GB/s for the 920MX. The GT 755M also has more shading units, TMUs, and ROPs: 384, 32, and 16 respectively, compared to 256, 24, and 8 for the 920MX.

The pixel rate and texture rate favor the GT 755M, with 8.160 GPixel/s and 32.64 GTexel/s versus 7.944 GPixel/s and 23.83 GTexel/s. FP32 compute is 783.4 GFLOPS for the GT 755M and 508.4 GFLOPS for the 920MX. Neither GPU has RT cores or tensor cores.

The TDP is a major difference, with the GT 755M at 50 W and the 920MX at 16 W. Both use MXM modules with no power connectors. The bus interface differs as well, with the GT 755M using PCIe 3.0 x16 and the 920MX using PCIe 3.0 x8. The display outputs are listed as portable device dependent for both.

API support is identical for DirectX and OpenGL, with both at DirectX 12 (11_0 and OpenGL 4.6. Vulkan support differs by one version, with the GT 755M supporting Vulkan 1.2.175 and the 920MX supporting Vulkan 1.4. The release dates show the GT 755M from the GeForce 700M generation, released on 2013-06-15, and the 920MX from the GeForce 900M generation, released on 2016-03-24. The GT 755M’s predecessors and successors are the GeForce 600M and GeForce 800M, while the 920MX’s predecessor and successor are the GeForce 800M and GeForce 10 Mobile, respectively.