GPU Comparison

The NVIDIA GeForce 920MX and NVIDIA GeForce GT 730M are two end-of-life mobile graphics solutions that target the entry-level laptop segment, separated by three years of architectural evolution. The 920MX, built on the Maxwell architecture and released in 2016, faces off against the 2013-era Kepler-based GT 730M. While both are equipped with 2 GB of DDR3 memory and share the same 28 nm TSMC process node, their underlying designs and benchmark results tell divergent stories depending on the workload. The data reveals a clear split: the 920MX dominates in OpenCL compute, while the GT 730M counters with superior Vulkan performance, making the choice highly dependent on the specific application.

Head-to-Head Benchmarks

The most striking difference in the benchmark data appears in the Geekbench OpenCL test, where the NVIDIA GeForce 920MX delivers a commanding victory. The 920MX scores 4068 points, while the GT 730M manages only 3107 points. This translates to a 30.9% advantage for the Maxwell-based part, a substantial margin that underscores the efficiency gains of the newer architecture. In raw compute throughput, the 920MX’s FP32 rating of 508.4 GFLOPS is actually lower than the GT 730M’s 556.8 GFLOPS, yet the OpenCL result shows the 920MX pulling far ahead. This suggests that architectural efficiency and driver optimizations play a larger role than peak theoretical numbers in this particular workload.

The Vulkan benchmark flips the script completely. Here, the GT 730M scores 3524 points, outperforming the 920MX’s 2987 points by a 15.2% margin. This is a notable reversal, as the 920MX supports Vulkan 1.4 while the GT 730M is limited to Vulkan 1.2.175. The newer API version on the 920MX does not translate into better performance in this test, indicating that the GT 730M’s larger memory bus and higher shading unit count provide a tangible advantage in graphics-oriented workloads. The GT 730M features 384 shading units compared to the 920MX’s 256, and a 128-bit memory bus versus the 920MX’s 64-bit interface, which likely explains its Vulkan superiority.

When looking at average benchmark scores across both tests, the 920MX holds a narrow edge. Its average score is 3528, while the GT 730M averages 3316, a difference of roughly 6.4%. The 920MX sits in the 21st percentile of all GPUs, while the GT 730M is just below at the 20th percentile. Both are firmly in the entry-level tier, but the 920MX’s position is marginally higher. In terms of nearest rivals, the 920MX is most closely matched with the NVIDIA GeForce GT 545, which scores 3594 (a 1.8% delta), while the GT 730M’s closest competitor is the Intel HD Graphics 530 at 3332 (a 0.5% delta). These comparisons reinforce that neither GPU is a performance powerhouse, but the 920MX has a slight overall edge in aggregate benchmarks.

Architecture Differences

The architectural divide between these two GPUs is significant, starting with the chip itself. The 920MX uses the GM108S chip based on the Maxwell architecture, while the GT 730M relies on the GK107 chip from the older Kepler architecture. Both are fabricated on the same 28 nm process at TSMC, but the transistor counts differ notably. The GT 730M packs 1,270 million transistors on a 118 mm² die, while the 920MX uses 1,020 million transistors on a much smaller 77 mm² die. This results in a higher transistor density for the 920MX at 13.2M per mm², compared to the GT 730M’s 10.8M per mm², reflecting the more compact and efficient Maxwell design.

Clock speeds also diverge sharply. The 920MX runs at a base clock of 965 MHz with a boost up to 993 MHz, while the GT 730M is locked at a flat 725 MHz for both base and boost. This 240 MHz base clock advantage for the 920MX contributes to its higher pixel rate of 7.944 GPixel/s, compared to the GT 730M’s 5.800 GPixel/s. Interestingly, the texture rates are nearly identical, 23.83 GTexel/s for the 920MX versus 23.20 GTexel/s for the GT 730M, despite the GT 730M having 32 TMUs against the 920MX’s 24. The GT 730M’s higher TMU count compensates for its lower clock speed in texture throughput.

Memory configuration is another major differentiator. Both GPUs feature 2 GB of DDR3 memory at an effective 1800 Mbps, but the memory bus widths diverge. The GT 730M uses a 128-bit bus, delivering 28.80 GB/s of bandwidth, while the 920MX is constrained to a 64-bit bus, yielding just 14.40 GB/s. This doubling of memory bandwidth for the GT 730M is a critical advantage in memory-intensive workloads and likely explains its Vulkan victory. The GT 730M also has double the ROPs (16 versus 8) and 384 shading units versus 256 on the 920MX, giving it a raw pixel-pushing capability that the newer part lacks.

Power consumption tells the opposite story. The 920MX is rated at a 16 W TDP, less than half of the GT 730M’s 33 W TDP. This efficiency gain is a hallmark of the Maxwell architecture and makes the 920MX a far better fit for thin-and-light laptops where thermal and power budgets are tight. Both cards use MXM modules with no power connectors and rely on PCIe 3.0, but the 920MX runs at x8 lanes while the GT 730M uses the full x16 interface. API support is similar for DirectX (12 with 11_0 feature level) and OpenGL 4.6 on both, but Vulkan support differs: the 920MX supports Vulkan 1.4, while the GT 730M is capped at Vulkan 1.2.175.

The Verdict

The data presents a nuanced picture with no clear overall winner. The NVIDIA GeForce 920MX wins the OpenCL compute test decisively with a 30.9% lead over the GT 730M, making it the preferred choice for general-purpose GPU compute and OpenCL-accelerated applications. Its higher clock speeds, newer architecture, and significantly lower TDP of 16 W versus 33 W make it the more efficient and modern solution for everyday tasks that leverage compute APIs. The 920MX’s average benchmark score of 3528 also edges out the GT 730M’s 3316, and its 21st percentile ranking is one point higher, indicating slightly better overall standing in the GPU landscape.

However, the GT 730M is the clear victor in Vulkan graphics performance, beating the 920MX by 15.2% in that specific test. Its double memory bandwidth (28.80 GB/s versus 14.40 GB/s), larger memory bus, and higher shading unit count (384 versus 256) give it a tangible advantage in modern graphics APIs. The GT 730M also matches the 920MX in DirectX 12 and OpenGL 4.6 support, so it remains a viable option for gaming and graphics workloads that rely on those APIs. Users prioritizing Vulkan-based titles or applications that are bandwidth-sensitive would find the GT 730M more capable.

For most users, the 920MX is the safer recommendation due to its newer architecture, better compute performance, lower power draw, and superior average benchmark score. The GT 730M’s strengths are more specialized, appealing primarily to those who run Vulkan workloads or need the higher memory bandwidth for specific tasks. The 920MX’s 30.9% OpenCL advantage is a massive margin that outweighs the 15.2% Vulkan deficit for general-purpose use. Ultimately, the choice comes down to workload: pick the 920MX for compute and efficiency, or the GT 730M for graphics-centric Vulkan applications.

FAQ

Q: Which GPU performs better in OpenCL benchmarks?

A: The NVIDIA GeForce 920MX is significantly faster, scoring 4068 in Geekbench OpenCL compared to the GT 730M’s 3107, a 30.9% advantage.

Q: Does the GT 730M have any advantage in modern graphics APIs?

A: Yes, the GT 730M wins the Geekbench Vulkan test with a score of 3524 versus the 920MX’s 2987, a 15.2% margin, despite the 920MX supporting Vulkan 1.4 compared to the GT 730M’s Vulkan 1.2.175.

Q: How do their memory bandwidths compare?

A: The GT 730M has double the memory bandwidth at 28.80 GB/s thanks to its 128-bit bus, while the 920MX is limited to 14.40 GB/s over a 64-bit bus, both using 2 GB of DDR3 memory.

Q: What are the power consumption differences?

A: The 920MX is far more power-efficient with a 16 W TDP, while the GT 730M consumes more than double at 33 W.

Q: Which GPU has a higher average benchmark score?

A: The 920MX averages 3528 across its benchmarks, placing it in the 21st percentile of all GPUs, while the GT 730M averages 3316 and sits in the 20th percentile.

Q: Are there significant architectural differences between the two?

A: Yes, the 920MX uses the Maxwell architecture with 256 shading units and 1,020 million transistors, while the GT 730M uses Kepler with 384 shading units and 1,270 million transistors, though both are on a 28 nm TSMC process.

Where Each One Wins

The NVIDIA GeForce 920MX is the clear winner in compute-oriented tasks. Its 30.9% lead in OpenCL performance over the GT 730M makes it the superior choice for applications that leverage GPGPU compute, such as video encoding, data processing, and OpenCL-accelerated productivity software. The 920MX also wins on efficiency, with a 16 W TDP that allows for thinner, cooler, and longer-lasting laptops. Its higher pixel rate of 7.944 GPixel/s and boost clock of 993 MHz give it an edge in tasks that benefit from raw clock speed. For users who prioritize battery life and compute performance over graphics-specific workloads, the 920MX is the obvious pick.

The NVIDIA GeForce GT 730M, conversely, wins in graphics-heavy scenarios, particularly those using Vulkan. Its 15.2% Vulkan benchmark victory demonstrates that its larger memory bus and higher shading unit count matter in modern graphics APIs. The GT 730M’s 28.80 GB/s memory bandwidth is double that of the 920MX, making it better suited for textures, framebuffer operations, and memory-intensive rendering. Its 384 shading units and 16 ROPs provide more parallel processing capability for pixel and vertex workloads. Gamers or users running Vulkan-based titles, or applications that are sensitive to memory bandwidth, will find the GT 730M more responsive.

In terms of overall standing, the 920MX edges ahead with an average score of 3528 versus 3316, and a higher percentile rank. The GT 730M’s nearest rival, the Intel HD Graphics 530, is just 0.5% behind it, while the 920MX’s closest competitor, the GT 545, is 1.8% ahead. The 920MX’s OpenCL win is more decisive than the GT 730M’s Vulkan win, suggesting that the 920MX has a broader performance advantage in compute-centric use cases. Ultimately, the 920MX is the better all-rounder, while the GT 730M is a niche pick for Vulkan enthusiasts or those needing high memory bandwidth on a tight budget.