NVIDIA GeForce GT 730M vs NVIDIA GeForce GT 735M Comparison

The NVIDIA GeForce GT 735M and GT 730M are two end-of-life mobile graphics solutions from the GeForce 700M generation, both built on a 28 nm process at TSMC. They share the same 33 W TDP and 2 GB of DDR3 memory, but the benchmark data reveals a clear performance hierarchy that is not immediately obvious from their similar naming. The GT 735M holds a decisive advantage in the one head-to-head test available, while the GT 730M counters with a broader feature set and architectural differences that matter for certain workloads.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and the results are stark. The GT 735M scores 3616 points, while the GT 730M trails with 3107 points. That represents a 16.4% lead for the GT 735M, a substantial margin in mobile graphics where every frame matters. The GT 735M’s average benchmark score of 3616 is also its only score, meaning it consistently delivers that level of compute performance. In contrast, the GT 730M’s OpenCL result is its weakest showing, as it also posts a Vulkan score of 3524, which is significantly higher than its OpenCL number.

Looking at the rival landscape reinforces the GT 735M’s position. Its score of 3616 places it within 0.3% of the NVIDIA GeForce GTX 1050, which averages 3629 points. It is also 0.6% ahead of both the NVIDIA RTX 5000 Mobile Ada Generation (3596) and the NVIDIA GeForce GT 545 (3594). The only rival that edges it out is the AMD Radeon HD 6770 at 3649 points, a 0.9% gap. This clustering around the 3600-point mark suggests the GT 735M sits at a performance plateau where it trades blows with much newer and older hardware alike.

The GT 730M, by contrast, finds itself in a lower tier. Its average benchmark score of 3316 (the mean of its OpenCL and Vulkan results) puts it just 0.5% behind the Intel HD Graphics 530 (3332) and 0.9% ahead of the NVIDIA GeForce 920M (3287). It is 2.2% behind the Intel HD Graphics P4600 (3389) and 3.3% ahead of the NVIDIA GeForce GT 640 (3210). The 16.4% delta between the two GT 700M parts is far larger than any gap between the GT 730M and its closest rivals, underscoring that the GT 735M is the clear compute winner.

The Verdict

The data is unambiguous: the NVIDIA GeForce GT 735M is the faster GPU for raw compute workloads. Its 16.4% OpenCL advantage over the GT 730M is not a marginal difference; it is a decisive one that will translate into measurably higher frame rates in applications that leverage OpenCL acceleration. The GT 735M’s average score of 3616 also places it in the 21st percentile of all GPUs, while the GT 730M’s average of 3316 lands in the 20th percentile. That one-percentile gap, combined with the head-to-head result, means users prioritizing compute performance should choose the GT 735M without hesitation.

However, the GT 730M is not without its merits. It offers a Vulkan score of 3524, which is only 2.5% lower than the GT 735M’s OpenCL score. Since the GT 735M has no Vulkan result in the data, the GT 730M is the only one of the two proven to work with that modern API. For users running Vulkan-based titles or applications, the GT 730M provides a verified path forward, even if its OpenCL performance lags. The GT 730M also comes in a MXM Module form factor, whereas the GT 735M is an IGP (integrated graphics processor), which may influence upgradeability or system integration decisions.

The verdict is a split decision. For compute-heavy tasks measured by OpenCL, the GT 735M is the superior choice. For those needing Vulkan support or a discrete MXM module, the GT 730M is the viable option despite its lower raw scores.

Where Each One Wins

The GT 735M wins in every scenario where OpenCL compute performance is the primary metric. Its 3616 OpenCL score is 16.4% higher than the GT 730M’s 3107, and it matches or exceeds the performance of far more modern parts like the GTX 1050 (within 0.3%) and the RTX 5000 Mobile Ada Generation (0.6% ahead). This makes it the better choice for general-purpose GPU computing, video encoding, and any workload that harnesses OpenCL. Its higher 482.3 GFLOPS of FP32 performance, compared to the GT 730M’s 556.8 GFLOPS, might seem contradictory, but the benchmark results show the GT 735M’s architecture executes OpenCL more efficiently despite lower theoretical throughput.

The GT 730M wins in the Vulkan arena, where it posts a 3524 score that is not matched by any GT 735M data point. This is a critical differentiator for users running Vulkan-based games or modern graphics APIs. The GT 730M also benefits from a wider 128-bit memory bus, delivering 28.80 GB/s of bandwidth versus the GT 735M’s 64-bit bus and 14.40 GB/s. While this does not help its OpenCL score, it suggests the GT 730M may handle memory-intensive tasks more gracefully in scenarios not covered by the benchmarks. Additionally, the GT 730M’s MXM module form factor offers physical flexibility that the GT 735M’s IGP design cannot match.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The NVIDIA GeForce GT 735M scores 3616 on Geekbench OpenCL, which is 16.4% higher than the GT 730M’s score of 3107.

Q: Does the GT 730M support Vulkan?

A: Yes, the GT 730M has a Geekbench Vulkan score of 3524. The GT 735M has no Vulkan score listed in the data.

Q: How does the GT 735M compare to a GTX 1050?

A: The GT 735M scores 3616, which is 0.3% lower than the GTX 1050’s average score of 3629. The difference is negligible.

Q: What is the memory bandwidth difference between the two?

A: The GT 730M has a 128-bit bus with 28.80 GB/s bandwidth, exactly double the GT 735M’s 64-bit bus with 14.40 GB/s.

Q: Do both GPUs have the same transistor count?

A: No, the GT 730M uses the GK107 chip with 1,270 million transistors, while the GT 735M uses the GK208 chip with 1,020 million transistors.

Q: Which GPU has a higher pixel fill rate?

A: The GT 730M has a pixel rate of 5.800 GPixel/s, while the GT 735M has a pixel rate of 5.024 GPixel/s.

Architecture Differences

The two GPUs are built on different chips, which explains their divergent performance profiles. The GT 735M uses the GK208 chip, which is based on the Kepler 2.0 architecture, while the GT 730M uses the GK107 chip on the original Kepler architecture. Both are fabricated on a 28 nm process at TSMC, but the die sizes differ: the GK208 measures 87 mm² with 1,020 million transistors, while the GK107 is larger at 118 mm² with 1,270 million transistors. The transistor density tells the story: the GK208 packs 11.7 million transistors per mm² versus 10.8 million per mm² on the GK107, indicating a denser, more optimized design on the GT 735M.

Memory configuration is another major divergence. The GT 730M features a 128-bit memory bus with 28.80 GB/s of bandwidth, while the GT 735M is limited to a 64-bit bus with 14.40 GB/s. Both use DDR3 memory at 900 MHz with 1800 Mbps effective speed. The memory clock is identical, but the bus width halves the GT 735M’s bandwidth. Despite this disadvantage, the GT 735M still wins the compute benchmark, suggesting its architecture compensates with better execution efficiency.

The render output units (ROPs) also differ: the GT 730M has 16 ROPs versus 8 on the GT 735M, which contributes to its higher pixel rate of 5.800 GPixel/s versus 5.024 GPixel/s. Both have 384 shading units and 32 texture mapping units, but the GT 730M’s higher base and boost clock of 725 MHz (versus 575 MHz base and 628 MHz boost on the GT 735M) gives it a theoretical texture rate of 23.20 GTexel/s versus 20.10 GTexel/s. The FP32 compute is also higher on the GT 730M at 556.8 GFLOPS versus 482.3 GFLOPS, yet the GT 735M still wins in practice, highlighting that raw specs do not always predict real-world results.

The bus interface and form factor differ as well. The GT 735M uses a PCIe 3.0 x8 interface and is an IGP, while the GT 730M uses PCIe 3.0 x16 and is an MXM module. Both have no power connectors and share the same 33 W TDP. Their API support is identical: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The GT 730M launched earlier on January 19, 2013, while the GT 735M followed on March 31, 2013. Both are end-of-life products with no launch MSRP listed, and both succeed the GeForce 600M series while being succeeded by the GeForce 800M series.