NVIDIA GeForce GT 735M vs NVIDIA GeForce GT 755M Comparison

The NVIDIA GeForce GT 755M and NVIDIA GeForce GT 735M are both members of the GeForce 700M mobile generation, both built on TSMC's 28 nm process, and both long past their production lives, yet the recorded data shows they are not close performers. The GT 755M's average benchmark score of 4033 sits in the 24th percentile of all GPUs in the database, while the GT 735M's average of 3616 places it in the 21st percentile. The gap widens dramatically in the direct head-to-head OpenCL measurement, where the GT 755M leads by 36.4 percent. Both chips share the same 384 shading units and 32 texture mapping units, but the GT 755M pairs those resources with GDDR5 memory on a 128-bit bus and a substantially higher clock, while the GT 735M is constrained by DDR3 on a narrow 64-bit bus and less than two-thirds of the pixel output hardware. For any workload where throughput matters, the data clearly favors the GT 755M; the GT 735M's only quantified advantage is a lower thermal envelope.

The Verdict

The GT 755M is the pick whenever compute or graphics performance is the priority. Its OpenCL score of 4934 against the GT 735M's 3616 is a 36.4 percent advantage, and that margin understates the structural differences: 86.40 GB/s of memory bandwidth versus 14.40 GB/s, 16 render output units versus 8, and a boost clock of 1020 MHz versus 628 MHz. Any workload that streams textures, writes pixels, or shifts large buffers will feel the six-fold bandwidth difference most keenly. The GT 755M also carries a Metal score of 3132 in the database, a test for which the GT 735M has no recorded result.

The GT 735M makes sense only where power and thermal headroom dominate the decision. Its 33 W thermal design point against the GT 755M's 50 W is a meaningful reduction, and its 87 mm² die is considerably smaller than the 755M's 118 mm², which allows it to fit into tighter system designs, reflected in its IGP-class slot classification versus the 755M's MXM module format. Neither card is competitive with modern hardware: percentile placements of 24 and 21 against all recorded GPUs put both firmly in the entry-level bracket of the database.

Notably, the rivals clustering around each chip tell their own story. The GT 755M's average score of 4033 lands within a percent of the AMD FirePro M4150 (4013 average, a 0.5 percent gap) and the Intel HD Graphics 630 (4075, a 1 percent gap in Intel's favor), while the GT 735M's 3616 sits in the immediate neighborhood of the NVIDIA GeForce GTX 1050 (3629 average, 0.3 percent ahead) and the AMD Radeon HD 6770 (3649, 0.9 percent ahead). These are the performance neighborhoods buyers should expect.

Architecture Differences

Both GPUs descend from NVIDIA's Kepler lineage, but they are not the same design generation. The GT 755M uses the GK107 chip with the original Kepler architecture, while the GT 735M uses the GK208 chip under the Kepler 2.0 label. Both were fabricated by TSMC on a 28 nm process, and both belong to the GeForce 700M generation, sandwiched between the GeForce 600M predecessors and GeForce 800M successors.

The silicon budgets differ considerably. The GK107 die in the GT 755M measures 118 mm² and carries 1,270 million transistors, giving a density of 10.8M transistors per mm². The GK208 die is smaller and leaner at 87 mm² and 1,020 million transistors, but thanks to its tighter footprint it actually achieves a higher density of 11.7M per mm². The smaller die is part of why the GT 735M fits the IGP slot classification while the GT 755M ships as an MXM module.

Despite the shared 384 shading units and 32 texture units, the back ends diverge sharply. The GT 755M has 16 render output units; the GT 735M has 8. Clocks compound this: the GT 755M runs a 980 MHz base and 1020 MHz boost, while the GT 735M manages 575 MHz base and 628 MHz boost. The resulting compute outputs are 783.4 GFLOPS FP32 for the GT 755M versus 482.3 GFLOPS for the GT 735M, a difference of roughly 62 percent in raw shader throughput.

The memory subsystems are the widest gulf. The GT 755M uses 2 GB of GDDR5 on a 128-bit bus, clocked at 1350 MHz for 5.4 Gbps effective, delivering 86.40 GB/s. The GT 735M uses 2 GB of DDR3 on a 64-bit bus at 900 MHz for 1800 Mbps effective, delivering just 14.40 GB/s. The interface differs too: the GT 755M connects over PCIe 3.0 x16 while the GT 735M is limited to PCIe 3.0 x8.

Software support is identical. Both report DirectX 12 (11_0 feature level), OpenGL 4.6, and Vulkan 1.2.175, and both list portable-device-dependent display outputs. Neither chip has ray tracing or tensor hardware, as expected for this era.

FAQ

Q: How much faster is the GT 755M than the GT 735M in the recorded benchmarks?

A: In the Geekbench OpenCL test, the GT 755M scored 4934 against the GT 735M's 3616, a 36.4 percent advantage. This is the only head-to-head test in the database, and it went decisively to the GT 755M.

Q: Do the two GPUs have the same number of shader cores?

A: Yes. Both have 384 shading units and 32 texture mapping units. The performance gap comes from clock speeds, memory bandwidth, and render output units, not core counts.

Q: How much more memory bandwidth does the GT 755M have?

A: The GT 755M delivers 86.40 GB/s from its GDDR5 memory on a 128-bit bus. The GT 735M delivers 14.40 GB/s from DDR3 on a 64-bit bus. That is a six-fold advantage for the GT 755M.

Q: Which GPU consumes less power?

A: The GT 735M, with a 33 W thermal design point versus 50 W for the GT 755M. Neither requires external power connectors.

Q: Where do these GPUs rank against everything in the database?

A: The GT 755M sits in the 24th percentile of all GPUs by average benchmark score; the GT 735M sits in the 21st percentile. Both are entry-level performers by current database standards.

Q: Which rival GPUs perform most similarly?

A: The GT 755M's average of 4033 is within 0.5 percent of the AMD FirePro M4150 (4013) and within 1.4 percent of the AMD Radeon HD 6850 X2 (3977). The GT 735M's 3616 is within 0.3 percent of the NVIDIA GeForce GTX 1050 (3629) and within 0.6 percent of the NVIDIA GeForce GT 545 (3594) and the NVIDIA RTX 5000 Mobile Ada Generation (3596).

Specification Differences

The table below lists only the fields where the two GPUs differ.

| Field | NVIDIA GeForce GT 755M | NVIDIA GeForce GT 735M |

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

| Chip | GK107 | GK208 |

| Architecture | Kepler | Kepler 2.0 |

| Transistors | 1,270 million | 1,020 million |

| Die size | 118 mm² | 87 mm² |

| Transistor density | 10.8M / mm² | 11.7M / mm² |

| Base clock | 980 MHz | 575 MHz |

| Boost clock | 1020 MHz | 628 MHz |

| Memory clock | 1350 MHz (5.4 Gbps effective) | 900 MHz (1800 Mbps effective) |

| Memory type | GDDR5 | DDR3 |

| Bus width | 128 bit | 64 bit |

| Memory bandwidth | 86.40 GB/s | 14.40 GB/s |

| Render output units | 16 | 8 |

| Pixel rate | 8.160 GPixel/s | 5.024 GPixel/s |

| Texture rate | 32.64 GTexel/s | 20.10 GTexel/s |

| FP32 compute | 783.4 GFLOPS | 482.3 GFLOPS |

| Thermal design point | 50 W | 33 W |

| Slot width | MXM Module | IGP |

| Bus interface | PCIe 3.0 x16 | PCIe 3.0 x8 |

| Release date | 2013-06-24 | 2013-03-31 |

| Average benchmark score | 4033 | 3616 |

| Percentile vs all GPUs | 24 | 21 |

Fields that match: manufacturer (NVIDIA), process node (28 nm, TSMC), memory size (2 GB), shading units (384), texture units (32), API support (DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.2.175), production status (End-of-life), generation (GeForce 700M), predecessor (GeForce 600M), successor (GeForce 800M), and power connectors (none).

Head-to-Head Benchmarks

The database contains one direct head-to-head result between these two GPUs, and it is not close.

Geekbench OpenCL: GT 755M wins, 4934 vs 3616 (36.4 percent). This is the defining measurement of the pairing. A 36.4 percent margin in a general compute test reflects the compound effect of every hardware difference: the GT 755M's 1020 MHz boost clock is 62 percent higher than the GT 735M's 628 MHz, its FP32 throughput of 783.4 GFLOPS exceeds the 735M's 482.3 GFLOPS by the same proportion, and its 86.40 GB/s of memory bandwidth ensures the shader array is fed rather than starved. The GT 735M's 14.40 GB/s of DDR3 bandwidth is the single most restrictive figure in its column of the spec sheet, and it is the most plausible explanation for why the realized performance gap, while large, does not fully track the compute throughput difference in workloads that lean on memory.

The supporting rates tell the same story. Pixel fill rate favors the GT 755M at 8.160 GPixel/s against 5.024 GPixel/s, driven by both the doubled render output unit count (16 versus 8) and the higher clock. Texture fill rate is 32.64 GTexel/s versus 20.10 GTexel/s, a 62 percent edge that mirrors the clock and shader arithmetic exactly, since texture unit counts are equal at 32.

Against the broader database, the GT 755M's average score of 4033 keeps company with the AMD FirePro M4150 at 4013 (0.5 percent behind the 755M), the AMD Radeon HD 6850 X2 at 3977 (1.4 percent behind), and sits within striking distance of the Intel HD Graphics 630 at 4075 (1 percent ahead of the 755M) and the AMD Radeon RX 9060 XT 8 GB at 4093 (1.5 percent ahead). The GT 735M's 3616 average is bracketed by the NVIDIA GeForce GTX 1050 at 3629, the NVIDIA RTX 5000 Mobile Ada Generation at 3596, the NVIDIA GeForce GT 545 at 3594, and the AMD Radeon HD 6770 at 3649, all within a percent of it.

The final tally in the recorded head-to-head data: one win for the GT 755M, zero for the GT 735M. The GT 735M's case rests entirely on its 33 W thermal design point, its smaller 87 mm² die, and its IGP-class integration, none of which translate into benchmark victories. Where performance is the criterion, the GT 755M wins every recorded comparison.