NVIDIA GeForce GT 545 vs NVIDIA GeForce GT 635M Comparison

The NVIDIA GeForce GT 635M and NVIDIA GeForce GT 545 are both end-of-life Fermi-based parts, but they target very different segments of the market. The GT 635M is a mobile IGP designed for laptops, while the GT 545 is a desktop card with a larger die and higher power envelope. Benchmark data shows a close contest in raw compute, but the architectural and physical differences tell a more nuanced story.

Head-to-Head Benchmarks

The only available benchmark comparison is Geekbench OpenCL, where the GeForce GT 635M scores 3740 against the GeForce GT 545's 3594. This gives the GT 635M a 4.1% lead, a modest but definitive margin. In the context of their respective peer groups, this result is interesting. The GT 635M sits in the 22nd percentile of all GPUs, while the GT 545 rests just one point lower at the 21st percentile, indicating that both are entry-level performers by modern standards.

The 4.1% delta is small enough to be considered within run-to-run variance on some workloads, but the data shows a consistent winner in the GT 635M. When examining the GT 635M's nearest rivals, its score of 3740 places it 0.6% ahead of both the Quadro 3000M (3718) and the GeForce GT 740M (3717), while also edging out the GeForce 825M (3694) by 1.2%. It trails the Intel UHD Graphics 710 (3792) by 1.4%, however. For the GT 545, its 3594 score puts it just 0.1% behind the RTX 5000 Mobile Ada Generation (3596) — a signal of how low that professional mobile part scores in this particular test — and 0.6% behind the GeForce GT 735M (3616). It falls further behind the GeForce GTX 1050 (3629) by 1% and the AMD Radeon HD 6770 (3649) by 1.5%.

The fact that the mobile GT 635M outperforms the desktop GT 545 in this compute-oriented benchmark is noteworthy, especially given the GT 545's substantially higher theoretical FP32 throughput of 414.7 GFLOPS versus 182.4 GFLOPS for the GT 635M. The GT 545 also has a significant advantage in pixel rate (4.320 GPixel/s vs 1.900 GPixel/s) and texture rate (17.28 GTexel/s vs 7.600 GTexel/s). Yet in the OpenCL test, the GT 635M comes out ahead, suggesting that the benchmark may favor memory bandwidth efficiency or driver optimizations for the newer mobile architecture. The GT 635M's memory runs at 1800 Mbps effective versus 1600 Mbps for the GT 545, which could partially explain the result despite the GT 545's wider 192-bit bus.

FAQ

Q: Which GPU has a higher Geekbench OpenCL score?

A: The NVIDIA GeForce GT 635M scores 3740, which is 4.1% higher than the NVIDIA GeForce GT 545's score of 3594.

Q: How does the GT 635M compare to the GT 740M?

A: The GT 635M is 0.6% faster, scoring 3740 versus the GT 740M's 3717, making them nearly identical in performance.

Q: What is the performance gap between the GT 545 and the GeForce GTX 1050?

A: The GTX 1050 scores 3629, which is 1% higher than the GT 545's 3594.

Q: Does the GT 545 outperform the GT 635M in any benchmark category?

A: In the single reported benchmark (Geekbench OpenCL), the GT 635M wins. However, the GT 545 has significantly higher theoretical pixel rate (4.320 GPixel/s vs 1.900 GPixel/s) and texture rate (17.28 GTexel/s vs 7.600 GTexel/s), suggesting it would win in fill-rate-bound scenarios.

Q: How does the GT 635M compare to the Intel UHD Graphics 710?

A: The Intel UHD Graphics 710 scores 3792, which is 1.4% higher than the GT 635M's 3740, making the Intel part slightly faster in this test.

Q: What is the percentile ranking of each GPU?

A: The GT 635M ranks in the 22nd percentile of all GPUs, while the GT 545 ranks in the 21st percentile.

Where Each One Wins

The GeForce GT 635M wins the only direct benchmark comparison, taking the Geekbench OpenCL test by 4.1%. This makes it the better choice for general compute workloads and OpenCL-accelerated applications. Its higher memory clock (1800 Mbps effective vs 1600 Mbps) likely contributes to this result, and its lower 35 W TDP makes it far more suitable for thin-and-light laptops where power efficiency is critical.

The GeForce GT 545, despite losing the benchmark, wins decisively in raw throughput metrics. Its FP32 performance of 414.7 GFLOPS is more than double the GT 635M's 182.4 GFLOPS, and its pixel rate of 4.320 GPixel/s is 127% higher. The GT 545 also has 16 ROPs compared to just 4 on the GT 635M, which means it would dramatically outperform the mobile part in traditional rasterization workloads like older games or 2D rendering. Its 38.40 GB/s memory bandwidth also exceeds the GT 635M's 28.80 GB/s, giving it an edge in bandwidth-sensitive tasks. For desktop users with a 250 W power supply, the GT 545 offers a more capable fill-rate performance profile, albeit with a 70 W TDP that is double the GT 635M's.

Specification Differences

The two GPUs diverge significantly in memory configuration. The GT 635M features 2 GB of DDR3 memory on a 128-bit bus, while the GT 545 has 1536 MB (1.5 GB) on a 192-bit bus. Despite the smaller capacity, the GT 545's wider bus provides higher bandwidth at 38.40 GB/s versus 28.80 GB/s. Memory clocks also differ, with the GT 635M running at 900 MHz (1800 Mbps effective) and the GT 545 at 800 MHz (1600 Mbps effective).

Compute resources show a clear hierarchy: the GT 545 has 144 shading units, 24 TMUs, and 16 ROPs, while the GT 635M has 96 shading units, 16 TMUs, and only 4 ROPs. The GT 545's pixel rate is 4.320 GPixel/s compared to 1.900 GPixel/s, and its texture rate is 17.28 GTexel/s versus 7.600 GTexel/s. The FP32 compute rating is also lopsided: 414.7 GFLOPS for the GT 545 versus 182.4 GFLOPS for the GT 635M.

Physical specifications differ substantially. The GT 635M is an IGP (integrated graphics processor) with slot width listed as "IGP," meaning it is designed for portable devices with no power connectors and a 35 W TDP. The GT 545 is a single-slot desktop card measuring 145 mm (5.7 inches) in length, with a 70 W TDP and a suggested PSU of 250 W. Display outputs also differ: the GT 635M has "Portable Device Dependent" outputs, while the GT 545 offers 1x DVI, 1x HDMI 1.3a, and 1x VGA.

Architecture Differences

Both GPUs are built on the 40 nm process at TSMC, but they use different chips and architecture revisions. The GT 635M uses the GF108 chip with Fermi architecture, while the GT 545 uses the GF116 chip with Fermi 2.0 architecture. This is a notable distinction: Fermi 2.0 on the GT 545 represents a refined iteration of the Fermi design, though both share the same fundamental architecture family.

The transistor counts reveal a major difference in complexity. The GT 545's GF116 chip packs 1,170 million transistors on a 238 mm² die, while the GT 635M's GF108 has only 585 million transistors on a 116 mm² die. This means the GT 545 has exactly double the transistor count and more than twice the die area. The transistor density is nearly identical (5.0M / mm² for the GT 635M and 4.9M / mm² for the GT 545), indicating that the manufacturing process is the same and the difference is purely in die size and complexity.

In terms of feature support, both cards expose DirectX 12 (11_0) and OpenGL 4.6, and neither lists Vulkan support. The GT 635M belongs to the GeForce 600M generation and is the successor to the GeForce 500M, while the GT 545 belongs to the GeForce 500 generation and is the successor to the GeForce 400. The GT 635M's successor is the GeForce 700M, and the GT 545's successor is the GeForce 600. The GT 545 had a launch MSRP of 149 USD, while no MSRP is listed for the GT 635M. Both are end-of-life products, with the GT 545 releasing in May 2011 and the GT 635M following in March 2012.