NVIDIA GeForce 930M vs NVIDIA GeForce GT 635M Comparison

Head-to-Head Benchmarks

The recorded benchmark data shows a single head-to-head result between the NVIDIA GeForce 930M and the NVIDIA GeForce GT 635M, and that result is a decisive win for the newer part. In the Geekbench OpenCL test, the GeForce 930M scores 5,046, while the GeForce GT 635M scores 3,740. That is a delta of 34.9%, meaning the 930M is roughly a third faster in raw compute throughput. This is not a marginal gap; it is a substantial generational leap that separates the two in almost every practical scenario.

The average benchmark score tells the same story. The GeForce 930M carries an average of 4,388 across its recorded tests, which includes both Geekbench OpenCL and Geekbench Vulkan results. The GeForce GT 635M has no Vulkan score in the database, and its average is simply its OpenCL result of 3,740. The difference in average scores is 648 points, or about 17.3% when calculated from the database figures. The 930M also sits at the 26th percentile of all GPUs in the database, while the GT 635M sits at the 22nd percentile. Both are low-end mobile parts, but the 930M is meaningfully higher in the distribution.

Looking at the nearest rivals in the database provides context for each chip's position. The GeForce 930M's closest competitor is the NVIDIA GeForce GT 645M, which averages 4,411, a delta of -0.5% relative to the 930M. That means the 930M and GT 645M are effectively tied in performance. The 930M also edges out the Intel Iris Pro Graphics 5200 (4,360, delta 0.7%), the NVIDIA GeForce RTX 4070 GDDR6 (4,335, delta 1.2%), and the AMD FirePro W2100 (4,295, delta 2.2%). It is worth remembering the RTX 4070 GDDR6 appears in this list only because its average score is oddly low in this particular database, not because it is a comparable part in any architectural sense.

The GeForce GT 635M's nearest rivals are similarly close in score. The NVIDIA Quadro 3000M averages 3,718, a delta of 0.6% above the GT 635M. The NVIDIA GeForce GT 740M averages 3,717, also a delta of 0.6%. The NVIDIA GeForce 825M averages 3,694, a delta of 1.2%. The Intel UHD Graphics 710 averages 3,792, which is 1.4% higher than the GT 635M. These deltas are all within a couple of percentage points, showing that the GT 635M sits in a tight cluster of low-end parts from roughly the same era.

The one head-to-head benchmark result is the only direct comparison available, but it is unambiguous. The 930M wins the only recorded test, and the win count in the database is 1 for the 930M and 0 for the GT 635M. No test in the database shows the GT 635M ahead.

The Verdict

Based strictly on the recorded data, the NVIDIA GeForce 930M is the stronger GPU. It wins the only head-to-head benchmark by 34.9%, it has a higher average benchmark score (4,388 versus 3,740), and it sits higher in the overall percentile ranking (26th versus 22nd). Anyone choosing between these two for a laptop purchase or an upgrade path should favor the 930M without hesitation, as the performance gap is large enough to be felt in everyday tasks that offload to the GPU.

The GeForce GT 635M is not without merit, but its merits are not in raw performance. It matches or slightly trails its own nearest rivals in the database, with deltas of less than 1.4% in either direction. It is a competent entry-level part from the Fermi era, but the 930M's Maxwell architecture simply delivers more compute per watt and per clock. The data shows no scenario where the GT 635M wins a recorded benchmark.

The verdict is clear: the 930M is the better choice for anyone who needs GPU compute performance, whether for OpenCL workloads, Vulkan-capable applications, or general multimedia acceleration. The GT 635M remains a usable part, but it is outclassed by a significant margin.

Where Each One Wins

The GeForce 930M wins in the only recorded benchmark category, Geekbench OpenCL, with a score of 5,046 versus 3,740. That is the sole head-to-head victory in the database. The 930M also wins on API support, as it records a Geekbench Vulkan score of 3,729, while the GT 635M has no Vulkan score at all. This means the 930M is capable of running Vulkan workloads, whereas the GT 635M's API support tops out at DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan entry listed.

The GeForce GT 635M does not win any recorded benchmark. Its only test result is the OpenCL score of 3,740, which is below the 930M's OpenCL result. However, the GT 635M does have a higher memory bandwidth in the specifications: 28.80 GB/s versus 12.80 GB/s for the 930M. This is due to the 128-bit bus width on the GT 635M versus the 64-bit bus on the 930M. That bandwidth advantage does not translate into a benchmark win in the recorded data, but it suggests the GT 635M could be less bottlenecked in memory-heavy tasks, even though its compute throughput is far lower.

The 930M wins on compute throughput in every measured category. Its FP32 performance is 421.6 GFLOPS, more than double the GT 635M's 182.4 GFLOPS. Its pixel rate is 4.392 GPixel/s versus 1.900 GPixel/s, and its texture rate is 13.18 GTexel/s versus 7.600 GTexel/s. These are not benchmark scores, but they are the underlying hardware rates that explain the benchmark gap.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 930M has an average benchmark score of 4,388, while the NVIDIA GeForce GT 635M has an average of 3,740.

Q: What is the performance difference in the head-to-head OpenCL test?

A: The GeForce 930M scores 5,046, and the GeForce GT 635M scores 3,740. That is a 34.9% delta in favor of the 930M.

Q: Does the GeForce GT 635M support Vulkan?

A: No, the database lists no Vulkan score for the GT 635M, and its API support does not include Vulkan. The GeForce 930M does have a Vulkan score of 3,729.

Q: How do these two compare to their nearest rivals?

A: The 930M is within 2.2% of its nearest rivals, with the GT 645M at -0.5%, the Intel Iris Pro Graphics 5200 at 0.7%, the RTX 4070 GDDR6 at 1.2%, and the FirePro W2100 at 2.2%. The GT 635M is within 1.4% of its nearest rivals, with the Quadro 3000M at 0.6%, the GT 740M at 0.6%, the GeForce 825M at 1.2%, and the Intel UHD Graphics 710 at -1.4%.

Q: Which GPU has more shading units?

A: The GeForce 930M has 384 shading units, while the GeForce GT 635M has 96 shading units.

Q: What is the memory bus width difference?

A: The GeForce GT 635M has a 128-bit memory bus, while the GeForce 930M has a 64-bit bus. The GT 635M also has higher memory bandwidth at 28.80 GB/s versus 12.80 GB/s.

Architecture Differences

The two GPUs come from different architectural generations and different manufacturing processes. The GeForce 930M is built on the Maxwell architecture, using the GM108S chip, fabricated on a 28 nm process at TSMC. The GeForce GT 635M uses the Fermi architecture, with the GF108 chip, on a 40 nm process, also at TSMC. The process node difference is significant: 28 nm versus 40 nm, which allows the 930M to pack more transistors into a smaller die.

The transistor counts illustrate this density difference. The 930M contains 1,020 million transistors on a die size of 77 mm², giving it a transistor density of 13.2 million transistors per square millimeter. The GT 635M contains 585 million transistors on a 116 mm² die, with a density of 5.0 million transistors per square millimeter. The 930M is both smaller and denser, a direct result of the newer process node.

The shading unit counts are drastically different. The 930M has 384 shading units, 24 texture mapping units, and 8 raster output units. The GT 635M has 96 shading units, 16 TMUs, and 4 ROPs. The 930M has four times the shading units, which explains its much higher FP32 throughput. The 930M also supports Vulkan 1.4, while the GT 635M has no Vulkan support listed. Both support DirectX 12 (11_0) and OpenGL 4.6.

Memory architecture differs as well. Both use DDR3 memory with 2 GB capacity, but the GT 635M uses a 128-bit bus with a bandwidth of 28.80 GB/s, while the 930M uses a 64-bit bus with a bandwidth of 12.80 GB/s. The GT 635M's memory clock is 900 MHz (1800 Mbps effective), while the 930M's is 800 MHz (1600 Mbps effective). Despite the GT 635M's wider bus and higher bandwidth, its compute cores are far fewer, so the bandwidth advantage does not overcome the compute deficit.

Specification Differences

The two parts differ in several key specification fields. The GeForce 930M has a base clock of 549 MHz and a boost clock of 549 MHz, while the GeForce GT 635M has no base or boost clock listed in the database. The memory clocks also differ: 800 MHz (1600 Mbps effective) for the 930M versus 900 MHz (1800 Mbps effective) for the GT 635M.

The memory bus width is a major difference, with the GT 635M at 128 bits and the 930M at 64 bits. This gives the GT 635M a bandwidth of 28.80 GB/s versus 12.80 GB/s for the 930M. The shading unit count is 384 versus 96, the TMU count is 24 versus 16, and the ROP count is 8 versus 4.

Compute rates differ sharply. The 930M has a pixel rate of 4.392 GPixel/s, a texture rate of 13.18 GTexel/s, and an FP32 performance of 421.6 GFLOPS. The GT 635M has a pixel rate of 1.900 GPixel/s, a texture rate of 7.600 GTexel/s, and an FP32 performance of 182.4 GFLOPS. The 930M is roughly 2.3 times faster in FP32, 2.3 times faster in pixel rate, and 1.7 times faster in texture rate.

The power draw is similar, with the 930M rated at 33 W and the GT 635M at 35 W, despite the 930M being significantly faster. The bus interface differs: PCIe 3.0 x8 for the 930M versus PCIe 2.0 x16 for the GT 635M. The release dates are also different, with the 930M released in March 2015 and the GT 635M in March 2012. The predecessor and successor fields differ as well: the 930M follows the GeForce 800M and precedes the GeForce 10 Mobile, while the GT 635M follows the GeForce 500M and precedes the GeForce 700M. The production status for both is end-of-life.