NVIDIA GeForce 930M vs NVIDIA GeForce GTX 650 Comparison

Head-to-Head Benchmarks

The recorded data shows a split decision between these two NVIDIA parts, with each taking one of the two available benchmark tests. In Geekbench OpenCL, the NVIDIA GeForce 930M posts a score of 5046 against the NVIDIA GeForce GTX 650's 4545, giving the 930M an 11% advantage. That is a meaningful margin for a mobile-oriented chip, and it aligns with the 930M's higher average benchmark score of 4388 compared to the GTX 650's 3823. The OpenCL result is the 930M's strongest showing in the database, and it outpaces its own nearest rival, the NVIDIA GeForce GT 645M, by 0.5% while sitting 1.2% above the NVIDIA GeForce RTX 4070 GDDR6, an unusual comparison that places the 930M in an unexpected tier for compute workloads.

The Vulkan test flips the script. Here, the GTX 650 scores 4524, which is 17.6% higher than the 930M's 3729. That is a substantial reversal, and it suggests the older Kepler architecture retains an edge in API-specific workloads. The GTX 650's Vulkan result is notably higher than its OpenCL score of 4545, while the 930M drops sharply from 5046 to 3729 when moving from OpenCL to Vulkan. This 27.9% internal swing for the 930M is the largest performance variance recorded between the two tests for either card, and it highlights how driver and architecture optimization can skew results in different compute environments.

Looking at the broader database context, the 930M sits at the 26th percentile among all GPUs, while the GTX 650 is at the 22nd percentile. The 930M's average score of 4388 places it just 0.5% behind the GT 645M and 0.7% ahead of the Intel Iris Pro Graphics 5200, indicating a tightly clustered group of mid-range parts. The GTX 650's average of 3823 is 0.3% behind the NVIDIA GeForce MX110 and 1.5% behind the AMD Radeon R5 Graphics, showing it competes in a lower performance band despite its desktop pedigree. The head-to-head record is even at one win apiece, but the deltas are asymmetric: the 930M's OpenCL win is a moderate 11%, while the GTX 650's Vulkan win is a more decisive 17.6%.

Where Each One Wins

The NVIDIA GeForce 930M is the clear choice for OpenCL-based compute tasks. Its 5046 score beats not only the GTX 650 but also every one of its own nearest rivals, including the GT 645M, the Iris Pro 5200, the RTX 4070 GDDR6, and the AMD FirePro W2100. This suggests that Maxwell's architecture, with its higher transistor density of 13.2 million per square millimeter versus the GTX 650's 11.5 million, translates into efficient raw compute throughput for general-purpose GPU workloads. The 930M also carries a smaller die at 77 mm² with 1,020 million transistors, which allows for better thermal headroom in a 33 W package, potentially sustaining higher clocks under load in OpenCL scenarios.

The NVIDIA GeForce GTX 650 dominates in Vulkan, where its 4524 score is 17.6% ahead of the 930M. This is a significant gap, and it aligns with the GTX 650's higher absolute hardware specifications: 812.5 GFLOPS of FP32 performance, 33.86 GTexel/s texture rate, and 8.464 GPixel/s pixel rate. These figures are roughly double the 930M's 421.6 GFLOPS, 13.18 GTexel/s, and 4.392 GPixel/s, respectively. The GTX 650 also benefits from a 128-bit memory bus with 80 GB/s bandwidth, compared to the 930M's 64-bit bus at 12.8 GB/s. That six-fold bandwidth advantage is likely the primary driver of its Vulkan success, as Vulkan workloads often stress memory throughput more heavily than OpenCL.

For real-world usage splits, the data suggests the 930M is better suited for lightweight compute or productivity tasks that leverage OpenCL, while the GTX 650 is preferable for any application or game that uses Vulkan as its rendering API. The GTX 650's desktop form factor, with a single-slot design and a 6-pin power connector, also implies it was designed for sustained performance in a fixed chassis, whereas the 930M is an integrated-class part for portable devices, with no power connectors and a 33 W TDP. The GTX 650's 65 W TDP and suggested PSU of 250 W further indicate a system-level commitment to performance, whereas the 930M can run in a notebook without external power.

Architecture Differences

The two GPUs come from different NVIDIA generations. The 930M uses the GM108S chip based on Maxwell architecture, manufactured on a 28 nm TSMC process, and belongs to the GeForce 900M series. The GTX 650 uses the GK106 chip based on Kepler architecture, also on 28 nm TSMC, but from the GeForce 600 series. The transistor counts differ substantially: the 930M packs 1,020 million transistors into a 77 mm² die, while the GTX 650 has 2,540 million transistors on a 221 mm² die. This makes the 930M denser at 13.2 million transistors per mm² versus the GTX 650's 11.5 million, but the GTX 650 has more than twice the absolute transistor budget.

Memory configurations are the starkest architectural contrast. The 930M uses 2 GB of DDR3 on a 64-bit bus, yielding 12.8 GB/s bandwidth. The GTX 650 uses 1024 MB of GDDR5 on a 128-bit bus, yielding 80 GB/s bandwidth. That is a 6.25x difference in memory throughput, which explains why the GTX 650 excels in bandwidth-sensitive workloads like Vulkan. The GTX 650 also has twice the ROPs (16 versus 8) and more TMUs (32 versus 24), while both share the same shading unit count at 384. The pixel rate and texture rate differences follow: the GTX 650 delivers 8.464 GPixel/s and 33.86 GTexel/s, versus the 930M's 4.392 GPixel/s and 13.18 GTexel/s.

Clock behavior also differs. The 930M has a fixed base and boost clock of 549 MHz, with no variation. The GTX 650's clock fields are null in the database, meaning no base or boost figures are recorded; only the memory clock is listed at 1250 MHz (5 Gbps effective). The 930M's memory clock is 800 MHz (1600 Mbps effective), which is lower in absolute terms but also reflects the slower DDR3 standard. Both GPUs support DirectX 12 (11_0) and OpenGL 4.6, but the 930M supports Vulkan 1.4 while the GTX 650 is limited to Vulkan 1.2.175, a version difference that may contribute to the 930M's weaker Vulkan showing if driver support for newer Vulkan features is less optimized on Maxwell.

The bus interface also differs: the 930M uses PCIe 3.0 x8, while the GTX 650 uses PCIe 3.0 x16. The GTX 650's wider interface allows more host communication bandwidth, which can matter for data transfer in compute workloads. The 930M is classified as an IGP (integrated graphics package) with no slot width specification, while the GTX 650 is a single-slot add-in card measuring 147 mm (5.8 inches) in length. The GTX 650's display outputs include 1x DVI and 2x DisplayPort 1.2, while the 930M's outputs are listed as "Portable Device Dependent," reflecting its mobile target.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 930M has an average benchmark score of 4388, while the NVIDIA GeForce GTX 650 averages 3823. The 930M also ranks at the 26th percentile among all GPUs, compared to the GTX 650's 22nd percentile.

Q: How do the two GPUs compare in OpenCL performance?

A: The 930M scores 5046 in Geekbench OpenCL, which is 11% higher than the GTX 650's 4545. The 930M's OpenCL score is also its best recorded benchmark result, while the GTX 650's OpenCL score is nearly identical to its Vulkan score.

Q: What is the difference in Vulkan performance?

A: The GTX 650 scores 4524 in Geekbench Vulkan, which is 17.6% higher than the 930M's 3729. This is the GTX 650's strongest test result, and it represents a significant margin over the 930M's Vulkan showing.

Q: Which GPU has more memory bandwidth?

A: The GTX 650 has 80 GB/s bandwidth from its 128-bit GDDR5 memory, while the 930M has 12.8 GB/s from its 64-bit DDR3 memory. This is a 6.25x difference in favor of the GTX 650.

Q: What are the TDP requirements for each card?

A: The 930M has a TDP of 33 W and requires no power connectors, while the GTX 650 has a TDP of 65 W and uses a single 6-pin power connector. The GTX 650 also has a suggested PSU rating of 250 W.

Q: Which GPU has the higher transistor density?

A: The 930M has a transistor density of 13.2 million per mm², while the GTX 650 has 11.5 million per mm². The 930M achieves this with 1,020 million transistors on a 77 mm² die, whereas the GTX 650 uses 2,540 million transistors on a 221 mm² die.

The Verdict

The benchmark data paints a clear picture for different use cases. If the primary workload is OpenCL-based compute, the NVIDIA GeForce 930M is the superior choice. Its 5046 OpenCL score beats the GTX 650 by 11%, and its average score of 4388 is higher across the board. The 930M also benefits from a more efficient design: a smaller die, higher transistor density, and a 33 W TDP that allows deployment in power-constrained systems without external power. For users running productivity tools, scientific applications, or any OpenCL-accelerated software, the 930M delivers better measured performance.

If the workload involves Vulkan rendering or any modern graphics API that leverages the Vulkan path, the NVIDIA GeForce GTX 650 is the definitive winner. Its 4524 Vulkan score is 17.6% ahead of the 930M, and its hardware specifications support this: 812.5 GFLOPS, 80 GB/s bandwidth, and 16 ROPs. The GTX 650's 128-bit memory bus is the decisive factor, providing over six times the bandwidth of the 930M, which is critical for Vulkan's memory-intensive nature. Desktop users with a 250 W PSU and a single-slot slot can comfortably accommodate the GTX 650's 65 W draw and 6-pin connector.

For users who need a mobile or integrated solution, the 930M is the only option between the two, as the GTX 650 is a desktop card with a 147 mm length and dual DisplayPort outputs. The 930M's portable-device-dependent outputs and IGP classification make it suitable for notebooks, while the GTX 650 requires a chassis with expansion space. The database also shows that the 930M's nearest rivals include the GT 645M and Iris Pro 5200, with score differences under 1%, meaning it competes in a tight mobile segment. The GTX 650's nearest rivals, such as the MX110 and Radeon R5 Graphics, show it sits in a slightly lower performance band overall, despite its Vulkan strength.

The final recommendation depends on the API environment. For general compute and OpenCL, choose the 930M. For Vulkan-specific tasks and higher raw throughput, choose the GTX 650. The tie in head-to-head wins reflects this balance, but the 17.6% Vulkan margin is more decisive than the 11% OpenCL margin, suggesting the GTX 650 has the larger performance swing in its favor. Neither card is a comprehensive winner, and the choice hinges entirely on the software stack being used.