NVIDIA GeForce GTX 680M vs NVIDIA GeForce MX330 Comparison

Head-to-Head Benchmarks

The database records a single direct comparison between the NVIDIA GeForce MX330 and the NVIDIA GeForce GTX 680M: the Geekbench OpenCL test. In this measurement, the GTX 680M scores 9230, while the MX330 scores 7896. The delta of -14.5% indicates that the MX330 trails the GTX 680M by 14.5% in this compute-oriented workload. This is a meaningful gap, placing the older Kepler-based part clearly ahead in raw OpenCL throughput.

Looking at the broader benchmark averages, the picture becomes more complex. The MX330 holds an average benchmark score of 8458 across all recorded tests, while the GTX 680M averages 7023. This is a reversal of the single head-to-head result. The MX330's average is 20.4% higher than the GTX 680M's average, driven by its additional Vulkan score of 9019, a test the GTX 680M does not appear in the database for. The GTX 680M's only other recorded benchmark, Geekbench Metal, yields 4815, which drags its average down considerably. The MX330's two scores, 7896 and 9019, are both well above that Metal result.

When placed against their respective nearest rivals, both GPUs occupy similar tiers of the performance spectrum. The MX330 sits at the 43rd percentile of all GPUs in the database, with its nearest rival being the AMD Radeon HD 8870M at an average score of 8462, a delta of 0%. The GTX 680M sits at the 39th percentile, with its nearest rival being the NVIDIA T600 at 7035, a delta of -0.2%. Notably, the MX330's average score of 8458 is higher than the GTX 680M's average of 7023, yet the GTX 680M's single OpenCL win shows that raw compute performance favors the older chip in that specific test.

The delta values in the nearest rival lists are informative. The MX330 is within 0.3% of the AMD Radeon 880M (8436) and within 0.4% of the NVIDIA GeForce GTX 675MX (8427), while trailing the Intel Arc A380 (8558) by 1.2%. The GTX 680M is within 0.7% of the AMD Radeon R5 M240 (6975) and 1.1% of the NVIDIA GeForce GTX 675M (6946), while trailing the NVIDIA GeForce GTX 970 (7157) by 1.9%. Both GPUs are clustered tightly among mid-range peers, with their percentile rankings differing by only four points.

FAQ

Q: Which GPU wins the only direct head-to-head benchmark?

A: The NVIDIA GeForce GTX 680M wins the Geekbench OpenCL test with a score of 9230 versus the MX330's 7896, a delta of -14.5% for the MX330.

Q: How do their average benchmark scores compare?

A: The MX330 has an average benchmark score of 8458, while the GTX 680M averages 7023. This means the MX330's average is roughly 20% higher, despite losing the single head-to-head test.

Q: What benchmark scores does each GPU have in the database?

A: The MX330 has a Geekbench OpenCL score of 7896 and a Geekbench Vulkan score of 9019. The GTX 680M has a Geekbench OpenCL score of 9230 and a Geekbench Metal score of 4815.

Q: How do their percentile rankings compare?

A: The MX330 sits at the 43rd percentile of all GPUs, while the GTX 680M sits at the 39th percentile. The MX330 is four percentile points higher despite the OpenCL loss.

Q: Which rivals are closest to each GPU?

A: The MX330's closest rival is the AMD Radeon HD 8870M with an average score of 8462, a delta of 0%. The GTX 680M's closest rival is the NVIDIA T600 with an average score of 7035, a delta of -0.2%.

Q: Does the GTX 680M have any benchmark advantage besides OpenCL?

A: The database shows no other direct comparison. The GTX 680M's Metal score of 4815 is its only other recorded result, and it is lower than both of the MX330's scores.

The Verdict

The data presents a split decision. In the single recorded head-to-head test, Geekbench OpenCL, the NVIDIA GeForce GTX 680M is decisively ahead by 14.5%. For users prioritizing OpenCL compute performance, the GTX 680M is the clear choice based on this measurement. Its score of 9230 exceeds the MX330's 7896 by a substantial margin.

However, the average benchmark score tells a different story. The MX330 averages 8458 across its two tests, while the GTX 680M averages 7023 across its two tests. This 20.4% advantage for the MX330 is driven by the GTX 680M's weak Metal result of 4815, which is far below both of the MX330's scores. The MX330 also holds a higher percentile ranking, 43rd versus 39th, and its Vulkan score of 9019 is close to the GTX 680M's best OpenCL result.

Users who rely on Vulkan workloads should favor the MX330, as it records 9019 in that API while the GTX 680M has no Vulkan score in the database. Users who rely on Metal should note that the GTX 680M has a recorded Metal score of 4815, but the MX330 has no Metal result at all. The choice ultimately hinges on which API and workload matters most, as neither GPU dominates across all recorded metrics.

Specification Differences

The two GPUs differ in nearly every major specification category. The MX330 uses a GP108B chip built on a 14 nm process at Samsung, while the GTX 680M uses a GK104 chip built on a 28 nm process at TSMC. The MX330 has 1,800 million transistors on a 74 mm² die, giving a transistor density of 24.3 million per mm². The GTX 680M has 3,540 million transistors on a 294 mm² die, with a density of 12.0 million per mm². The MX330's die is far smaller and denser, reflecting its newer process node.

Clock speeds differ significantly. The MX330 has a base clock of 1531 MHz and a boost clock of 1594 MHz, while the GTX 680M has a base clock of 719 MHz and a boost clock of 758 MHz. The MX330 runs at more than double the clock speed. Memory clocks also differ: the MX330 runs at 1752 MHz with 7 Gbps effective, while the GTX 680M runs at 900 MHz with 3.6 Gbps effective.

Memory configuration is another major divergence. The MX330 has 2 GB of GDDR5 on a 64-bit bus, yielding 56.06 GB/s of bandwidth. The GTX 680M has 4 GB of GDDR5 on a 256-bit bus, yielding 115.2 GB/s of bandwidth. The GTX 680M has twice the memory capacity and more than double the bandwidth.

Compute resources are heavily in the GTX 680M's favor. The MX330 has 384 shading units, 24 texture mapping units, and 16 raster output units. The GTX 680M has 1344 shading units, 112 TMUs, and 32 ROPs. The GTX 680M has 3.5 times the shading units, 4.7 times the TMUs, and exactly twice the ROPs. Pixel rates are close, with the MX330 at 25.50 GPixel/s and the GTX 680M at 21.22 GPixel/s, but texture rates favor the GTX 680M at 84.90 GTexel/s versus 38.26 GTexel/s. FP32 performance also favors the GTX 680M at 2.038 TFLOPS versus 1,224.2 GFLOPS for the MX330. The MX330 has a recorded FP16 figure of 19.13 GFLOPS, while the GTX 680M has no FP16 measurement.

Power and form factor differences are stark. The MX330 has a TDP of 10 W and uses an IGP slot width with no power connectors. The GTX 680M has a TDP of 100 W and uses an MXM Module slot width, also with no power connectors. The MX330 connects via PCIe 3.0 x4, while the GTX 680M uses MXM-B (3.0). Both list portable device dependent display outputs. The MX330 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 680M supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6. Neither GPU has a launch MSRP recorded in the database.

Architecture Differences

The MX330 is built on the Pascal architecture, while the GTX 680M uses the Kepler architecture. This generational split explains many of the measured differences. Pascal is a newer design than Kepler, and the MX330 benefits from a 14 nm process node at Samsung, while the GTX 680M uses a 28 nm node at TSMC. The smaller node allows the MX330 to pack 24.3 million transistors per mm², compared to the GTX 680M's 12.0 million per mm², even though the GTX 680M has nearly twice the total transistor count.

The GTX 680M's GK104 chip is a large, high-transistor design (3,540 million on 294 mm²), typical of a high-end mobile part from 2012. The MX330's GP108B chip is a small, efficient design (1,800 million on 74 mm²), typical of an entry-level mobile part from 2020. The GTX 680M's much higher shading unit count (1344 versus 384) and TMU count (112 versus 24) reflect its role as a throughput-oriented chip, while the MX330's higher clocks (1531 MHz base versus 719 MHz base) reflect its efficiency-oriented design.

The memory architecture also reflects their different roles. The GTX 680M has a 256-bit bus with 115.2 GB/s of bandwidth, essential for feeding its large shader array. The MX330 has a 64-bit bus with 56.06 GB/s, sufficient for its smaller array. The GTX 680M has 4 GB of memory versus 2 GB on the MX330, a difference that matters for texture-heavy workloads or large datasets.

API support shows another generational gap. The MX330 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 680M supports DirectX 12 (11_0) and Vulkan 1.2.175. The MX330's higher Vulkan version suggests better support for modern graphics features. The MX330 also has a recorded FP16 capability of 19.13 GFLOPS, while the GTX 680M has none. The GTX 680M's predecessor is listed as GeForce 500M and its successor as GeForce 700M, while the MX330 has no predecessor or successor listed. Both are end-of-life products, with the MX330 released in 2020 and the GTX 680M in 2012.

Where Each One Wins

The NVIDIA GeForce GTX 680M wins in raw compute and memory bandwidth. Its OpenCL score of 9230 beats the MX330's 7896 by 14.5%, and its 115.2 GB/s of memory bandwidth is more than double the MX330's 56.06 GB/s. Its 1344 shading units and 112 TMUs give it a large advantage in texture-heavy workloads, and its FP32 throughput of 2.038 TFLOPS exceeds the MX330's 1,224.2 GFLOPS. Users running OpenCL compute tasks, particularly those that scale with shader count and memory bandwidth, should expect the GTX 680M to perform better.

The NVIDIA GeForce MX330 wins in efficiency and modern API support. Its TDP of 10 W is one-tenth of the GTX 680M's 100 W, making it suitable for thin-and-light systems where power draw is critical. Its higher clocks (1531 MHz base versus 719 MHz base) and smaller die (74 mm² versus 294 mm²) reflect a design focused on low power consumption. Its Vulkan 1.4 support is newer than the GTX 680M's Vulkan 1.2.175, and its DirectX 12 (12_1) support is more advanced than the GTX 680M's DirectX 12 (11_0). Its average benchmark score of 8458 is 20.4% higher than the GTX 680M's 7023, driven by its strong Vulkan result of 9019.

The MX330 also wins on pixel rate, at 25.50 GPixel/s versus 21.22 GPixel/s for the GTX 680M, despite having only half the ROPs. This is a direct result of its much higher clock speed. For fill-rate-bound workloads, the MX330 holds a small advantage. The MX330's percentile ranking of 43rd versus 39th for the GTX 680M further supports its overall standing in the database.

The GTX 680M's Metal score of 4815 is its weakest recorded result, and the MX330 has no Metal score, so comparisons in that API are impossible. The MX330's Vulkan score of 9019 is its strongest, while the GTX 680M has no Vulkan result. For users targeting Vulkan, the MX330 is the only option with data. For users targeting Metal, the GTX 680M has a recorded score, but it is lower than either of the MX330's results. The GTX 680M also has 4 GB of memory versus 2 GB, which may benefit workloads with larger memory footprints. Ultimately, the GTX 680M is the pick for compute-heavy OpenCL tasks and memory-intensive workloads, while the MX330 is the pick for efficiency, Vulkan workloads, and higher average performance across the recorded benchmarks.