NVIDIA GeForce GTX 680M vs NVIDIA Quadro K620 Comparison

Head-to-Head Benchmarks

The database records one direct head-to-head benchmark between these two GPUs, and the result is not close. In Geekbench OpenCL, the NVIDIA GeForce GTX 680M scores 9230, while the NVIDIA Quadro K620 scores 6693. That is a 37.9% advantage for the GTX 680M, a substantial margin that frames the entire comparison.

Looking at average benchmark scores across the database, the GTX 680M sits at 7023, while the Quadro K620 sits at 6282. This places the GTX 680M in the 39th percentile of all GPUs, compared to the Quadro K620's 36th percentile. The GTX 680M is ahead, but the percentile gap is modest, which suggests the two cards occupy similar tiers in the broader GPU landscape despite the wide OpenCL delta.

The nearest rivals in the database put both cards in context. The GTX 680M's closest competitor is the NVIDIA T600, which averages 7035, a delta of only -0.2% relative to the GTX 680M. That means the GTX 680M is statistically tied with the T600 in aggregate performance. On the other side, the AMD Radeon R5 M240 averages 6975, which is 0.7% behind the GTX 680M, and the NVIDIA GeForce GTX 675M averages 6946, a 1.1% gap. The GTX 680M also trails the NVIDIA GeForce GTX 970 by 1.9%, as that card averages 7157.

The Quadro K620's nearest rivals tell a different story. The NVIDIA GeForce RTX 5070 Ti SUPER and the NVIDIA GeForce RTX 4070 Ti SUPER AD102 both average 6270, which is 0.2% behind the Quadro K620. The AMD Radeon R7 M350 averages 6327, a 0.7% lead over the Quadro K620, and the AMD Radeon Pro WX 4100 averages 6330, a 0.8% lead. The Quadro K620 is therefore bracketed by a group of cards that are all within roughly one percent of its average score, making it a very middle-of-the-pack performer in its immediate neighborhood.

The single head-to-head OpenCL result is the only direct measurement the database offers, but it aligns with the broader architectural picture. The GTX 680M's 37.9% OpenCL win is much larger than the 11.8% gap in average scores would suggest, which indicates that the GTX 680M's advantage is more pronounced in compute-oriented workloads than in the mixed set of benchmarks that feed the average.

Where Each One Wins

The GTX 680M wins the only recorded head-to-head benchmark, and it wins on average benchmark score, and it wins on percentile ranking. There is no benchmark in the database where the Quadro K620 comes out ahead. The GTX 680M also has a second benchmark in its profile, a Geekbench Metal score of 4815, which the Quadro K620 does not have a corresponding result for. The Quadro K620 does have a Geekbench Vulkan score of 5870, which the GTX 680M lacks, but without a shared test, those results cannot be compared directly.

For compute workloads measured by OpenCL, the GTX 680M is clearly the stronger part. Its 37.9% lead in that specific test is decisive. The Quadro K620's Vulkan result of 5870 suggests it has some competency in that API, but the database provides no GTX 680M Vulkan score to measure against.

The GTX 680M's higher average score, 7023 versus 6282, indicates that it holds the edge across the mixed benchmark pool. The Quadro K620's average is dragged down by its lower OpenCL result, and its Vulkan score of 5870 is also below the GTX 680M's OpenCL score of 9230, though cross-API comparisons are not directly meaningful.

In practical terms, the GTX 680M is the pick for any workload that leans on raw compute throughput. The Quadro K620 is not without merit, but the recorded data shows no scenario where it beats the GTX 680M in a shared test. The Quadro K620's single-slot form factor and lower power draw may make it attractive for constrained systems, but those are physical characteristics, not performance wins.

Architecture Differences

The two GPUs come from different NVIDIA architectures, and those differences explain much of the performance gap. The GTX 680M is built on the GK104 chip using the Kepler architecture, manufactured on a 28 nm process at TSMC. The Quadro K620 uses the GM107 chip with the Maxwell architecture, also on a 28 nm process at TSMC. Both are 28 nm parts, but the transistor counts diverge sharply.

The GTX 680M packs 3,540 million transistors on a 294 mm² die, giving it a transistor density of 12.0M per mm². The Quadro K620 has 1,870 million transistors on a 148 mm² die, with a density of 12.6M per mm². The GTX 680M has nearly twice the transistors and nearly twice the die area, which translates directly into a much larger compute footprint.

The shading unit counts reflect this gulf. The GTX 680M has 1344 shading units, 112 texture mapping units, and 32 ROPs. The Quadro K620 has 384 shading units, 24 TMUs, and 16 ROPs. That is a 3.5x advantage in shading units for the GTX 680M, a 4.7x advantage in TMUs, and a 2x advantage in ROPs. Pixel rate for the GTX 680M is 21.22 GPixel/s versus 17.98 GPixel/s for the Quadro K620, while texture rate is 84.90 GTexel/s versus 26.98 GTexel/s. The FP32 throughput is 2.038 TFLOPS for the GTX 680M versus 863.2 GFLOPS for the Quadro K620.

Clock speeds tell a different story. The Quadro K620 runs at a base clock of 1058 MHz and a boost clock of 1124 MHz, while the GTX 680M runs at 719 MHz base and 758 MHz boost. The Quadro K620's clocks are roughly 47% higher at base and 48% higher at boost, but the GTX 680M's massive shader count overwhelms that clock advantage in raw throughput.

Memory is another major divergence. The GTX 680M has 4 GB of GDDR5 on a 256 bit bus, delivering 115.2 GB/s of bandwidth. The Quadro K620 has 2 GB of DDR3 on a 128 bit bus, delivering only 28.80 GB/s. The GTX 680M has a 4x bandwidth advantage, and it doubles the memory capacity. Memory clock rates are 900 MHz for both cards, with the GTX 680M rated at 3.6 Gbps effective and the Quadro K620 at 1800 Mbps effective, but the bus width difference dominates.

Power and physical design also differ. The GTX 680M has a 100 W TDP and uses an MXM module form factor with an MXM-B (3.0) interface. The Quadro K620 has a 45 W TDP, is a single-slot card measuring 160 mm by 69 mm, uses a PCIe 2.0 x16 interface, and has a suggested PSU of 200 W. Neither card requires external power connectors. The Quadro K620's display outputs are 1x DVI and 1x DisplayPort 1.2, while the GTX 680M's outputs are listed as portable device dependent.

The generation labels are notable. The GTX 680M belongs to the GeForce 600M generation, while the Quadro K620 is listed under the Quadro Kepler (Kx200) generation, despite using the Maxwell architecture. The GTX 680M's predecessor is the GeForce 500M and its successor is the GeForce 700M. The Quadro K620's predecessor is the Quadro Fermi and its successor is the Quadro Maxwell.

API support is similar. Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 680M supports Vulkan 1.2.175, while the Quadro K620 supports Vulkan 1.4, a newer version. Neither card has ray tracing cores or tensor cores.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 680M has an average benchmark score of 7023, while the NVIDIA Quadro K620 has an average benchmark score of 6282.

Q: How big is the difference in the Geekbench OpenCL test?

A: The GTX 680M scores 9230 in Geekbench OpenCL, while the Quadro K620 scores 6693, giving the GTX 680M a 37.9% advantage.

Q: What are the memory specifications of each card?

A: The GTX 680M has 4 GB of GDDR5 on a 256 bit bus with 115.2 GB/s bandwidth. The Quadro K620 has 2 GB of DDR3 on a 128 bit bus with 28.80 GB/s bandwidth.

Q: Which card has more shading units?

A: The GTX 680M has 1344 shading units, while the Quadro K620 has 384 shading units.

Q: What is the power draw difference?

A: The GTX 680M has a TDP of 100 W, while the Quadro K620 has a TDP of 45 W.

Q: Do both cards support the same APIs?

A: Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 680M supports Vulkan 1.2.175, while the Quadro K620 supports Vulkan 1.4.

The Verdict

The data is unambiguous. The NVIDIA GeForce GTX 680M wins the only shared benchmark by 37.9%, has a higher average score by 741 points, and sits higher in the percentile rankings at 39 versus 36. Anyone choosing between these two for raw performance should take the GTX 680M. Its 3.5x shading unit advantage, 4x memory bandwidth advantage, and doubled memory capacity make it the stronger compute part across the board.

The Quadro K620 does have advantages in specific areas. It draws less than half the power at 45 W versus 100 W, runs in a compact single-slot form factor, and has a newer Vulkan API version at 1.4 versus 1.2.175. For a system with tight power or space constraints, the Quadro K620 is the more practical physical fit. Its higher clock speeds, 1058 MHz base and 1124 MHz boost, also show that Maxwell's architecture is more clock-efficient, but that efficiency does not overcome the GTX 680M's raw resource advantage.

The nearest rival data reinforces the GTX 680M's position. It trades blows with the NVIDIA T600 at -0.2% and beats the AMD Radeon R5 M240 by 0.7% and the NVIDIA GeForce GTX 675M by 1.1%. The Quadro K620 sits in a tighter cluster, with its nearest rivals all within one percent. Neither card is a top-tier performer, but the GTX 680M is the stronger of the two.

For compute-heavy workloads, OpenCL specifically, the GTX 680M is the clear choice. For a low-power, single-slot workstation card with newer Vulkan support, the Quadro K620 has a role. But on the basis of the recorded benchmarks, the GTX 680M is the faster GPU, and the Quadro K620 cannot claim a single measured win.