NVIDIA GeForce MX130 vs NVIDIA Quadro K620 Comparison

The NVIDIA Quadro K620 and the NVIDIA GeForce MX130 are both end-of-life mobile or compact desktop parts built around NVIDIA’s Maxwell architecture, but they target very different workloads. The Quadro K620 is a professional workstation card, while the MX130 is an entry-level notebook GPU. Benchmark data from the database shows a clear split: the Quadro K620 wins both recorded head-to-head tests, but the margins and the context behind those wins matter for real-world use cases. The K620 leads in OpenCL and Vulkan compute, while the MX130 offers a higher texture rate and a more modern memory interface in a lower-power package.

Where Each One Wins

The Quadro K620 is the stronger choice for compute-centric professional tasks. In the database’s head-to-head comparison, the K620 wins both available benchmarks. Its OpenCL score of 6693 beats the MX130’s 6102 by 9.7%, and its Vulkan score of 5870 beats the MX130’s 4914 by a larger 19.5%. These are not small gaps. The Vulkan result in particular shows that the K620 has a meaningful advantage in modern GPU compute APIs, which are often used in rendering, simulation, and data-processing workloads. The K620 also sits at the 36th percentile among all GPUs in the database, while the MX130 sits at the 32nd percentile, meaning the K620 is ranked slightly higher overall.

The MX130, by contrast, is not the winner in any recorded benchmark. Its average benchmark score is 5508, compared to the K620’s 6282. That is a substantial difference in aggregate performance. However, the MX130 does have a few raw specification advantages that do not show up in the two scored tests. It has a higher base clock (1109 MHz versus 1058 MHz) and a higher boost clock (1189 MHz versus 1124 MHz). It also has a higher texture rate: 28.54 GTexel/s versus 26.98 GTexel/s for the K620. In texture-heavy workloads, such as certain older games or image-processing pipelines that rely on TMUs, the MX130 could edge ahead despite losing the compute benchmarks. The MX130 also uses GDDR5 memory with a much higher effective speed of 5 Gbps, compared to the K620’s DDR3 at 1800 Mbps effective.

For users who prioritize compute performance, the Quadro K620 is the clear pick. For users who need a lower-power, integrated-style GPU with a faster memory clock and higher texture throughput, the MX130 has its niche. The K620 wins on compute, the MX130 wins on memory bandwidth per watt and texture rate, though the raw bandwidth figure favors the MX130: 40.10 GB/s versus 28.80 GB/s.

Architecture Differences

Both GPUs are built on the Maxwell architecture and fabricated by TSMC on a 28 nm process, but they use different chips. The Quadro K620 uses the GM107 chip, while the MX130 uses the GM108S chip. The K620’s die is 148 mm² with 1,870 million transistors, giving a transistor density of 12.6 million per mm². The MX130’s die is much smaller at 77 mm² with 1,020 million transistors, yielding a higher density of 13.2 million per mm². This means the MX130 packs slightly more transistors per area, but the K620 has nearly double the die area and almost double the transistor count.

The shading unit count is identical: both GPUs have 384 shading units and 24 texture mapping units. The key difference lies in the render output units (ROPs). The K620 has 16 ROPs, while the MX130 has only 8. This directly explains the K620’s much higher pixel rate: 17.98 GPixel/s versus 9.512 GPixel/s for the MX130. In any workload that is pixel-bound, such as rasterization or certain post-processing effects, the K620 has a clear hardware advantage. The MX130’s lower ROP count is a major architectural limitation.

Memory architecture also differs significantly. The K620 uses a 128-bit memory bus with DDR3 memory, while the MX130 uses a 64-bit bus with GDDR5 memory. The narrower bus on the MX130 is offset by the faster GDDR5 standard, resulting in higher bandwidth: 40.10 GB/s for the MX130 versus 28.80 GB/s for the K620. However, the K620’s wider bus reduces latency concerns in some access patterns. The MX130 also has a different bus interface: PCIe 3.0 x4, whereas the K620 uses PCIe 2.0 x16. For compute workloads that stream data from system memory, the K620’s wider PCIe link is likely more capable, though the MX130’s newer PCIe generation provides higher per-lane bandwidth.

Both GPUs expose the same API feature set: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. The K620 has no tensor cores or ray tracing cores, and the MX130 also lacks these. The K620 is a single-slot card with 1x DVI and 1x DisplayPort 1.2 outputs, while the MX130 is an integrated GPU (IGP) with portable-device-dependent outputs. This means the K620 is designed for fixed workstation setups, while the MX130 is meant for laptops where display outputs vary by laptop model.

Head-to-Head Benchmarks

The database records two head-to-head benchmark results between these GPUs, and the Quadro K620 wins both. In the OpenCL test, the K620 scores 6693 against the MX130’s 6102, a delta of 9.7%. This is a solid but not overwhelming lead. OpenCL is commonly used in professional applications like video editing, CAD, and scientific computing, so the K620’s advantage here aligns with its workstation positioning. The MX130’s score of 6102 is not far behind, suggesting that for simple OpenCL workloads, the two GPUs are comparable, but the K620 pulls ahead under sustained compute load.

The Vulkan test shows a much larger gap. The K620 scores 5870, while the MX130 scores 4914, a delta of 19.5%. Vulkan is a lower-overhead API that scales well with core count and memory bandwidth. The K620’s higher ROP count and wider memory bus likely contribute to this advantage. A 19.5% lead is significant in real-world terms; it means the K620 can complete Vulkan-based compute tasks nearly a fifth faster than the MX130. This could matter for applications that use Vulkan for GPGPU tasks, such as certain game engines or renderers.

The average benchmark scores reinforce the K620’s overall lead. The K620 averages 6282 across all recorded benchmarks, while the MX130 averages 5508. That is a 14% difference in aggregate performance. The K620’s nearest rivals in the database include the NVIDIA GeForce RTX 5070 Ti SUPER and RTX 4070 Ti SUPER AD102, both scoring 6270 with a delta of 0.2%, essentially tied with the K620. On the other side, the AMD Radeon R7 M350 scores 6327 (0.7% ahead) and the AMD Radeon Pro WX 4100 scores 6330 (0.8% ahead). The MX130’s rivals include the NVIDIA GeForce GTX 765M at 5501 (0.1% ahead), the AMD Radeon R7 M440 at 5483 (0.5% behind), the AMD FirePro M4000 at 5537 (0.5% ahead), and the NVIDIA Quadro M4000 at 5467 (0.7% behind). These rival comparisons show that the K620 competes at a higher performance tier than the MX130.

Specification Differences

The two GPUs differ in several key specification fields. The most important is the memory subsystem. The K620 has 2 GB of DDR3 memory on a 128-bit bus, delivering 28.80 GB/s of bandwidth. The MX130 also has 2 GB of memory, but it is GDDR5 on a 64-bit bus, delivering 40.10 GB/s. Despite the narrower bus, the MX130 has higher bandwidth due to the faster memory type.

Clock speeds favor the MX130. The MX130 has a base clock of 1109 MHz and a boost clock of 1189 MHz, both higher than the K620’s 1058 MHz base and 1124 MHz boost. The memory clock also differs: the K620 runs at 900 MHz (1800 Mbps effective), while the MX130 runs at 1253 MHz (5 Gbps effective). The MX130’s memory clock is more than twice as fast in effective terms.

The ROP count is a major differentiator. The K620 has 16 ROPs, the MX130 has 8. This leads to a pixel rate of 17.98 GPixel/s for the K620 versus 9.512 GPixel/s for the MX130. The texture rate slightly favors the MX130: 28.54 GTexel/s versus 26.98 GTexel/s. FP32 compute is close: the K620 delivers 863.2 GFLOPS, the MX130 delivers 913.2 GFLOPS, a small win for the MX130.

Power consumption favors the MX130. The K620 has a TDP of 45 W and a suggested PSU of 200 W, while the MX130 has a TDP of 30 W and no suggested PSU (it is an IGP). The K620 is a single-slot card with its own display outputs, while the MX130 is integrated with portable-device-dependent outputs. The bus interface differs as well: the K620 uses PCIe 2.0 x16, the MX130 uses PCIe 3.0 x4. The K620 has a length of 160 mm (6.3 inches) and height of 69 mm (2.7 inches), while the MX130 has no recorded dimensions. The K620 uses the GM107 chip, the MX130 uses GM108S. The K620 has 1,870 million transistors on a 148 mm² die, the MX130 has 1,020 million on 77 mm². Release dates also differ: the K620 launched on 2014-07-21, the MX130 on 2017-11-16.

FAQ

Q: Which GPU has higher raw compute performance in FP32?

A: The MX130 has a slightly higher FP32 rate of 913.2 GFLOPS compared to the K620’s 863.2 GFLOPS. However, the K620 wins the recorded OpenCL and Vulkan benchmarks, indicating that real-world compute performance depends on more than just peak FP32 throughput.

Q: Why does the Quadro K620 win the Vulkan benchmark by such a large margin?

A: The K620 scores 5870 in Vulkan versus 4914 for the MX130, a 19.5% lead. The K620 has 16 ROPs versus 8 on the MX130, and a 128-bit memory bus versus 64-bit, which likely contribute to better Vulkan compute behavior despite the MX130’s higher nominal bandwidth.

Q: Does the MX130 have any specification advantage over the K620?

A: Yes, the MX130 has higher base and boost clocks (1109 MHz and 1189 MHz versus 1058 MHz and 1124 MHz), higher memory bandwidth (40.10 GB/s versus 28.80 GB/s), higher texture rate (28.54 GTexel/s versus 26.98 GTexel/s), and a lower TDP (30 W versus 45 W).

Q: Are both GPUs from the same architecture generation?

A: Both use the Maxwell architecture and are fabricated by TSMC on a 28 nm process. The K620 uses the GM107 chip, while the MX130 uses the GM108S chip. Both support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4.

Q: What is the memory configuration difference?

A: The K620 has 2 GB of DDR3 on a 128-bit bus with 28.80 GB/s bandwidth. The MX130 has 2 GB of GDDR5 on a 64-bit bus with 40.10 GB/s bandwidth. The MX130 has higher bandwidth despite the narrower bus.

Q: Which GPU has a higher overall benchmark average?

A: The K620 has an average benchmark score of 6282, while the MX130 averages 5508. The K620 sits at the 36th percentile among all GPUs, compared to the 32nd percentile for the MX130.