NVIDIA GeForce GTX 660 vs NVIDIA Quadro M2000M Comparison

The NVIDIA Quadro M2000M and NVIDIA GeForce GTX 660 are both end-of-life GPUs built on a 28 nm TSMC process, but they target entirely different segments: one is a professional mobile workstation part, the other a desktop gaming card from an earlier generation. Benchmark data shows the GTX 660 leads in raw compute tests, yet the M2000M holds its own in specific professional workloads. Below is a breakdown of their architectures, specifications, and benchmark performance, based strictly on the provided data.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Quadro M2000M has an average benchmark score of 9832, while the GTX 660 scores 9022. This places the M2000M in the 47th percentile of all GPUs, versus the 45th percentile for the GTX 660.

Q: How do their memory configurations differ?

A: The M2000M comes with 4 GB of GDDR5 on a 128-bit bus, yielding 80.19 GB/s bandwidth. The GTX 660 has 2 GB of GDDR5 on a 192-bit bus, delivering 144.2 GB/s. The GTX 660’s wider bus provides significantly more memory bandwidth despite less capacity.

Q: Which card wins in OpenCL and Vulkan benchmarks?

A: The GTX 660 wins both head-to-head tests. It scores 11347 in OpenCL versus 10057 for the M2000M (an 11.4% advantage), and 11415 in Vulkan versus 9606 (a 15.8% advantage). The M2000M has no Vulkan score in its own benchmark list, but the head-to-head data confirms the GTX 660’s lead.

Q: What are the power requirements for each card?

A: The Quadro M2000M has a TDP of 55 W and uses an MXM module with no power connectors. The GTX 660 has a TDP of 140 W, requires a dual-slot cooler with a 1x 6-pin power connector, and a suggested 300 W power supply.

Q: Which card supports a newer version of Vulkan?

A: The M2000M supports Vulkan 1.4, while the GTX 660 supports Vulkan 1.2.175. Both cards support DirectX 12 (11_0) and OpenGL 4.6.

Q: What is the launch MSRP of the GTX 660?

A: The GTX 660 had a launch MSRP of 229 USD. The M2000M has no launch MSRP listed in the data.

Architecture Differences

The two GPUs come from different NVIDIA architectures. The Quadro M2000M is based on the GM107 chip (Maxwell architecture) from the Quadro Maxwell-M generation, while the GTX 660 uses the GK106 chip (Kepler architecture) from the GeForce 600 series. Both are fabricated on a 28 nm process at TSMC, but the transistor counts differ: the M2000M packs 1,870 million transistors on a 148 mm² die (12.6M / mm² density), while the GTX 660 contains 2,540 million transistors on a 221 mm² die (11.5M / mm² density). The GTX 660’s larger die gives it more raw hardware: 960 shading units, 80 TMUs, and 24 ROPs, versus the M2000M’s 640 shading units, 40 TMUs, and 16 ROPs. Neither card features ray tracing or tensor cores.

Clock speeds also differ. The M2000M runs at a base of 1098 MHz and boost of 1137 MHz, with memory at 1253 MHz (5 Gbps effective). The GTX 660 has a lower base clock of 980 MHz and boost of 1032 MHz, but its memory runs at 1502 MHz (6 Gbps effective). The M2000M’s Maxwell architecture is newer, but the GTX 660 compensates with more execution units and faster memory. The M2000M uses an MXM-A (3.0) bus interface, while the GTX 660 uses PCIe 3.0 x16. Display outputs are portable-device-dependent for the M2000M, whereas the GTX 660 offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2.

The Verdict

The data points to a clear split: the GTX 660 is the stronger performer in compute benchmarks, winning both head-to-head tests with margins of 11.4% in OpenCL and 15.8% in Vulkan. Its average benchmark score, however, is lower than the M2000M’s (9022 vs 9832), which suggests the M2000M benefits from a more consistent performance profile across other workloads not listed here. The M2000M also offers double the VRAM (4 GB vs 2 GB) and a much lower TDP (55 W vs 140 W), making it more suitable for mobile workstations where power and space are constrained. The GTX 660, with its higher bandwidth (144.2 GB/s vs 80.19 GB/s) and more shading units, is better suited for tasks that scale with raw throughput, such as gaming or GPU compute that doesn’t require large memory pools. Pick the M2000M if you need professional-grade reliability, low power draw, and more memory; choose the GTX 660 if raw compute speed and bandwidth are your priority, and you can tolerate a desktop card with a higher power draw.

Specification Differences

The two cards diverge on nearly every major specification beyond the process node. The M2000M uses the GM107 chip with 1,870 million transistors on a 148 mm² die, while the GTX 660 uses the GK106 chip with 2,540 million transistors on a 221 mm² die. Transistor density favors the M2000M (12.6M / mm² vs 11.5M / mm²). Clock speeds: the M2000M has a base of 1098 MHz and boost of 1137 MHz; the GTX 660 has a base of 980 MHz and boost of 1032 MHz. Memory clocks: 1253 MHz (5 Gbps effective) for the M2000M, 1502 MHz (6 Gbps effective) for the GTX 660. Memory capacity: 4 GB vs 2 GB. Bus width: 128-bit vs 192-bit. Bandwidth: 80.19 GB/s vs 144.2 GB/s. Shading units: 640 vs 960. TMUs: 40 vs 80. ROPs: 16 vs 24. Pixel rate: 18.19 GPixel/s vs 20.64 GPixel/s. Texture rate: 45.48 GTexel/s vs 82.56 GTexel/s. FP32 performance: 1,455.4 GFLOPS vs 1.981 TFLOPS. TDP: 55 W vs 140 W. Slot width: MXM Module vs Dual-slot. Power connectors: None vs 1x 6-pin. Suggested PSU: not listed vs 300 W. Bus interface: MXM-A (3.0) vs PCIe 3.0 x16. Display outputs: Portable Device Dependent vs 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2. Vulkan version: 1.4 vs 1.2.175. Release dates: 2015-12-02 vs 2012-09-05. The GTX 660 also has a listed length of 241 mm (9.5 inches), while the M2000M has no dimensions listed.

Head-to-Head Benchmarks

The head-to-head data includes two tests, and the GTX 660 wins both. In Geekbench OpenCL, the GTX 660 scores 11347 against the M2000M’s 10057, a delta of -11.4% from the M2000M’s perspective. This means the GTX 660 outperforms the M2000M by over 11% in this compute workload. In Geekbench Vulkan, the gap widens: the GTX 660 scores 11415, while the M2000M scores 9606, a delta of -15.8%. The GTX 660’s lead in Vulkan is nearly 19% relative to the M2000M’s score, indicating a stronger advantage in this API. Notably, the M2000M’s own benchmark list includes an OpenCL score of 10057 and a Vulkan score of 9606, matching the head-to-head figures. The GTX 660 also has a Geekbench Metal score of 4305, which has no counterpart in the M2000M’s data. The M2000M wins zero head-to-head tests, while the GTX 660 wins two.

Where Each One Wins

The GTX 660 is the clear winner in compute benchmarks. It excels in OpenCL, where its 11347 score outpaces the M2000M by 1290 points, and in Vulkan, where its 11415 score beats the M2000M by 1809 points. This makes it the better choice for applications that rely on general-purpose GPU compute, such as rendering, physics simulation, or any workload that can leverage OpenCL or Vulkan. Its higher texture rate (82.56 GTexel/s vs 45.48 GTexel/s) and pixel rate (20.64 GPixel/s vs 18.19 GPixel/s) further support its advantage in graphics-intensive tasks. Additionally, the GTX 660’s 144.2 GB/s memory bandwidth is nearly double the M2000M’s 80.19 GB/s, which benefits tasks that stream large amounts of data.

The M2000M wins on efficiency and capacity. Its 55 W TDP is less than half the GTX 660’s 140 W, making it far more suitable for laptops or compact MXM-based systems. Its 4 GB VRAM is double the GTX 660’s 2 GB, which is critical for workloads that need to hold large datasets in memory, such as professional 3D modeling, CAD, or machine learning inference. The M2000M also has a higher average benchmark score (9832 vs 9022) and a better percentile ranking (47th vs 45th), suggesting that in non-head-to-head tests, it performs more consistently across a broader range of applications. Its newer Maxwell architecture (released in December 2015 vs September 2012) provides Vulkan 1.4 support, a more recent API version than the GTX 660’s Vulkan 1.2.175. For a mobile workstation where power draw and memory capacity are paramount, the M2000M is the practical pick; for a desktop system where raw compute speed is the sole metric, the GTX 660 dominates.