NVIDIA GeForce GT 640 vs NVIDIA GeForce MX110 Comparison

Head-to-Head Benchmarks

The two benchmark results in the database split cleanly, with each GPU taking one win. In the Geekbench OpenCL test, the NVIDIA GeForce MX110 scores 4255 against the GeForce GT 640's 3736, a 13.9% advantage. That is a decisive margin in compute workloads that scale with raw shader throughput. The MX110's Maxwell architecture delivers higher per-clock efficiency, and its FP32 rating of 515.1 GFLOPS, while lower than the GT 640's 692.7 GFLOPS on paper, translates to a faster OpenCL result in practice.

The GeForce GT 640 strikes back in the Geekbench Vulkan test. It scores 3809 versus the MX110's 3413, a 10.4% lead. Vulkan workloads often reward memory bandwidth and the GT 640's 128-bit bus, which provides 28.51 GB/s of bandwidth. The MX110's 64-bit bus, despite using faster GDDR5 memory at 40.10 GB/s, cannot overcome the GT 640's architectural advantages in this specific API. The GT 640 also has more shading units (384 versus 256) and double the texture mapping units (32 versus 16), which helps in certain draw-call-heavy scenarios.

Looking at the average benchmark scores, the MX110 holds the overall edge. Its average of 3834 across all recorded tests compares favorably to the GT 640's 3210. That 624-point gap is substantial, representing roughly a 19.4% difference in favor of the MX110. However, the GT 640's average is dragged down by its Metal benchmark score of 2086, a test that only appears in its record. The MX110 does not have a Metal score in the database, so the comparison of averages should be weighted toward the shared OpenCL and Vulkan results.

The percentile rankings reinforce the MX110's position. The MX110 sits at the 23rd percentile of all GPUs, while the GT 640 sits at the 20th percentile. Both are entry-level parts, but the MX110 is measurably closer to mid-range performance. Its nearest rivals include the NVIDIA GeForce GTX 650 (0.3% higher average score), the Intel UHD Graphics 710 (1.1% higher), the AMD Radeon R5 Graphics (1.2% lower), and the NVIDIA Quadro 2000 (1.6% lower). The GT 640's nearest rivals cluster similarly: the Intel Arc Pro B60 (0.9% higher), the NVIDIA Quadro P1000 (1.5% higher), the NVIDIA GeForce 920M (2.3% lower), and the NVIDIA GeForce GT 730M (3.2% lower). Both GPUs sit in a tight competitive band, but the MX110's rivals are slightly closer in raw score, indicating less variance in its performance tier.

Where Each One Wins

The MX110 wins in compute-oriented OpenCL workloads. Its 13.9% lead in that test suggests it handles general-purpose GPU tasks, such as physics simulations, video encoding helpers, and OpenCL-accelerated filters, better than the GT 640. The Maxwell architecture's improved instruction scheduling and lower overhead per thread group likely contribute to this. The MX110 also achieves a higher pixel rate of 8.048 GPixel/s versus the GT 640's 7.216 GPixel/s, which means it can fill the screen with rendered pixels faster in some scenarios, despite having half the ROPs (8 versus 16). This counterintuitive result comes from the Maxwell generation's more efficient ROP design.

The GT 640 wins in Vulkan workloads. Its 10.4% lead in the Vulkan test points to better driver optimization for modern graphics APIs on the Kepler architecture, or at least a better match between its hardware layout and Vulkan's command buffer model. The GT 640's texture rate of 28.86 GTexel/s versus the MX110's 16.10 GTexel/s is a massive advantage in texture-heavy scenes, and its 128-bit memory bus provides more parallel memory channels. For older DirectX 11 games that rely heavily on texture sampling, the GT 640 likely pulls ahead. Its display outputs (1x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2) also make it a more flexible option for multi-monitor setups, whereas the MX110's outputs are listed as "Portable Device Dependent," meaning it is designed for laptops and small form factor systems.

The MX110's power efficiency is another key differentiator. Its TDP of 30 W is less than half of the GT 640's 65 W, and it requires no power connectors, just like the GT 640. The MX110 is an IGP (integrated graphics processor) soldered to the motherboard, while the GT 640 is a single-slot add-in card measuring 145 mm (5.7 inches) in length. The MX110's lower power draw makes it suitable for thin-and-light laptops, while the GT 640's higher power budget and physical expansion slot make it a desktop-only part.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce MX110, with an average of 3834 across all tests, versus 3210 for the GeForce GT 640. That is a 624-point difference in favor of the MX110.

Q: Does the GeForce GT 640 win any benchmark against the MX110?

A: Yes, the GT 640 wins the Geekbench Vulkan test with a score of 3809 versus 3413 for the MX110, a 10.4% advantage.

Q: How much faster is the MX110 in OpenCL?

A: The MX110 scores 4255 in Geekbench OpenCL, which is 13.9% higher than the GT 640's 3736.

Q: Which GPU has more shading units?

A: The GeForce GT 640 has 384 shading units, while the MX110 has 256. The GT 640 also has 32 TMUs versus 16, and 16 ROPs versus 8.

Q: What is the memory configuration difference?

A: Both have 2 GB of memory, but the MX110 uses GDDR5 on a 64-bit bus with 40.10 GB/s bandwidth, while the GT 640 uses DDR3 on a 128-bit bus with 28.51 GB/s bandwidth.

Q: Which GPU is more power-efficient?

A: The MX110 has a TDP of 30 W, while the GT 640 has a TDP of 65 W. The MX110 uses less than half the power.

Specification Differences

The two GPUs differ across nearly every core specification. The MX110 uses the GM108S chip with 1,020 million transistors on a 77 mm² die, while the GT 640 uses the GK107 chip with 1,270 million transistors on a 118 mm² die. The transistor density favors the MX110 at 13.2M per mm² versus 10.8M per mm², reflecting the newer manufacturing process maturity. Both are built on TSMC's 28 nm node, but the MX110's denser layout points to design improvements.

Clock speeds differ significantly. The MX110 has a base clock of 978 MHz and a boost clock of 1006 MHz. The GT 640 has no listed base or boost clock in the database. Memory clocks also differ: the MX110 runs at 1253 MHz (5 Gbps effective), while the GT 640 runs at 891 MHz (1782 Mbps effective). The MX110's memory is GDDR5, while the GT 640 uses the slower DDR3 type.

Compute rates show a mixed picture. The MX110 has a pixel rate of 8.048 GPixel/s, higher than the GT 640's 7.216 GPixel/s. However, the GT 640's texture rate of 28.86 GTexel/s is nearly double the MX110's 16.10 GTexel/s. FP32 throughput favors the GT 640 at 692.7 GFLOPS versus 515.1 GFLOPS.

The bus interface differs: the MX110 uses PCIe 3.0 x4, while the GT 640 uses PCIe 3.0 x16. This affects maximum data transfer rates between the GPU and the rest of the system. The GT 640 has a suggested PSU of 250 W, while the MX110 has no suggested PSU listed. Display outputs also differ: the GT 640 offers 1x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the MX110's outputs are portable-device dependent.

The GT 640 has a launch MSRP of 99 USD. The MX110's launch MSRP is not listed in the database.

Architecture Differences

The MX110 is built on the Maxwell architecture, while the GT 640 uses the older Kepler architecture. This generational gap explains many of the performance differences. Maxwell introduced improved color compression, more efficient scheduling, and better power management compared to Kepler. The MX110's 28 nm process node is the same as the GT 640's, but the Maxwell design squeezes more performance per transistor.

The chip designs are fundamentally different. The MX110's GM108S is a small, efficient chip designed for mobile and low-power applications. The GT 640's GK107 is a larger desktop-oriented chip with more raw resources. The GT 640 has 384 shading units, 32 TMUs, and 16 ROPs, versus the MX110's 256 shading units, 16 TMUs, and 8 ROPs. Despite having half the resources, the MX110 nearly matches the GT 640 in OpenCL and beats it in pixel rate, which speaks to Maxwell's architectural efficiency.

Memory subsystems diverge sharply. The MX110 uses a 64-bit bus with GDDR5, which achieves 40.10 GB/s of bandwidth. The GT 640 uses a 128-bit bus with DDR3, which achieves only 28.51 GB/s. The MX110's narrower but faster memory interface is better suited to its compact design. The GT 640's wider bus, paired with slower memory, provides more parallel access but lower overall throughput.

API support differs in Vulkan version. The MX110 supports Vulkan 1.4, while the GT 640 supports Vulkan 1.2.175. Both support DirectX 12 (11_0) and OpenGL 4.6. The MX110's newer Vulkan version likely accounts for part of its competitive positioning, though the GT 640 still wins the actual Vulkan benchmark.

Physical design reflects their intended use cases. The MX110 is an IGP (integrated graphics processor) with no slot width, no power connectors, and portable-device-dependent outputs. The GT 640 is a single-slot desktop card with a 145 mm length and standard display outputs. The production status for both is end-of-life.

The Verdict

The data points to the NVIDIA GeForce MX110 as the stronger all-around GPU for most users. Its higher average benchmark score (3834 versus 3210), better OpenCL performance (13.9% ahead), higher pixel rate, and dramatically lower power consumption (30 W versus 65 W) make it the more practical choice. The MX110's 23rd percentile ranking versus the GT 640's 20th percentile confirms that it sits slightly higher in the overall performance hierarchy.

Choose the MX110 if you need efficient compute performance in a laptop or compact system. Its OpenCL lead and lower power draw make it suitable for portable devices where battery life and thermals matter. The Maxwell architecture's efficiency means it punches above its weight class, as evidenced by beating the GT 640 in pixel rate despite having half the ROPs.

Choose the GeForce GT 640 if you primarily run Vulkan-based applications or older games that favor texture throughput. Its 10.4% Vulkan win and double the texture rate (28.86 GTexel/s versus 16.10 GTexel/s) give it an edge in specific scenarios. The GT 640's 128-bit memory bus and 384 shading units also make it more capable in workloads that scale with raw resource counts. Its desktop form factor with standard display outputs is better suited for older systems or multi-monitor setups.

The GT 640's higher FP32 throughput (692.7 GFLOPS versus 515.1 GFLOPS) suggests it should win compute tasks, but the benchmark data shows the MX110 winning OpenCL anyway. This is a case where architecture efficiency trumps raw specs. The MX110's newer Maxwell design extracts more usable performance from fewer resources. For most practical purposes, the MX110 is the better GPU. The GT 640 remains relevant only for users who specifically need its Vulkan strength or require a PCIe x16 desktop card with dedicated display outputs.