NVIDIA GeForce 830M vs NVIDIA GeForce GT 640 Comparison

Head-to-Head Benchmarks

The recorded benchmark data presents a split decision between the NVIDIA GeForce 830M and the NVIDIA GeForce GT 640, with each card claiming one victory in the two available tests. In Geekbench OpenCL, the 830M posts a score of 4324 against the GT 640's 3736, a delta of 15.7% in favor of the newer mobile part. This is a substantial margin, placing the 830M clearly ahead in compute workloads that leverage OpenCL acceleration. The GT 640, by contrast, counters in Geekbench Vulkan with a score of 3809 versus the 830M's 3590, a 5.7% advantage. The Vulkan result is narrower, indicating that while the GT 640 holds the edge in this API, the gap is far less pronounced than the OpenCL lead enjoyed by the 830M.

Looking at the broader average benchmark score, the 830M reaches 3957, while the GT 640 trails at 3210. This aggregate figure, however, includes the GT 640's Geekbench Metal score of 2086, which drags its average down; the 830M has no recorded Metal result. When isolating the shared tests, OpenCL and Vulkan, the 830M's average across those two is 3957, while the GT 640's average across its OpenCL and Vulkan runs is 3772.5. The head-to-head deltas, therefore, tell the cleaner story: a decisive OpenCL win for the 830M and a modest Vulkan win for the GT 640. The percentile placement reinforces this split, as the 830M ranks at the 24th percentile among all GPUs, while the GT 640 sits lower at the 20th percentile, suggesting that the 830M's overall standing in the database is slightly stronger despite the Vulkan reversal.

Where Each One Wins

The GeForce 830M demonstrates its advantage in OpenCL-based workloads, which typically include general-purpose compute tasks such as video encoding, image processing, and scientific calculations that offload to the GPU. Its 15.7% lead in this test suggests that the Maxwell architecture, with its higher base clock of 1082 MHz and boost up to 1150 MHz, provides a measurable compute throughput benefit. The 830M's FP32 rating of 588.8 GFLOPS, while lower than the GT 640's 692.7 GFLOPS, does not prevent it from winning in this particular benchmark, indicating that architectural efficiency in Maxwell may compensate for raw shader count. The 830M also carries a higher pixel rate at 9.200 GPixel/s versus the GT 640's 7.216 GPixel/s, which could translate to faster fill-rate-bound operations.

The GeForce GT 640, conversely, asserts its superiority in Vulkan, a modern low-overhead API that benefits from the card's larger memory bus and higher bandwidth. With a 128-bit bus and 28.51 GB/s of bandwidth, the GT 640 more than doubles the 830M's 64-bit bus and 14.40 GB/s. This memory throughput advantage likely explains its Vulkan win, as Vulkan workloads often stress memory access patterns. The GT 640 also has more shading units (384 versus 256) and more texture mapping units (32 versus 16), which can aid in geometry-heavy or texture-heavy scenes. In the database's nearest rival comparisons, the 830M sits near the AMD Radeon R5 M420, with a delta of 0%, and the NVIDIA Quadro K2000, trailing by 0.2%, showing it competes with mid-range mobile and workstation parts. The GT 640, meanwhile, is close to the Intel Arc Pro B60, leading by 0.9%, and the NVIDIA Quadro P1000, ahead by 1.5%, placing it in a slightly lower performance tier overall.

Architecture Differences

The two GPUs come from different architectural generations, and the data reflects that divide. The GeForce 830M uses the GM108 chip built on the Maxwell architecture, manufactured by TSMC on a 28 nm process node. It packs 1,020 million transistors into a die size of 77 mm², yielding a transistor density of 13.2 million per square millimeter. The GeForce GT 640, in contrast, employs the GK107 chip from the Kepler architecture, also on TSMC's 28 nm node, but with a larger die of 118 mm² and 1,270 million transistors, resulting in a lower density of 10.8 million per square millimeter. The Maxwell chip is therefore more compact and denser, which aligns with its lower TDP of 33 W compared to the GT 640's 65 W.

The memory subsystem differs markedly. The 830M uses 2 GB of DDR3 on a 64-bit bus, delivering 14.40 GB/s of bandwidth and a memory clock of 900 MHz (1800 Mbps effective). The GT 640 also has 2 GB of DDR3, but on a 128-bit bus, achieving 28.51 GB/s with a memory clock of 891 MHz (1782 Mbps effective). This doubles the bandwidth for the GT 640, a critical factor in many workloads. The core configurations also diverge: the 830M has 256 shading units, 16 TMUs, and 8 ROPs, while the GT 640 has 384 shading units, 32 TMUs, and 16 ROPs. The GT 640's higher texture rate of 28.86 GTexel/s versus 18.40 GTexel/s for the 830M underscores its advantage in texture-heavy rendering, though the 830M's pixel rate of 9.200 GPixel/s exceeds the GT 640's 7.216 GPixel/s.

Feature support shows some differences as well. Both cards support DirectX 12 (11_0) and OpenGL 4.6, but the Vulkan implementation differs: the 830M supports Vulkan 1.4, while the GT 640 is limited to Vulkan 1.2.175. The 830M is an integrated-style part with an IGP slot width and no power connectors, designed for portable devices, while the GT 640 is a single-slot add-in card with a 145 mm length, one DVI, one HDMI 1.4a, and one DisplayPort 1.2 output. The GT 640 also lists a suggested PSU of 250 W, whereas the 830M has no such recommendation. Both are end-of-life products, with the 830M released in 2014 as part of the GeForce 800M generation, succeeding the 700M series, and the GT 640 released in 2012 in the GeForce 600 generation, succeeding the 500 series.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 830M leads with an average benchmark score of 3957, while the NVIDIA GeForce GT 640 averages 3210.

Q: What is the performance difference in OpenCL between the two cards?

A: In Geekbench OpenCL, the 830M scores 4324 against the GT 640's 3736, giving the 830M a 15.7% advantage.

Q: Does the GT 640 win any benchmarks?

A: Yes, the GT 640 wins Geekbench Vulkan with a score of 3809 versus the 830M's 3590, a 5.7% lead.

Q: How do their memory bandwidths compare?

A: The GT 640 offers 28.51 GB/s over a 128-bit bus, while the 830M provides 14.40 GB/s over a 64-bit bus, so the GT 640 has roughly double the bandwidth.

Q: Are both GPUs end-of-life products?

A: Yes, both the GeForce 830M and the GeForce GT 640 are listed as end-of-life in the database.

Q: Which card has more shading units?

A: The GT 640 has 384 shading units, while the 830M has 256, a difference of 128 units in favor of the GT 640.

Specification Differences

The recorded specifications show clear divergence across several key fields. The 830M uses the GM108 chip from the Maxwell architecture, while the GT 640 uses the GK107 from Kepler. The transistor counts differ: 1,020 million for the 830M versus 1,270 million for the GT 640. Die size also varies, with the 830M at 77 mm² and the GT 640 at 118 mm², leading to transistor densities of 13.2M per mm² and 10.8M per mm², respectively. Base and boost clocks are only listed for the 830M, at 1082 MHz and 1150 MHz, whereas the GT 640 has no recorded base or boost clock. Memory clocks show 900 MHz (1800 Mbps effective) for the 830M and 891 MHz (1782 Mbps effective) for the GT 640.

Memory bus width is a major differentiator: the 830M has a 64-bit bus, the GT 640 a 128-bit bus. Bandwidth follows suit at 14.40 GB/s versus 28.51 GB/s. Core counts differ, with the 830M featuring 256 shading units, 16 TMUs, and 8 ROPs, while the GT 640 features 384 shading units, 32 TMUs, and 16 ROPs. Pixel rates are 9.200 GPixel/s for the 830M and 7.216 GPixel/s for the GT 640, while texture rates are 18.40 GTexel/s and 28.86 GTexel/s, respectively. FP32 performance is 588.8 GFLOPS for the 830M and 692.7 GFLOPS for the GT 640. TDP values are 33 W for the 830M and 65 W for the GT 640, with slot widths of IGP versus single-slot.

Power connectors are absent on both, but the GT 640 lists a suggested PSU of 250 W, while the 830M has none. The bus interface is PCIe 3.0 x8 for the 830M and PCIe 3.0 x16 for the GT 640. Display outputs are portable-device dependent for the 830M, while the GT 640 offers one DVI, one HDMI 1.4a, and one DisplayPort 1.2. Vulkan support differs, with the 830M at version 1.4 and the GT 640 at version 1.2.175. The GT 640 has a recorded length of 145 mm (5.7 inches) and a launch MSRP of 99 USD, while the 830M has no length or MSRP data. Release dates are 2014-03-11 for the 830M and 2012-06-04 for the GT 640. Predecessors and successors also differ: the 830M follows the GeForce 700M and precedes the GeForce 900M, while the GT 640 follows the GeForce 500 and precedes the GeForce 700.

The Verdict

The data supports a nuanced choice. For users prioritizing OpenCL compute performance, the GeForce 830M is the better option, posting a 15.7% higher score and a superior average benchmark of 3957, which places it at the 24th percentile among all GPUs. Its higher pixel rate and lower TDP (33 W versus 65 W) make it suitable for power-constrained portable systems, and its modern Vulkan 1.4 support ensures better API compatibility. The GT 640, however, wins in Vulkan by 5.7%, and its doubled memory bandwidth (28.51 GB/s versus 14.40 GB/s) plus higher shading unit count (384 versus 256) give it an edge in memory-bound or texture-heavy scenarios. Its OpenCL score is lower, and its average benchmark of 3210 sits at the 20th percentile, reflecting weaker overall standing.

The GT 640 also offers a physical card form factor with discrete display outputs, a 250 W suggested PSU, and a longer track record, having launched in 2012. The 830M, from 2014, has no discrete outputs and is designed for integration. For a desktop user with a PCIe x16 slot and a need for Vulkan performance, the GT 640 is the data-backed selection. For a mobile user or anyone prioritizing OpenCL workloads and lower power consumption, the 830M is the stronger pick. The split wins in head-to-head tests mean the decision hinges on the specific workload: OpenCL favors the 830M, Vulkan favors the GT 640, and no single card dominates across both.