NVIDIA GeForce 840M vs NVIDIA GeForce GTX 560M Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 840M has an average benchmark score of 5322, while the NVIDIA GeForce GTX 560M scores 4855. The 840M leads by 18.7% in the Geekbench OpenCL test.

Q: What architecture does each chip use?

A: The GeForce 840M uses the Maxwell architecture with the GM108S chip, while the GeForce GTX 560M uses Fermi 2.0 with the GF116 chip.

Q: Which GPU has more shading units?

A: The GeForce 840M has 384 shading units, compared to 192 on the GeForce GTX 560M. This is a 2x difference in raw compute lanes.

Q: What is the memory configuration difference?

A: The GeForce 840M has 2 GB DDR3 memory on a 64-bit bus with 16.02 GB/s bandwidth. The GeForce GTX 560M has 1536 MB GDDR5 on a 192-bit bus with 60.00 GB/s bandwidth.

Q: How do their transistor counts compare?

A: The GeForce 840M has 1,020 million transistors on a 77 mm² die, while the GeForce GTX 560M has 1,170 million transistors on a 238 mm² die. The 840M packs more transistors per area.

Q: Which GPU supports Vulkan?

A: The GeForce 840M supports Vulkan 1.4, while the GeForce GTX 560M does not list Vulkan support. Both support DirectX 12 (11_0) and OpenGL 4.6.

Architecture Differences

The GeForce 840M is built on Maxwell, using the GM108S chip fabricated at 28 nm by TSMC. It uses 1,020 million transistors across a 77 mm² die, giving a transistor density of 13.2M per mm². This is a modern, power-efficient design with a 33 W TDP and an integrated (IGP) slot form factor.

The GeForce GTX 560M uses the older Fermi 2.0 architecture with the GF116 chip, fabricated at 40 nm. It packs 1,170 million transistors on a much larger 238 mm² die, resulting in a density of just 4.9M per mm². The TDP is 75 W, more than double the 840M, and it uses an MXM Module slot.

The 840M has 384 shading units, 16 texture mapping units, and 8 ROPs. The GTX 560M has 192 shading units, 32 TMUs, and 24 ROPs. The 840M has more shader cores but fewer texture units and ROPs, reflecting their different design priorities.

Memory configuration is a major architectural split. The 840M uses 2 GB DDR3 on a 64-bit bus with 16.02 GB/s bandwidth. The GTX 560M uses 1536 MB GDDR5 on a 192-bit bus with 60.00 GB/s bandwidth, giving it nearly 4x the memory bandwidth. This matters for texture-heavy workloads and higher resolutions.

The 840M supports PCIe 3.0 x8, while the GTX 560M uses an MXM-B (3.0) interface. The 840M has a base clock of 1029 MHz with a boost of 1124 MHz and memory at 1001 MHz (2 Gbps effective). The GTX 560M has no base or boost clock listed, only memory at 625 MHz (2.5 Gbps effective).

The 840M is part of the GeForce 800M generation, succeeding GeForce 700M, and is followed by GeForce 900M. The GTX 560M belongs to the GeForce 500M generation, succeeding GeForce 400M, and is followed by GeForce 600M. Both are end-of-life products.

Head-to-Head Benchmarks

The database contains one head-to-head benchmark result between these two GPUs. In Geekbench OpenCL, the GeForce 840M scores 5764 while the GeForce GTX 560M scores 4855. The 840M wins by 18.7%, a substantial margin in raw compute throughput.

This result aligns with the 840M's 863.2 GFLOPS FP32 performance versus the GTX 560M's 595.2 GFLOPS. The 840M has 45% more floating-point capability, which the OpenCL score reflects. The 840M's higher shader count (384 vs 192) and higher clocks (1029 MHz base) drive this advantage.

The GTX 560M does have wins in specific metrics. Its texture rate is 24.80 GTexel/s versus 17.98 GTexel/s on the 840M, a 38% advantage. Its pixel rate is 6.200 GPixel/s versus 8.992 GPixel/s on the 840M, so the 840M actually leads there by 45%. The 560M's 32 TMUs and 24 ROPs help texture and pixel work, but not compute.

The GTX 560M's memory bandwidth is 60.00 GB/s versus 16.02 GB/s on the 840M. That is a 3.7x bandwidth advantage, which historically helps in fill-rate-bound scenarios and high-resolution texture streaming. However, the recorded Geekbench OpenCL test does not show this translating into a benchmark win.

The 840M also has a Geekbench Vulkan score of 4880, a test the GTX 560M cannot run since it lacks Vulkan support. This gives the 840M an additional measured capability.

In the nearest rival context, the 840M's average score of 5322 sits just 0.1% above the GeForce 930A (5317) and 0.7% above the GeForce 940M (5284). It is 0.3% above the GeForce GTX 980M (5308) and 0.7% below the AMD Radeon R7 M445 (5358). The GTX 560M's average score of 4855 is 0.2% above the GeForce 940MX (4844), 0.8% below the GeForce GTS 450 (4893), and 0.9% below the GeForce RTX 5060 Ti 8 GB (4901).

The Verdict

The data shows a clear overall winner in the GeForce 840M. It wins the only head-to-head benchmark by 18.7%, has a higher average benchmark score (5322 vs 4855), and sits at the 31st percentile versus the 28th for the GTX 560M. For compute-oriented tasks like OpenCL workloads, the 840M is measurably faster.

The GTX 560M is not without merit. It has 60.00 GB/s of memory bandwidth versus 16.02 GB/s, a 3.7x advantage. It also has more TMUs (32 vs 16) and ROPs (24 vs 8), which historically benefit texture filtering and pixel fill. But the recorded benchmark data does not show these advantages producing a win.

For a buyer choosing between these two end-of-life mobile GPUs today, the 840M is the better choice for general compute and modern API support. It offers Vulkan, higher FP32 throughput, and lower power draw (33 W vs 75 W). The GTX 560M may still be preferable in niche scenarios where memory bandwidth is the bottleneck, but the benchmark evidence favors the 840M.

Specification Differences

The two GPUs differ across nearly every major specification.

Process and Die: The 840M uses 28 nm with a 77 mm² die and 1,020 million transistors. The GTX 560M uses 40 nm with a 238 mm² die and 1,170 million transistors. Transistor density is 13.2M / mm² for the 840M versus 4.9M / mm² for the 560M.

Clocks: The 840M has a base clock of 1029 MHz and boost of 1124 MHz. The GTX 560M has no base or boost clock listed. Memory clock is 1001 MHz (2 Gbps effective) on the 840M versus 625 MHz (2.5 Gbps effective) on the 560M.

Memory: The 840M has 2 GB DDR3 on a 64-bit bus with 16.02 GB/s bandwidth. The GTX 560M has 1536 MB GDDR5 on a 192-bit bus with 60.00 GB/s bandwidth.

Compute Units: The 840M has 384 shading units, 16 TMUs, and 8 ROPs. The GTX 560M has 192 shading units, 32 TMUs, and 24 ROPs.

Rates: Pixel rate is 8.992 GPixel/s on the 840M versus 6.200 GPixel/s on the 560M. Texture rate is 17.98 GTexel/s on the 840M versus 24.80 GTexel/s on the 560M. FP32 is 863.2 GFLOPS on the 840M versus 595.2 GFLOPS on the 560M.

Power and Form: TDP is 33 W for the 840M versus 75 W for the 560M. The 840M uses an IGP slot, the 560M uses an MXM Module. The 840M has PCIe 3.0 x8, the 560M has MXM-B (3.0).

APIs: Both support DirectX 12 (11_0) and OpenGL 4.6. The 840M supports Vulkan 1.5, the 560M has no Vulkan support.

Generation and Release: The 840M released in 2011-05-29, the 560M released in 2014-03-11. Neither has a launch MSRP.

Where Each One Wins

The 840M wins in compute-heavy workloads. It leads the Geekbench OpenCL test by 18.7%, has higher FP32 throughput (863.2 GFLOPS vs 595.2 GFLOPS, 45% more), and higher pixel rate (8.992 GPixel/s vs 6.200 GPixel/s). It also has Vulkan support and a measured Vulkan score of 4880, giving it access to modern API workloads the 560M cannot run.

The GTX 560M wins in memory-bandwidth-bound tasks. It has 60.00 GB/s versus 60.00 GB/s, a 3.7x advantage. It also has higher texture rate (24.80 GTexel/s vs 17.98 GTexel/s, 38% more) and more TMUs (32 vs 16). Scenarios that stress texture streaming or high-resolution filtering would favor the 560M on paper.

The 840M wins on efficiency metrics. At 33 W TDP versus 75 W, it draws less power for a higher average score. The 840M has a 77 mm² die, the smallest, with a density of 13.2M / mm². The 840M also has more shading units at 384 versus 192, double the shader count.

The GTX 560M wins on pure memory capacity per bus width. It has 192-bit bus width versus 64-bit, and more memory bandwidth per GB/s at 60.00 GB/s. The 560M has 32 TMUs at 24.80 GTexel/s, a higher texture rate than the 840M.

For users weighting modern API support and compute, the 840M is the clear pick. For users who need maximum memory bandwidth and texture throughput in legacy DX11-era workloads, the 560M has measurable advantages. The percentile ranks confirm this: 31st percentile for the 840M, 28th for the 560M. The benchmark data supports the 840M as the better all-around performer.