NVIDIA GeForce 840M vs NVIDIA GeForce GTX 550 Ti Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 550 Ti has an average benchmark score of 5731, while the NVIDIA GeForce 840M has an average score of 5322. This puts the GTX 550 Ti in the 33rd percentile of all GPUs, ahead of the 840M's 31st percentile.

Q: How do the two compare in the Geekbench OpenCL test?

A: The GeForce 840M wins the single recorded OpenCL head-to-head with a score of 5764 versus 5731 for the GTX 550 Ti, a margin of 0.6%. This is the only direct benchmark comparison in the database, and the 840M takes it by a narrow edge.

Q: What are the power requirements for each card?

A: The GTX 550 Ti has a TDP of 116 W, requires a 300 W suggested power supply, uses a dual-slot cooler, and needs one 6-pin power connector. The GeForce 840M has a TDP of 33 W, uses no power connectors, and is an IGP (integrated graphics processor) with no slot width specified.

Q: Which GPU has more shading units?

A: The GeForce 840M has 384 shading units, which is exactly double the 192 shading units found in the GTX 550 Ti. However, the GTX 550 Ti has more texture mapping units (32 versus 16) and more render output units (24 versus 8).

Q: What memory configurations do these cards use?

A: The GTX 550 Ti uses 1024 MB of GDDR5 memory on a 192-bit bus with a bandwidth of 98.50 GB/s. The GeForce 840M uses 2 GB of DDR3 memory on a 64-bit bus with a bandwidth of only 16.02 GB/s.

Q: Are both GPUs still in production?

A: No, both are end-of-life products. The GTX 550 Ti was released in March 2011 and belongs to the GeForce 500 generation, while the GeForce 840M was released in March 2014 and belongs to the GeForce 800M generation.

Architecture Differences

The two GPUs come from different architectural generations and manufacturing processes, which explains most of their behavioral differences. The GTX 550 Ti is built on the Fermi 2.0 architecture using the GF116 chip, manufactured on a 40 nm process at TSMC. In contrast, the GeForce 840M uses the Maxwell architecture with the GM108S chip, produced on a 28 nm process, also at TSMC. The process shrink from 40 nm to 28 nm is significant: the 840M packs 1,020 million transistors into a 77 mm² die, while the GTX 550 Ti uses 1,170 million transistors across a much larger 238 mm² die. This results in a transistor density of 13.2 million per square millimeter for the 840M versus 4.9 million per square millimeter for the GTX 550 Ti, indicating a far more efficient design in the newer chip.

The memory subsystems are radically different. The GTX 550 Ti uses 1024 MB of GDDR5 memory on a 192-bit bus, delivering 98.50 GB/s of bandwidth. The GeForce 840M uses 2 GB of DDR3 memory on a 64-bit bus, delivering only 16.02 GB/s. This is a six-fold difference in memory bandwidth, heavily favoring the older card. Even though the 840M has double the memory capacity, the narrow bus and slower DDR3 memory severely limit its data throughput.

Compute resources are distributed differently as well. The 840M has 384 shading units versus 192 on the GTX 550 Ti, but the GTX 550 Ti has 32 TMUs versus 16, and 24 ROPs versus 8. The pixel rate is higher on the 840M at 8.992 GPixel/s versus 7.200 GPixel/s, but the texture rate is much higher on the GTX 550 Ti at 28.80 GTexel/s versus 17.98 GTexel/s. FP32 performance also favors the 840M at 863.2 GFLOPS versus 691.2 GFLOPS.

Feature support differs in the API stack. Both support DirectX 12 (11_0) and OpenGL 4.6, but the 840M adds Vulkan 1.4 support while the GTX 550 Ti has no recorded Vulkan support. The bus interface also differs: the GTX 550 Ti uses PCIe 2.0 x16, while the 840M uses PCIe 3.0 x8. The 840M is an IGP with display outputs described as "Portable Device Dependent," while the GTX 550 Ti offers 2x DVI and 1x mini-HDMI 1.3a outputs.

The Verdict

The data presents a nuanced picture. In the only direct head-to-head benchmark recorded, the Geekbench OpenCL test, the GeForce 840M wins by a slim 0.6% margin. However, the average benchmark score across all tests tells a different story: the GTX 550 Ti averages 5731, which is 7.7% higher than the 840M's 5322 average. The 840M's average is dragged down by its additional Vulkan test score of 4880, which is far below its OpenCL result.

For desktop users with a power supply and slot space, the GTX 550 Ti is the stronger overall performer based on the average benchmark data. It also offers vastly superior memory bandwidth (98.50 GB/s versus 16.02 GB/s), which matters for texture-heavy workloads and higher resolutions. Its 192-bit bus and GDDR5 memory are simply in a different class than the 840M's 64-bit DDR3 setup.

For laptop or portable-device users, the GeForce 840M is the obvious choice based on power characteristics alone. Its 33 W TDP versus 116 W, no power connectors, and IGP form factor make it suitable for mobile systems where the GTX 550 Ti's dual-slot cooler and 6-pin connector would be impractical. The 840M also has double the memory capacity and higher FP32 throughput, plus Vulkan support.

The recorded data suggests that the GTX 550 Ti is the better all-around card for raw performance, while the 840M is the better fit for constrained environments. Neither card is a clear winner in every metric; the choice depends on the use case and platform constraints.

Specification Differences

| Specification | NVIDIA GeForce GTX 550 Ti | NVIDIA GeForce 840M |

|---|---|---|

| Architecture | Fermi 2.0 | Maxwell |

| Chip | GF116 | GM108S |

| Process Node | 40 nm | 28 nm |

| Transistors | 1,170 million | 1,020 million |

| Die Size | 238 mm² | 77 mm² |

| Transistor Density | 4.9M / mm² | 13.2M / mm² |

| Base Clock | Not specified | 1029 MHz |

| Boost Clock | Not specified | 1124 MHz |

| Memory Clock | 1026 MHz (4.1 Gbps effective) | 1001 MHz (2 Gbps effective) |

| Memory Size | 1024 MB | 2 GB |

| Memory Type | GDDR5 | DDR3 |

| Memory Bus Width | 192 bit | 64 bit |

| Memory Bandwidth | 98.50 GB/s | 16.02 GB/s |

| Shading Units | 192 | 384 |

| TMUs | 32 | 16 |

| ROPs | 24 | 8 |

| Pixel Rate | 7.200 GPixel/s | 8.992 GPixel/s |

| Texture Rate | 28.80 GTexel/s | 17.98 GTexel/s |

| FP32 | 691.2 GFLOPS | 863.2 GFLOPS |

| TDP | 116 W | 33 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 6-pin | None |

| Suggested PSU | 300 W | Not specified |

| Bus Interface | PCIe 2.0 x16 | PCIe 3.0 x8 |

| Display Outputs | 2x DVI, 1x mini-HDMI 1.3a | Portable Device Dependent |

| Vulkan Support | Not specified | 1.4 |

| Release Date | 2011-03-14 | 2014-03-11 |

| Launch MSRP | 149 USD | Not specified |

Head-to-Head Benchmarks

The only direct benchmark comparison in the database is the Geekbench OpenCL test. Here, the GeForce 840M scores 5764 against the GTX 550 Ti's 5731, a difference of just 0.6% in favor of the 840M. This is a very tight margin, essentially a statistical tie in practical terms. The 840M's win in this test likely reflects its higher shading unit count and higher FP32 throughput.

Looking at the broader benchmark picture, the GTX 550 Ti's average score of 5731 places it slightly above its nearest rivals: the GTX 670MX (5721, 0.2% behind), Intel Iris Pro Graphics P6300 (5712, 0.3% behind), and AMD Radeon HD 8790M (5691, 0.7% behind). It trails the AMD Radeon R7 M465 (5841) by 1.9%. These are all very close competitors, with the GTX 550 Ti sitting in a tightly clustered performance band.

The 840M's average score of 5322 is lower, but its nearest rivals are similarly close: the GeForce 930A (5317, 0.1% behind), GeForce GTX 980M (5308, 0.3% behind), and GeForce 940M (5284, 0.7% behind) all sit within a fraction of a percent. The AMD Radeon R7 M445 (5358) is 0.7% ahead. The 840M's OpenCL score of 5764 is notably higher than its average, meaning the Vulkan score of 4880 substantially pulls down its overall ranking.

The most striking observation is that the 840M's OpenCL score exceeds not only the GTX 550 Ti's but also sits above the GTX 550 Ti's entire nearest-rival cluster. Yet the 840M's average performance is far lower because of the Vulkan result. This suggests that the 840M is competitive in OpenCL workloads but significantly weaker in Vulkan-based tasks.

Where Each One Wins

The GTX 550 Ti wins in memory bandwidth and texture throughput. Its 98.50 GB/s of bandwidth is more than six times the 840M's 16.02 GB/s, and its texture rate of 28.80 GTexel/s exceeds the 840M's 17.98 GTexel/s by 60%. For games and applications that are bandwidth-limited or texture-heavy, the GTX 550 Ti has a clear advantage. Its 24 ROPs versus 8 also helps with fill-rate-bound scenarios at lower resolutions.

The GeForce 840M wins in compute throughput and efficiency. It delivers 863.2 GFLOPS of FP32 performance versus 691.2 GFLOPS, a 25% advantage. Its pixel rate of 8.992 GPixel/s is also higher than the GTX 550 Ti's 7.200 GPixel/s. The 840M's 33 W TDP is a fraction of the GTX 550 Ti's 116 W, making it vastly more power-efficient. It also has double the memory capacity (2 GB versus 1 GB) and supports Vulkan 1.4, which the GTX 550 Ti does not.

For desktop gaming with high-resolution textures, the GTX 550 Ti is the pick. The combination of GDDR5 memory, a 192-bit bus, and higher texture rate makes it better suited for traditional gaming workloads where memory bandwidth is frequently the bottleneck.

For mobile systems and compute-oriented tasks, the GeForce 840M is the pick. Its IGP form factor, lack of power connectors, and low TDP make it feasible for laptops and compact devices. Its higher shading unit count and FP32 throughput also favor general-purpose compute workloads, particularly in OpenCL-based applications.

For Vulkan-based workloads, the 840M is the only option with recorded support. The GTX 550 Ti has no Vulkan support listed in the database, while the 840M supports Vulkan 1.4, though its Vulkan benchmark score of 4880 indicates modest performance in that API.