NVIDIA GeForce 940M vs NVIDIA GeForce GTX 770M Comparison

Head-to-Head Benchmarks

The database contains one direct head-to-head benchmark result for these two mobile GPUs, and it is decisive. In the Geekbench OpenCL test, the NVIDIA GeForce GTX 770M scores 8,552 points against the NVIDIA GeForce 940M's 6,018 points. That is a 42.1% advantage for the older Kepler-based card. This is not a marginal victory; it is a substantial performance gap that places the GTX 770M in a different performance tier entirely.

Looking at the broader benchmark picture, the GTX 770M's average benchmark score of 6,000 puts it in the 35th percentile of all GPUs in the database. The 940M's average score is 5,284, which places it in the 31st percentile. While both are near the lower-middle range of the database, the GTX 770M holds a meaningful edge in raw compute workloads, which is exactly what the OpenCL test measures.

The GTX 770M's closest rival in the database is the AMD Radeon RX 6400, which scores 6,001, a delta of 0%. This effectively means the GTX 770M performs on par with that desktop card in the database's average scoring. Its other near-neighbors include the NVIDIA RTX PRO 6000 Blackwell Server at 5,996 (0.1% behind) and the AMD FirePro W4100 at 5,987 (0.2% behind). These are tight groupings, but they show that the GTX 770M's compute performance is consistent with a well-established baseline.

For the 940M, the nearest rival is the NVIDIA GeForce GTX 980M, which scores 5,308, a delta of -0.4%. This is unusual because the GTX 980M is typically a much more powerful part, but the database's average scoring puts them close together. The 940M also sits near the GeForce 930A (5,317, -0.6%) and the GeForce 840M (5,322, -0.7%). The data indicates the 940M's performance is tightly clustered with other low-end Maxwell parts, and it trails the GTX 770M by a wide margin in the one test where they directly compete.

The only other benchmark for the GTX 770M is a Geekbench Metal score of 3,448, which is not directly comparable to the 940M's Vulkan score of 4,549 since they use different APIs. However, these scores do show that the GTX 770M supports Metal, while the 940M supports Vulkan, a difference that will matter depending on the software environment.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The GTX 770M has an average benchmark score of 6,000, while the 940M scores 5,284. The GTX 770M also holds a higher percentile ranking at 35th versus the 940M's 31st.

Q: How much faster is the GTX 770M in the OpenCL test?

A: The GTX 770M scores 8,552 versus the 940M's 6,018, which is a 42.1% advantage. This is the only direct head-to-head test in the database, and the GTX 770M wins it.

Q: What memory types do these GPUs use?

A: The GTX 770M uses 3 GB of GDDR5 memory on a 192-bit bus, while the 940M uses 2 GB of DDR3 memory on a 64-bit bus. This is a major architectural difference that affects bandwidth.

Q: Do both GPUs support the same DirectX version?

A: Yes, both support DirectX 12 (11_0) and OpenGL 4.6. However, the Vulkan support differs: the GTX 770M supports Vulkan 1.2.175, while the 940M supports Vulkan 1.4.

Q: Which GPU is newer?

A: The GTX 770M was released on May 29, 2013, while the 940M came later on March 12, 2015. The 940M is the newer part by almost two years.

Q: Are both GPUs the same power draw?

A: Yes, both have a TDP of 75 W and use MXM Module slot width with no power connectors. They are similarly power-constrained mobile parts.

Architecture Differences

The two GPUs come from different NVIDIA architectures, which explains much of the performance difference. The GTX 770M uses the GK106 chip built on the Kepler architecture, while the 940M uses the GM107 chip on the Maxwell architecture. Both are manufactured by TSMC on a 28 nm process, but the similarities end there.

The Kepler chip is physically larger, with a die size of 221 mm² and 2,540 million transistors. This gives it a transistor density of 11.5 million per mm². The Maxwell chip, in contrast, is smaller at 148 mm² with 1,870 million transistors, but it achieves a higher transistor density of 12.6 million per mm². This density improvement is a hallmark of Maxwell's design efficiency.

The GTX 770M has significantly more compute resources: 960 shading units, 80 texture mapping units, and 24 ROPs. The 940M, by comparison, has 512 shading units, 32 TMUs, and 16 ROPs. This means the GTX 770M has nearly double the shader count and more than double the texture units, which directly translates to higher fill rates. The GTX 770M also has a higher texture rate of 63.76 GTexel/s versus the 940M's 35.14 GTexel/s, and a slightly lower pixel rate of 15.94 GPixel/s versus 17.57 GPixel/s.

Clock speeds tell a different story. The 940M runs at a higher base clock of 1020 MHz and a boost of 1098 MHz, compared to the GTX 770M's 706 MHz base and 797 MHz boost. This higher clock speed helps the Maxwell part in some throughput calculations, but the sheer number of execution units on the Kepler part overwhelms the clock advantage. The FP32 compute rating shows this clearly: the GTX 770M delivers 1.530 TFLOPS, while the 940M manages 1,124.4 GFLOPS.

Memory architecture is a critical differentiator. The GTX 770M uses 3 GB of GDDR5 with a 192-bit bus, yielding a bandwidth of 96.19 GB/s. The 940M uses 2 GB of DDR3 with a 64-bit bus, which limits bandwidth to just 14.40 GB/s. This is a massive gap in memory throughput, and it will severely bottleneck the 940M in any texture-heavy or large-data workload.

The GTX 770M also has a higher memory clock at 1002 MHz (4 Gbps effective) versus the 940M's 900 MHz (1,800 Mbps effective). Both GPUs share the same MXM-B (3.0) bus interface and portable-device-dependent display outputs. Neither has dedicated RT or tensor cores, as both predate those features.

The Verdict

The data is unambiguous: the NVIDIA GeForce GTX 770M is the stronger GPU in compute performance. It wins the only direct benchmark test by 42.1%, has a higher average benchmark score, and sits in a higher percentile. Anyone choosing between these two for a laptop should pick the GTX 770M if raw performance is the priority.

The 940M's advantages are limited to newer architecture, higher clock speeds, and newer Vulkan support. It is also a later release by almost two years, which might suggest better driver optimization in the long run, but the benchmark data does not support that expectation. The 940M's average score of 5,284 is lower, and its closest rivals include other low-end Maxwell parts like the 840M and 930A, indicating it is firmly in the entry-level segment.

The GTX 770M's nearest rivals include desktop cards like the RX 6400 and workstation parts like the Quadro K4000M, which suggests its performance level is more versatile. The 940M's nearest rivals are all mobile parts with similar or slightly higher scores, reinforcing its position as a budget mobile solution.

For gaming, the GTX 770M's superior memory bandwidth and higher shader count will deliver better frame rates in most scenarios, especially at higher resolutions or with higher texture quality settings. The 940M's lower bandwidth will struggle with modern game assets, and its 2 GB VRAM is a limitation for current releases.

For compute tasks like OpenCL workloads, the GTX 770M is the clear winner by a wide margin. The 42.1% lead in the OpenCL test suggests it will handle GPGPU tasks with significantly less latency and higher throughput.

Specification Differences

The key specification differences between the two GPUs are as follows:

• Chip and architecture: GK106 on Kepler versus GM107 on Maxwell
• Transistors: 2,540 million versus 1,870 million
• Die size: 221 mm² versus 148 mm²
• Transistor density: 11.5M / mm² versus 12.6M / mm²
• Base clock: 706 MHz versus 1020 MHz
• Boost clock: 797 MHz versus 1098 MHz
• Memory clock: 1002 MHz (4 Gbps effective) versus 900 MHz (1,800 Mbps effective)
• Memory size: 3 GB versus 2 GB
• Memory type: GDDR5 versus DDR3
• Memory bus width: 192 bit versus 64 bit
• Memory bandwidth: 96.19 GB/s versus 14.40 GB/s
• Shading units: 960 versus 512
• TMUs: 80 versus 32
• ROPs: 24 versus 16
• Pixel rate: 15.94 GPixel/s versus 17.57 GPixel/s
• Texture rate: 63.76 GTexel/s versus 35.14 GTexel/s
• FP32 compute: 1.530 TFLOPS versus 1,124.4 GFLOPS
• Vulkan version: 1.2.175 versus 1.4
• Release date: 2013-05-29 versus 2015-03-12
• Predecessor: GeForce 600M versus GeForce 800M
• Successor: GeForce 800M versus GeForce 10 Mobile

Where Each One Wins

The GTX 770M wins in virtually every performance-oriented category. Its memory bandwidth is over six times higher, its texture rate is nearly double, and its FP32 compute is about 36% higher. This makes it the better choice for gaming at higher settings, for any application that relies on memory throughput, and for OpenCL-based compute workloads.

The 940M wins in efficiency of design, with a higher transistor density and higher clock speeds. It also has newer Vulkan support (1.4 versus 1.2.175), which may be relevant for users running Vulkan-based applications or emulators. Its higher pixel rate of 17.57 GPixel/s versus 15.94 GPixel/s gives it a small edge in pure pixel-fill workloads, though this is unlikely to overcome its bandwidth deficit in real-world scenarios.

For users who prioritize raw performance, the GTX 770M is the only sensible choice. For users who need newer API support or who are constrained by a system that requires a Maxwell part, the 940M is an acceptable fallback, but the benchmark data shows it is a clear step down in capability. The GTX 770M's end-of-life status and older release date do not diminish its performance advantage in the recorded data.