NVIDIA GeForce GTX 675M vs NVIDIA GeForce GTX 770M Comparison

Head-to-Head Benchmarks

The recorded benchmark data contains a single direct comparison between the NVIDIA GeForce GTX 675M and the NVIDIA GeForce GTX 770M. In the Geekbench OpenCL test, the GTX 770M scores 8552 points while the GTX 675M scores 6946 points. This gives the GTX 770M an 18.8 percent advantage, a decisive margin in raw compute workloads. The GTX 770M’s higher score reflects a substantial performance gap, not a marginal one.

The GTX 675M’s score of 6946 places it in the 39th percentile among all GPUs in the database. Its closest rivals include the AMD Radeon R5 M240 with an average score of 6975, just 0.4 percent higher, and the NVIDIA GeForce GTX 680M at 7023, which is 1.1 percent faster. The NVIDIA T600 scores 7035, 1.3 percent above the GTX 675M. Meanwhile, the AMD FirePro M5100 trails at 6830, meaning the GTX 675M is 1.7 percent faster than that card. These comparisons show that the GTX 675M sits in a tightly contested performance band, where small score differences separate it from several similarly positioned mobile GPUs.

The GTX 770M’s average benchmark score across all recorded tests is 6000 points, which places it in the 35th percentile. This average includes both its Geekbench OpenCL result of 8552 and a Geekbench Metal score of 3448. The Metal score pulls the average down, so the OpenCL result is the more representative measure of its raw compute capability. Its nearest rivals include the AMD Radeon RX 6400 with an average score of 6001, essentially tied at a 0 percent difference, and the NVIDIA RTX PRO 6000 Blackwell Server at 5996, which is 0.1 percent lower. The AMD FirePro W4100 scores 5987, 0.2 percent below, and the NVIDIA Quadro K4000M also lands at 5986, another 0.2 percent gap. These rivals cluster tightly around the GTX 770M’s average, though that average is skewed by the Metal benchmark.

Looking strictly at the head-to-head OpenCL result, the GTX 770M wins the only direct comparison recorded. The GTX 675M records zero benchmark wins while the GTX 770M records one. This is a limited dataset, but the magnitude of the OpenCL gap suggests a clear compute hierarchy. The GTX 770M’s 18.8 percent lead in this test is consistent with its newer architecture and higher shading unit count, which will be examined in later sections.

FAQ

Q: Which GPU scores higher in the only head-to-head benchmark?

A: The NVIDIA GeForce GTX 770M scores 8552 in Geekbench OpenCL, while the NVIDIA GeForce GTX 675M scores 6946. The GTX 770M leads by 18.8 percent.

Q: How does the GTX 675M compare to its nearest rivals?

A: The GTX 675M’s 6946 OpenCL score is 0.4 percent lower than the AMD Radeon R5 M240’s 6975, 1.1 percent lower than the NVIDIA GeForce GTX 680M’s 7023, and 1.3 percent lower than the NVIDIA T600’s 7035. It is 1.7 percent higher than the AMD FirePro M5100’s 6830.

Q: What is the GTX 770M’s average benchmark score and percentile?

A: The GTX 770M has an average benchmark score of 6000 across all tests, placing it in the 35th percentile of all GPUs. This average includes an OpenCL score of 8552 and a Metal score of 3448.

Q: Which GPUs are closest to the GTX 770M in average score?

A: The AMD Radeon RX 6400 averages 6001, a 0 percent difference. The NVIDIA RTX PRO 6000 Blackwell Server averages 5996, 0.1 percent lower. The AMD FirePro W4100 averages 5987, 0.2 percent lower, and the NVIDIA Quadro K4000M averages 5986, also 0.2 percent lower.

Q: Does the GTX 770M support Vulkan?

A: Yes, the GTX 770M supports Vulkan version 1.2.175. The GTX 675M has no Vulkan support listed in the database.

Q: What is the percentile ranking difference between the two GPUs?

A: The GTX 675M ranks in the 39th percentile, while the GTX 770M ranks in the 35th percentile. Despite the GTX 770M’s higher OpenCL score, its average is lower due to the Metal result.

Architecture Differences

The two GPUs come from different NVIDIA generations and architectures. The GTX 675M uses the GF114 chip based on Fermi 2.0 architecture, part of the GeForce 600M generation. The GTX 770M uses the GK106 chip based on Kepler architecture, part of the GeForce 700M generation. This generational shift is significant for compute efficiency and feature support.

The manufacturing process differs substantially. The GTX 675M is built on a 40 nanometer node at TSMC, while the GTX 770M uses a 28 nanometer node, also at TSMC. The smaller node allows the GTX 770M to pack more transistors into a smaller die. The GTX 675M contains 1,950 million transistors on a 332 square millimeter die, yielding a transistor density of 5.9 million per square millimeter. The GTX 770M contains 2,540 million transistors on a 221 square millimeter die, achieving a density of 11.5 million per square millimeter. This is nearly double the density, a direct result of the newer process technology.

The shading unit count differs dramatically. The GTX 675M has 384 shading units, while the GTX 770M has 960, a 2.5 times increase. Texture mapping units also increase from 64 to 80. However, the render output units drop from 32 on the GTX 675M to 24 on the GTX 770M. This trade-off means the GTX 770M excels at shading and texturing but has fewer pixel output pipelines.

Clock speeds are not directly comparable because the GTX 675M lacks recorded base and boost clocks. The GTX 770M has a base clock of 706 MHz and a boost clock of 797 MHz. Memory clocks differ as well: the GTX 675M runs at 750 MHz with 3 Gbps effective speed, while the GTX 770M runs at 1002 MHz with 4 Gbps effective speed. The GTX 675M has no Vulkan support listed, while the GTX 770M supports Vulkan 1.2.175. Both support DirectX 12 (11_0) and OpenGL 4.6.

The rated thermal design power differs notably. The GTX 675M is rated at 100 watts, while the GTX 770M is rated at 75 watts. This means the newer card delivers higher compute performance while drawing less power, a benefit of the 28 nanometer process and Kepler efficiency. Both use MXM modules with no power connectors and portable-device-dependent display outputs.

The Verdict

The benchmark data points clearly to the GTX 770M as the stronger compute performer. In the only direct comparison, Geekbench OpenCL, the GTX 770M leads by 18.8 percent with a score of 8552 versus 6946. This is not a small edge; it represents a meaningful generational improvement. The GTX 770M also has 960 shading units versus 384, 80 texture units versus 64, and higher effective memory speed. These architectural advantages translate directly into the observed score difference.

However, the GTX 675M holds a few distinctions. Its 39th percentile ranking is higher than the GTX 770M’s 35th percentile, though this is due to the GTX 770M’s Metal benchmark score of 3448 dragging down its average. The GTX 675M also has a wider 256 bit memory bus compared to the GTX 770M’s 192 bit bus, and it has 32 render output units versus 24. But the memory bandwidth figures are nearly identical: 96.00 GB/s for the GTX 675M and 96.19 GB/s for the GTX 770M. The wider bus compensates for the GTX 675M’s lower memory clock, resulting in a tie on bandwidth.

For a user prioritizing raw compute performance in OpenCL workloads, the GTX 770M is the clear choice. Its 18.8 percent lead in the head-to-head test, combined with double the shading units and a higher texture rate, makes it the superior card for general compute tasks. The GTX 770M also supports Vulkan, which the GTX 675M lacks, adding compatibility with modern graphics APIs. Its lower power draw of 75 watts versus 100 watts is an additional practical advantage in a laptop thermal envelope.

For a user focused on pixel throughput or memory bus width, the GTX 675M has specific strengths. Its pixel rate of 9.920 GPixel/s is lower than the GTX 770M’s 15.94 GPixel/s, so the GTX 770M wins there too. The GTX 675M’s only real advantages are the wider memory bus and higher render output unit count, but these do not overcome the GTX 770M’s overall performance lead in the recorded data.

The verdict from the data is straightforward: the GTX 770M outperforms the GTX 675M in the measured benchmark and offers better architectural efficiency. The GTX 675M remains a competent card from an earlier generation, but the GTX 770M represents the stronger choice for anyone comparing these two based on the database results.

Specification Differences

The following fields differ between the two GPUs:

• Chip: GTX 675M uses GF114; GTX 770M uses GK106.
• Architecture: GTX 675M uses Fermi 2.0; GTX 770M uses Kepler.
• Generation: GTX 675M is GeForce 600M; GTX 770M is GeForce 700M.
• Process Node: GTX 675M is 40 nm; GTX 770M is 28 nm.
• Transistors: GTX 675M has 1,950 million; GTX 770M has 2,540 million.
• Die Size: GTX 675M is 332 mm²; GTX 770M is 221 mm².
• Transistor Density: GTX 675M has 5.9M per mm²; GTX 770M has 11.5M per mm².
• Base Clock: GTX 675M has none listed; GTX 770M has 706 MHz.
• Boost Clock: GTX 675M has none listed; GTX 770M has 797 MHz.
• Memory Clock: GTX 675M runs at 750 MHz, 3 Gbps effective; GTX 770M runs at 1002 MHz, 4 Gbps effective.
• Memory Size: GTX 675M has 2 GB; GTX 770M has 3 GB.
• Memory Bus Width: GTX 675M has 256 bit; GTX 770M has 192 bit.
• Shading Units: GTX 675M has 384; GTX 770M has 960.
• Texture Mapping Units: GTX 675M has 64; GTX 770M has 80.
• Render Output Units: GTX 675M has 32; GTX 770M has 24.
• Pixel Rate: GTX 675M is 9.920 GPixel/s; GTX 770M is 15.94 GPixel/s.
• Texture Rate: GTX 675M is 39.68 GTexel/s; GTX 770M is 63.76 GTexel/s.
• FP32 Performance: GTX 675M is 952.3 GFLOPS; GTX 770M is 1.530 TFLOPS.
• TDP: GTX 675M is 100 W; GTX 770M is 75 W.
• Vulkan Support: GTX 675M has none listed; GTX 770M supports 1.2.175.
• Release Date: GTX 675M launched on 2012-03-21; GTX 770M launched on 2013-05-29.
• Predecessor: GTX 675M’s predecessor is GeForce 500M; GTX 770M’s predecessor is GeForce 600M.
• Successor: GTX 675M’s successor is GeForce 700M; GTX 770M’s successor is GeForce 800M.
• Benchmark Scores: GTX 675M has one OpenCL score of 6946; GTX 770M has OpenCL of 8552 and Metal of 3448.
• Average Benchmark Score: GTX 675M is 6946; GTX 770M is 6000.
• Percentile: GTX 675M is 39th; GTX 770M is 35th.

Where Each One Wins

The GTX 770M wins in every meaningful compute category from the recorded data. Its OpenCL score is 18.8 percent higher than the GTX 675M’s. Its shading unit count of 960 dwarfs the GTX 675M’s 384. Its texture rate of 63.76 GTexel/s is 60.7 percent higher than the GTX 675M’s 39.68 GTexel/s. Its pixel rate of 15.94 GPixel/s is 60.7 percent higher than the GTX 675M’s 9.920 GPixel/s. Its FP32 performance of 1.530 TFLOPS is 60.7 percent higher than the GTX 675M’s 952.3 GFLOPS. Its memory bandwidth is marginally higher at 96.19 GB/s versus 96.00 GB/s. It has 3 GB of memory versus 2 GB. It supports Vulkan while the GTX 675M does not. It draws 25 percent less power at 75 watts versus 100 watts. It has a smaller die at 221 mm² versus 332 mm², and a higher transistor density.

The GTX 675M wins in a few narrow specification categories. It has a wider memory bus at 256 bit versus 192 bit. It has more render output units at 32 versus 24. It has a larger die area, though this is not a performance advantage. It has a higher percentile ranking at 39th versus 35th, though this is due to the GTX 770M’s Metal score lowering its average. It has a higher average benchmark score of 6946 versus 6000 for the same reason. Its transistor count is lower at 1,950 million versus 2,540 million, which is not a win but a difference.

For use cases involving OpenCL compute, the GTX 770M is the obvious pick. For users who need Vulkan support, the GTX 770M is the only option between the two. For workloads that stress pixel output, the GTX 770M’s higher pixel rate wins. For memory bandwidth, the two are effectively tied, so neither has a practical edge. The GTX 675M’s wider memory bus and higher ROP count do not translate into any recorded benchmark win, and its higher percentile is an artifact of the GTX 770M’s Metal benchmark inclusion. The data supports the GTX 770M as the superior GPU in this comparison.