NVIDIA GeForce GT 755M vs NVIDIA GeForce GTX 650 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GT 755M leads with an average benchmark score of 4033, while the NVIDIA GeForce GTX 650 trails at 3823. This places the GT 755M at the 24th percentile of all GPUs, compared to the GTX 650’s 22nd percentile.

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

A: The GT 755M wins decisively, scoring 3132 against the GTX 650’s 2400, a 30.5% advantage. This is the largest performance gap between the two in any benchmark.

Q: What is the difference in OpenCL performance?

A: The GT 755M again comes out ahead, scoring 4934 versus 4545 for the GTX 650, an 8.6% margin. The GTX 650 does have an additional Vulkan score of 4524, but the GT 755M has no Vulkan result in the data.

Q: Which card has more memory?

A: The GT 755M has 2 GB of GDDR5 memory, double the GTX 650’s 1024 MB. Both use a 128-bit memory bus, but the GT 755M’s memory runs at 1350 MHz (5.4 Gbps effective) versus 1250 MHz (5 Gbps effective) for the GTX 650.

Q: Are both GPUs based on the same architecture?

A: Yes, both are built on NVIDIA’s Kepler architecture using a 28 nm process from TSMC. However, they use different chips: the GT 755M uses GK107, while the GTX 650 uses GK106.

Q: Which card has a higher TDP?

A: The GTX 650 has a higher TDP at 65 W, compared to the GT 755M’s 50 W. The GTX 650 also requires a 6-pin power connector and a 250 W suggested PSU, whereas the GT 755M uses an MXM module with no power connectors.

Architecture Differences

The two GPUs share NVIDIA’s Kepler architecture and the same 28 nm TSMC manufacturing process, but they are built on different silicon. The GT 755M uses the GK107 chip with 1,270 million transistors on a 118 mm² die, while the GTX 650 uses the larger GK106 chip with 2,540 million transistors on a 221 mm² die. This gives the GK106 a higher transistor density of 11.5M / mm² versus the GK107’s 10.8M / mm².

Despite the chip size difference, both GPUs have identical core configurations: 384 shading units, 32 texture mapping units, and 16 raster operations units. The clock speeds differ, however. The GT 755M has a base clock of 980 MHz with a boost clock of 1020 MHz, while the GTX 650’s base and boost clocks are not specified in the data. Instead, the GTX 650’s memory runs at 1250 MHz (5 Gbps effective), slightly lower than the GT 755M’s 1350 MHz (5.4 Gbps effective).

The GT 755M is a mobile-oriented MXM module with display outputs described as “Portable Device Dependent,” while the GTX 650 is a desktop single-slot card with 1x DVI and 2x DisplayPort 1.2 outputs. The GTX 650 also has a physical length of 147 mm (5.8 inches), a detail not listed for the GT 755M. Both support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175, and both use a PCIe 3.0 x16 bus interface.

The production status for both is end-of-life. The GT 755M belongs to the GeForce 700M generation with a release date of June 2013, while the GTX 650 is from the GeForce 600 generation with a later release date of November 2013. The GT 755M’s predecessor is GeForce 600M and its successor is GeForce 800M, whereas the GTX 650’s predecessor is GeForce 500 and its successor is GeForce 700.

Where Each One Wins

The data shows the GT 755M wins both head-to-head benchmarks, so the case for the GTX 650 rests on specific advantages rather than raw performance. The GT 755M is the clear winner in compute workloads, particularly in Metal-based applications where its 30.5% lead over the GTX 650 is substantial. Its OpenCL advantage of 8.6% is more moderate but still consistent.

The GTX 650’s wins are structural. It has a larger chip with double the transistor count (2,540 million versus 1,270 million), which suggests headroom for different workloads even if the benchmark scores do not reflect it. The GTX 650 also offers a Vulkan benchmark score of 4524, which the GT 755M lacks entirely — this means the GTX 650 is the only one of the two with data supporting Vulkan API performance. For users targeting Vulkan-based applications, the GTX 650 is the only option with verified results.

The GT 755M wins on efficiency and form factor. Its 50 W TDP is 15 W lower than the GTX 650’s 65 W, and it requires no power connectors, making it suitable for portable devices. The GTX 650 needs a 6-pin connector and a 250 W PSU, which limits it to desktop builds. The GT 755M’s 2 GB memory capacity is also double the GTX 650’s 1 GB, which matters for texture-heavy workloads.

In terms of memory bandwidth, the GT 755M edges out the GTX 650 at 86.40 GB/s versus 80.00 GB/s. Pixel rate and texture rate are nearly identical, with the GTX 650 slightly ahead at 8.464 GPixel/s and 33.86 GTexel/s versus the GT 755M’s 8.160 GPixel/s and 32.64 GTexel/s. FP32 performance also favors the GTX 650 marginally at 812.5 GFLOPS versus 783.4 GFLOPS.

Specification Differences

The specifications where the two GPUs differ are as follows:

• Chip: GK107 (GT 755M) vs GK106 (GTX 650)
• Generation: GeForce 700M vs GeForce 600
• Transistors: 1,270 million vs 2,540 million
• Die size: 118 mm² vs 221 mm²
• Transistor density: 10.8M / mm² vs 11.5M / mm²
• Base clock: 980 MHz vs not specified
• Boost clock: 1020 MHz vs not specified
• Memory clock: 1350 MHz (5.4 Gbps effective) vs 1250 MHz (5 Gbps effective)
• Memory size: 2 GB vs 1024 MB
• Bandwidth: 86.40 GB/s vs 80.00 GB/s
• Pixel rate: 8.160 GPixel/s vs 8.464 GPixel/s
• Texture rate: 32.64 GTexel/s vs 33.86 GTexel/s
• FP32: 783.4 GFLOPS vs 812.5 GFLOPS
• TDP: 50 W vs 65 W
• Slot width: MXM Module vs Single-slot
• Power connectors: None vs 1x 6-pin
• Suggested PSU: Not specified vs 250 W
• Display outputs: Portable Device Dependent vs 1x DVI, 2x DisplayPort 1.2
• Length: Not specified vs 147 mm (5.8 inches)
• Release date: June 2013 vs November 2013
• Predecessor: GeForce 600M vs GeForce 500
• Successor: GeForce 800M vs GeForce 700

Both GPUs share the same shading units (384), TMUs (32), ROPs (16), memory type (GDDR5), bus width (128 bit), foundry (TSMC), process node (28 nm), architecture (Kepler), APIs (DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.2.175), bus interface (PCIe 3.0 x16), and production status (end-of-life).

Head-to-Head Benchmarks

The head-to-head data covers two tests, and the GT 755M wins both. The most striking result is in Geekbench Metal, where the GT 755M scores 3132 against the GTX 650’s 2400. This 30.5% delta is the single largest performance gap between the two cards in any metric. The GT 755M’s advantage here is significant enough to be considered a class difference rather than a minor edge, especially for Metal-accelerated applications.

In Geekbench OpenCL, the GT 755M wins again with 4934 versus 4545, a more modest 8.6% lead. This narrower margin suggests that in general-purpose compute tasks, the two GPUs are closer in capability than the Metal results imply. The GTX 650’s FP32 rating of 812.5 GFLOPS is actually slightly higher than the GT 755M’s 783.4 GFLOPS, which partly explains why the OpenCL gap is smaller — raw compute throughput is similar, but the GT 755M’s higher memory bandwidth (86.40 GB/s vs 80.00 GB/s) may contribute to its advantage.

The GTX 650 has a Vulkan score of 4524, but there is no corresponding GT 755M Vulkan result, so a direct comparison is impossible. The GTX 650’s Vulkan score is higher than its OpenCL score (4524 vs 4545) and its Metal score (2400), indicating that Vulkan is its strongest API in the data. However, without a GT 755M number, this cannot be interpreted as a win — it is simply an unavailable data point.

The GT 755M’s 2 GB memory capacity versus the GTX 650’s 1024 MB likely plays a role in the Metal benchmark’s large delta, as modern workloads can exceed 1 GB VRAM. The GTX 650’s lower memory clock (1250 MHz vs 1350 MHz) compounds this, resulting in 6.4 GB/s less bandwidth. The GTX 650 does slightly outperform the GT 755M in pixel rate (8.464 vs 8.160 GPixel/s), texture rate (33.86 vs 32.64 GTexel/s), and FP32 (812.5 vs 783.4 GFLOPS), but these advantages are all under 4% and do not translate into benchmark wins.

The Verdict

The data is unambiguous: the NVIDIA GeForce GT 755M is the faster GPU in every head-to-head benchmark recorded. It wins Geekbench Metal by 30.5% and Geekbench OpenCL by 8.6%, and it holds advantages in memory capacity (2 GB vs 1024 MB) and memory bandwidth (86.40 GB/s vs 80.00 GB/s). Its average benchmark score of 4033 is 5.5% higher than the GTX 650’s 3823, and its 24th percentile ranking versus 22nd confirms the overall performance edge.

The GTX 650 is not without its own merits. It has a larger chip with more transistors, a slightly higher FP32 rating, and marginally better pixel and texture rates. It is the only one of the two with a Vulkan benchmark score, and its desktop form factor with dedicated display outputs makes it a drop-in card for standard systems. However, none of these advantages translate into a benchmark win, and its higher TDP (65 W vs 50 W) plus the need for a 6-pin power connector make it less efficient and less flexible.

For users who prioritize raw performance in Metal and OpenCL workloads, the GT 755M is the clear pick. Its 2 GB memory and higher bandwidth give it headroom for modern applications, and its lower power draw makes it suitable for portable devices. The GTX 650 should be chosen only when Vulkan support is a hard requirement (since it has verified Vulkan results), or when a desktop single-slot card with fixed display outputs is needed. In all other scenarios, the benchmark data favors the GT 755M without qualification.