NVIDIA GeForce GTX 675M vs NVIDIA GeForce GTX 750 Comparison

NVIDIA
GEFORCE

NVIDIA GeForce GTX 675M

CORE STATE GF114
VRAM 2 GB
CLOCK SPEED —
TDP 100 W
BUS WIDTH 256 bit
ARCHITECTURE Fermi 2.0
nm
PROCESS 40 nm
LAUNCH DATE 2012
VS
NVIDIA
GEFORCE

GeForce GTX 750

CORE STATE GM107
VRAM 1024 MB
CLOCK SPEED 1085 MHz
TDP 55 W
BUS WIDTH 128 bit
ARCHITECTURE Maxwell
nm
PROCESS 28 nm
LAUNCH DATE 2014

PERFORMANCE BENCHMARKS

geekbench_opencl
6,946
9,315
geekbench_metal
N/A
4,274
geekbench_vulkan
N/A
8,078

Analysis: NVIDIA GeForce GTX 675M vs NVIDIA GeForce GTX 750

The NVIDIA GeForce GTX 750 and the NVIDIA GeForce GTX 675M represent two very different approaches to graphics processing, separated by two years of architectural evolution. The desktop-oriented GTX 750, built on the Maxwell architecture, is the clear performance leader in the available head-to-head data, while the older mobile GTX 675M, based on Fermi 2.0, offers a larger memory pool and wider memory bus. Benchmark results show a decisive 34.1% victory for the GTX 750 in the only directly comparable test, yet the GTX 675M retains distinct advantages in memory capacity and texture throughput that matter in specific workloads.

Head-to-Head Benchmarks

The single available head-to-head benchmark, Geekbench OpenCL, delivers a definitive result. The NVIDIA GeForce GTX 750 scores 9,315 points against 6,946 points for the NVIDIA GeForce GTX 675M. This represents a 34.1% delta in favor of the GTX 750, a substantial margin that places the two cards in different performance tiers. The GTX 750's average benchmark score across all tests (7,222) is also higher than the GTX 675M's average (6,946), though this comparison is skewed by the fact that the GTX 675M only has a single benchmark result available.

Looking at the broader competitive landscape, the GTX 750's OpenCL score places it in the 40th percentile of all GPUs, while the GTX 675M sits just below at the 39th percentile. Despite the large head-to-head delta, both cards occupy a similar percentile ranking, indicating that the GTX 675M's single score is competitive within its own generation. The GTX 750's nearest rivals in average score include the AMD Radeon Vega 8 Mobile (7,203, a 0.3% delta), NVIDIA GeForce GTX 560 SE (7,171, a 0.7% delta), and Intel Iris Pro Graphics P580 (7,170, a 0.7% delta). Interestingly, the NVIDIA GeForce GTX 970 appears in the GTX 750's rival list with an average score of 7,157, which is actually 0.9% lower than the GTX 750's average, highlighting the GTX 750's strong compute showing.

The GTX 675M's nearest rivals tell a different story. The AMD Radeon R5 M240 (6,975, a -0.4% delta) is essentially tied, while the NVIDIA GeForce GTX 680M (7,023, a -1.1% delta) and NVIDIA T600 (7,035, a -1.3% delta) are slightly ahead. The AMD FirePro M5100 (6,830, a 1.7% delta) trails the GTX 675M. This data suggests the GTX 675M is competitive within its mobile segment, but the GTX 750's OpenCL advantage over it is substantial and consistent with its newer architecture.

Architecture Differences

The architectural gap between these two GPUs is generational. The GTX 750 uses the GM107 chip built on the Maxwell architecture, fabricated by TSMC on a 28 nm process. The GTX 675M uses the GF114 chip on the older Fermi 2.0 architecture, also from TSMC but on a 40 nm process. This process shrink is significant: the GTX 750 packs 1,870 million transistors into a 148 mm² die, achieving a transistor density of 12.6 million per mm². The GTX 675M, by contrast, contains 1,950 million transistors on a much larger 332 mm² die, yielding only 5.9 million transistors per mm². The GTX 750 is therefore more than twice as dense, a direct result of the newer manufacturing technology.

Core configuration differs markedly between the two. The GTX 750 has 512 shading units, 32 texture mapping units, and 16 render output units. The GTX 675M has fewer shading units (384) but double the TMUs (64) and double the ROPs (32). This explains the texture rate advantage: the GTX 675M achieves 39.68 GTexel/s versus 34.72 GTexel/s for the GTX 750. However, the GTX 750's pixel rate is superior at 17.36 GPixel/s compared to 9.92 GPixel/s for the GTX 675M, thanks to its much higher clock speeds. The GTX 750 operates at a base clock of 1020 MHz with a boost of 1085 MHz, while the GTX 675M's core clocks are not specified in the data.

Compute performance favors the GTX 750. Its FP32 throughput is 1,111.0 GFLOPS, compared to 952.3 GFLOPS for the GTX 675M. Neither card offers FP16 support. Memory architecture also diverges: the GTX 750 has 1024 MB of GDDR5 on a 128-bit bus, delivering 80.19 GB/s of bandwidth. The GTX 675M has 2 GB of GDDR5 on a 256-bit bus, achieving 96.00 GB/s. The GTX 675M's memory runs at 750 MHz (3 Gbps effective), while the GTX 750's memory runs at 1253 MHz (5 Gbps effective). The GTX 675M's wider bus compensates for its slower memory clock, resulting in higher overall bandwidth.

API support is another differentiator. Both cards support DirectX 12 (11_0) and OpenGL 4.6. However, the GTX 750 supports Vulkan 1.4, while the GTX 675M has no Vulkan support listed. Power requirements diverge sharply: the GTX 750 has a TDP of 55 W and requires no power connectors, with a suggested PSU of 250 W. The GTX 675M has a 100 W TDP and is an MXM module, meaning it draws power through the mobile platform rather than dedicated connectors.

FAQ

Q: Which card is faster in OpenCL compute workloads?

A: The NVIDIA GeForce GTX 750 is decisively faster, scoring 9,315 in Geekbench OpenCL versus 6,946 for the GTX 675M, a 34.1% advantage.

Q: Does the GTX 675M have any advantage over the GTX 750?

A: Yes, in memory configuration. The GTX 675M offers 2 GB of VRAM on a 256-bit bus with 96.00 GB/s bandwidth, versus 1 GB on a 128-bit bus with 80.19 GB/s for the GTX 750. It also has double the TMUs (64 vs 32) and ROPs (32 vs 16), giving it a higher texture rate of 39.68 GTexel/s.

Q: Why does the GTX 750 achieve a higher pixel rate despite fewer ROPs?

A: The GTX 750's much higher clock speeds (1020 MHz base, 1085 MHz boost) compensate for its 16 ROPs, resulting in 17.36 GPixel/s. The GTX 675M, with unspecified core clocks and 32 ROPs, only reaches 9.92 GPixel/s.

Q: Is the GTX 750 more power-efficient?

A: The data strongly indicates so. The GTX 750 has a TDP of 55 W, while the GTX 675M has a TDP of 100 W. The GTX 750 also achieves higher FP32 performance (1,111.0 GFLOPS vs 952.3 GFLOPS) at that lower power draw.

Q: Do both cards support modern APIs?

A: Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 750 additionally supports Vulkan 1.4, while the GTX 675M has no Vulkan support listed.

Q: Which card has a better transistor density?

A: The GTX 750, at 12.6 million transistors per mm², is more than double the GTX 675M's 5.9 million per mm². This reflects the newer 28 nm process versus the older 40 nm node.

Specification Differences

| Specification | NVIDIA GeForce GTX 750 | NVIDIA GeForce GTX 675M |

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

| Architecture | Maxwell | Fermi 2.0 |

| Process Node | 28 nm | 40 nm |

| Transistors | 1,870 million | 1,950 million |

| Die Size | 148 mm² | 332 mm² |

| Transistor Density | 12.6M / mm² | 5.9M / mm² |

| Base Clock | 1020 MHz | Not specified |

| Boost Clock | 1085 MHz | Not specified |

| Memory Clock | 1253 MHz (5 Gbps effective) | 750 MHz (3 Gbps effective) |

| Memory Size | 1024 MB | 2 GB |

| Memory Bus Width | 128 bit | 256 bit |

| Memory Bandwidth | 80.19 GB/s | 96.00 GB/s |

| Shading Units | 512 | 384 |

| TMUs | 32 | 64 |

| ROPs | 16 | 32 |

| Pixel Rate | 17.36 GPixel/s | 9.92 GPixel/s |

| Texture Rate | 34.72 GTexel/s | 39.68 GTexel/s |

| FP32 Performance | 1,111.0 GFLOPS | 952.3 GFLOPS |

| TDP | 55 W | 100 W |

| Slot Width | Single-slot | MXM Module |

| Bus Interface | PCIe 3.0 x16 | MXM-B (3.0) |

| Display Outputs | 2x DVI, 1x mini-HDMI 1.4a | Portable Device Dependent |

| Vulkan Support | 1.4 | Not specified |

| Release Date | 2014-02-17 | 2012-03-21 |

| Successor | GeForce 900 | GeForce 700M |

| Launch MSRP | 119 USD | Not specified |

The Verdict

The data presents a clear hierarchy. The NVIDIA GeForce GTX 750 outperforms the NVIDIA GeForce GTX 675M in the only directly comparable benchmark, taking a 34.1% lead in Geekbench OpenCL. Its compute capabilities are superior, with 1,111.0 GFLOPS FP32 throughput against 952.3 GFLOPS, and it achieves this while drawing nearly half the power (55 W vs 100 W). The GTX 750's pixel rate of 17.36 GPixel/s is 75% higher than the GTX 675M's 9.92 GPixel/s, and its modern Vulkan support adds future-proofing that the older card lacks.

The GTX 675M is not without merits. Its 2 GB VRAM on a 256-bit bus delivers 96.00 GB/s of bandwidth, which is 19.7% higher than the GTX 750's 80.19 GB/s. Its texture rate of 39.68 GTexel/s is also higher, thanks to double the TMUs. For workloads that are heavily texture-bound or that require large frame buffers, the GTX 675M could hold an edge, despite its age. However, these advantages come at the cost of significantly higher power consumption and a much larger die.

Where Each One Wins

The GTX 750 wins in raw compute, pixel throughput, and efficiency. Its 34.1% OpenCL lead is the headline number, and its higher FP32 performance (1,111.0 GFLOPS) makes it the better choice for general-purpose GPU compute tasks. The 17.36 GPixel/s pixel rate suggests stronger fill-rate performance for resolution scaling and effects-heavy rendering. Its lower TDP (55 W) and single-slot design make it suitable for compact desktop builds, and its Vulkan 1.4 support enables modern API features.

The GTX 675M wins in memory capacity and bandwidth. Its 2 GB VRAM is double that of the GTX 750, and its 96.00 GB/s bandwidth is 19.7% higher. The wider 256-bit bus and 64 TMUs make it potentially better suited for texture-heavy scenarios, such as high-resolution texture streaming or older games optimized for wide memory interfaces. Its MXM form factor means it is designed for mobile workstations or laptops, where the 100 W TDP is acceptable. The GTX 675M's 9.92 GPixel/s pixel rate, while lower, is not catastrophic, and its 39.68 GTexel/s texture rate is actually higher than the GTX 750's, indicating a specialization in texture processing. For users constrained to a mobile platform with a need for large VRAM, the GTX 675M remains a functional, if older, option.

DETAILED SPECIFICATIONS

SPECIFICATION
GTX 675M
GTX 750
Core Specs
Shading Units
384
512 +33.3%
Shaders
384
512 +33.3%
TMUs
64
32 -50.0%
ROPs
32
16 -50.0%
SM Count
8
—
Clocks
Base Clock
—
1020 MHz
Boost Clock
—
1085 MHz
GPU Clock
620 MHz
—
Shader Clock
1240 MHz
—
Memory Clock
750 MHz 3 Gbps effective
1253 MHz 5 Gbps effective
Memory
Memory Size
2 GB
1024 MB
VRAM (MB)
2,048
1,024 -50.0%
Memory Type
GDDR5
GDDR5
Memory Bus
256 bit
128 bit
Bandwidth
96.00 GB/s
80.19 GB/s
Cache
L1 Cache
64 KB (per SM)
64 KB (per SMM)
L2 Cache
512 KB
2 MB
Performance
Pixel Rate
9.920 GPixel/s
17.36 GPixel/s
Texture Rate
39.68 GTexel/s
34.72 GTexel/s
FP32 (TFLOPS)
952.3 GFLOPS
1,111.0 GFLOPS
FP64 (TFLOPS)
79.36 GFLOPS (1:12)
34.72 GFLOPS (1:32)
Power
TDP
100 W
55 W
TDP (W)
100
55 -45.0%
Suggested PSU
—
250 W
Power Connectors
None
None
Architecture
Architecture
Fermi 2.0
Maxwell
GPU Name
GF114
GM107
Generation
GeForce 600M
GeForce 700
Process Size
40 nm
28 nm
Transistors
1,950 million
1,870 million
Die Size
332 mm²
148 mm²
Foundry
TSMC
TSMC
Density
5.9M / mm²
12.6M / mm²
API Support
DirectX
12 (11_0)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
—
1.4
OpenCL
1.1
3.0
CUDA
2.1
5.0
Shader Model
5.1
6.7 (5.1)
Physical
Slot Width
MXM Module
Single-slot
Length
—
145 mm 5.7 inches
Outputs
Portable Device Dependent
2x DVI1x mini-HDMI 1.4a
Bus Interface
MXM-B (3.0)
PCIe 3.0 x16
Other
Launch Price
—
119 USD
Production
End-of-life
End-of-life
Predecessor
GeForce 500M
GeForce 600
Successor
GeForce 700M
GeForce 900
View GeForce GTX 675M Details View GeForce GTX 750 Details