NVIDIA GeForce GTX 675MX vs NVIDIA GeForce GTX 760 Comparison

NVIDIA
GEFORCE

NVIDIA GeForce GTX 675MX

CORE STATE GK104
VRAM 2 GB
CLOCK SPEED —
TDP 100 W
BUS WIDTH 256 bit
ARCHITECTURE Kepler
nm
PROCESS 28 nm
LAUNCH DATE 2012
VS
NVIDIA
GEFORCE

GeForce GTX 760

CORE STATE GK104
VRAM 2 GB
CLOCK SPEED 1032 MHz
TDP 170 W
BUS WIDTH 256 bit
ARCHITECTURE Kepler
nm
PROCESS 28 nm
LAUNCH DATE 2013

PERFORMANCE BENCHMARKS

geekbench_metal
6,131
4,524
geekbench_opencl
10,723
11,299
geekbench_vulkan
N/A
12,551

Analysis: NVIDIA GeForce GTX 675MX vs NVIDIA GeForce GTX 760

Head-to-Head Benchmarks

The recorded data for these two Kepler-based NVIDIA parts presents a fascinating split, with each card claiming a victory in one of the two shared benchmark tests. In the Geekbench Metal test, the NVIDIA GeForce GTX 675MX emerges as the clear winner, posting a score of 6131 against the GTX 760's 4524. This represents a substantial 26.2% advantage for the mobile-oriented 675MX, a surprising result given the desktop pedigree of the GTX 760. The delta is significant enough to suggest architectural tuning or driver optimizations that favor the Metal API on the 675MX, despite its lower overall compute specifications.

Conversely, the Geekbench OpenCL test tells a different story. Here, the NVIDIA GeForce GTX 760 reclaims the lead with a score of 11299, outpacing the GTX 675MX's 10723 by a narrower 5.4% margin. While the win is decisive, the relatively modest percentage gap indicates that the 675MX is no slouch in general-purpose compute workloads. The GTX 760's higher raw throughput, driven by its superior clock speeds and shader count, likely contributes to this edge. The result is a 1-1 tie in the head-to-head, which makes the choice between these two GPUs heavily dependent on the specific application or API in question.

Looking at the broader context, the GTX 760's average benchmark score of 9458 places it in the 46th percentile of all GPUs, while the GTX 675MX sits at 8427, landing in the 43rd percentile. The 760's nearest rivals in the database include the NVIDIA GeForce GTX TITAN BLACK (avg score 9474, a negligible 0.2% difference) and the AMD Radeon R7 M380 (avg score 9313, a 1.6% delta). This shows the GTX 760 is positioned in a tightly contested performance tier. The 675MX, meanwhile, is bracketed by the AMD Radeon 880M (avg score 8436, a 0.1% delta) and the NVIDIA GeForce MX330 (avg score 8458, a 0.4% delta), indicating it sits in a slightly lower, but still competitive, performance class. These percentile rankings reinforce the idea that while the 675MX wins the Metal test, its overall average performance is below the GTX 760, which is the faster card in aggregate.

Architecture Differences

Both the NVIDIA GeForce GTX 760 and the GeForce GTX 675MX share the same fundamental building blocks. They are both built on the GK104 chip, utilizing the Kepler architecture, and are fabricated by TSMC on a 28 nm process node. The transistor count is identical at 3,540 million, and the die size is the same at 294 mm². This results in an identical transistor density of 12.0M / mm². The core architecture is therefore essentially the same, but NVIDIA has made specific cuts and clock adjustments to differentiate the two products.

The most significant divergence lies in the execution resources. The GTX 760 is configured with 1152 shading units, 96 texture mapping units (TMUs), and 32 raster output units (ROPs). The GTX 675MX, in contrast, has fewer active resources: 960 shading units and 80 TMUs, while maintaining the same 32 ROPs. This reduction in shader and texture hardware directly impacts the 675MX's fill rates. The GTX 760 achieves a pixel rate of 24.77 GPixel/s and a texture rate of 99.07 GTexel/s, whereas the 675MX is rated at 13.08 GPixel/s and 52.32 GTexel/s. This is a massive difference in raw rasterization and texturing throughput, largely a consequence of the clock speed disparity.

Clock speeds are another major differentiator. The GTX 760 has a base clock of 980 MHz and a boost clock of 1032 MHz. The GTX 675MX has no listed base or boost clock in the database, but its memory clock is significantly lower at 900 MHz (3.6 Gbps effective) compared to the GTX 760's 1502 MHz (6 Gbps effective). This clock speed difference also explains the gap in FP32 performance: the GTX 760 delivers 2.378 TFLOPS, while the 675MX is rated at 1,255.7 GFLOPS. The memory subsystem is a point of similarity in width (both use a 256-bit bus) and capacity (both have 2 GB of GDDR5), but the 675MX's lower memory clock yields a bandwidth of just 115.2 GB/s compared to the GTX 760's 192.3 GB/s. The 675MX is clearly a power-constrained part, designed for a laptop environment, which necessitates these reductions.

Where Each One Wins

Based strictly on the benchmark data, the NVIDIA GeForce GTX 675MX wins in scenarios that leverage the Metal graphics API. Its 26.2% lead in the Geekbench Metal test is the most pronounced performance gap in this comparison. This suggests that for specific Metal-based applications, particularly those on Apple platforms or in certain professional visualization tools, the 675MX may offer a surprisingly smooth experience despite its lower theoretical specs. This is a curious data point, as it implies the 675MX's driver stack or hardware configuration is exceptionally well-optimized for this particular API, potentially due to its integration in systems where Metal is a primary graphics interface.

The NVIDIA GeForce GTX 760, on the other hand, wins in the Geekbench OpenCL test, which is a more general-purpose compute benchmark. Its 5.4% advantage here, combined with its significantly higher fill rates, texture rate (99.07 GTexel/s vs 52.32 GTexel/s), and memory bandwidth (192.3 GB/s vs 115.2 GB/s), indicates it is the stronger card for traditional rasterization workloads like gaming. The data suggests that for OpenCL-accelerated tasks, the 760's higher shader count and clock speeds provide a tangible benefit. For any use case that is not specifically optimized for Metal, the GTX 760's architecture and specifications point to it being the more capable performer. The 675MX's single win, while impressive, appears to be an exception rather than the rule, given its lower average benchmark score (8427 vs 9458) and lower percentile ranking (43rd vs 46th).

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 760 has a higher average benchmark score of 9458, compared to the NVIDIA GeForce GTX 675MX's 8427.

Q: Is the GTX 675MX better than the GTX 760 in any benchmark?

A: Yes, the GTX 675MX wins the Geekbench Metal test with a score of 6131, which is 26.2% higher than the GTX 760's score of 4524.

Q: How do the memory bandwidths compare?

A: The GTX 760 has a memory bandwidth of 192.3 GB/s, while the GTX 675MX has a lower bandwidth of 115.2 GB/s. Both use a 256-bit bus and 2 GB of GDDR5 memory.

Q: What is the difference in shading units?

A: The GTX 760 has 1152 shading units, while the GTX 675MX has 960 shading units. The GTX 760 also has 96 TMUs compared to the 675MX's 80 TMUs.

Q: Are these GPUs based on the same architecture?

A: Yes, both the GTX 760 and GTX 675MX are based on the GK104 chip and use the Kepler architecture, manufactured on a 28 nm process at TSMC.

Q: Which card has a higher pixel rate?

A: The GTX 760 has a significantly higher pixel rate of 24.77 GPixel/s, compared to the GTX 675MX's 13.08 GPixel/s.

The Verdict

The data paints a clear picture for specific use cases. If the primary workload involves applications that heavily utilize the Metal API, the NVIDIA GeForce GTX 675MX is the surprising winner, based on its 26.2% lead in the Geekbench Metal test. This is a significant margin and should not be ignored for users in that specific ecosystem.

For virtually everything else, the NVIDIA GeForce GTX 760 is the superior choice. It wins the OpenCL benchmark by 5.4%, and its underlying specifications are overwhelmingly in its favor. The 760 offers nearly double the pixel rate (24.77 GPixel/s vs 13.08 GPixel/s), almost double the texture rate (99.07 GTexel/s vs 52.32 GTexel/s), and significantly more memory bandwidth (192.3 GB/s vs 115.2 GB/s). Its higher FP32 compute (2.378 TFLOPS vs 1,255.7 GFLOPS) and higher average benchmark score (9458 vs 8427) confirm it as the more powerful and versatile GPU. The GTX 675MX is a capable mobile part, but the GTX 760 is the better performer for general compute and likely gaming, with the 675MX's Metal result being the only notable exception.

Specification Differences

The following specifications differ between the NVIDIA GeForce GTX 760 and the NVIDIA GeForce GTX 675MX:

  • Generation: The GTX 760 is from the GeForce 700 series, while the GTX 675MX is from the GeForce 600M series.
  • Base Clock: The GTX 760 has a base clock of 980 MHz; the GTX 675MX has no listed base clock.
  • Boost Clock: The GTX 760 has a boost clock of 1032 MHz; the GTX 675MX has no listed boost clock.
  • Memory Clock: The GTX 760 has a memory clock of 1502 MHz (6 Gbps effective), while the GTX 675MX has a memory clock of 900 MHz (3.6 Gbps effective).
  • Memory Bandwidth: The GTX 760 has a bandwidth of 192.3 GB/s, whereas the GTX 675MX has 115.2 GB/s.
  • Shading Units: The GTX 760 has 1152 shading units; the GTX 675MX has 960.
  • Texture Mapping Units (TMUs): The GTX 760 has 96 TMUs; the GTX 675MX has 80.
  • Pixel Rate: The GTX 760 is rated at 24.77 GPixel/s, and the GTX 675MX is rated at 13.08 GPixel/s.
  • Texture Rate: The GTX 760 is rated at 99.07 GTexel/s, and the GTX 675MX is rated at 52.32 GTexel/s.
  • FP32 Performance: The GTX 760 delivers 2.378 TFLOPS, while the GTX 675MX delivers 1,255.7 GFLOPS.
  • TDP: The GTX 760 has a TDP of 170 W, while the GTX 675MX has a TDP of 100 W.
  • Slot Width: The GTX 760 is a Dual-slot card, while the GTX 675MX is an MXM Module.
  • Power Connectors: The GTX 760 requires 2x 6-pin connectors, while the GTX 675MX has none.
  • Suggested PSU: The GTX 760 suggests a 450 W power supply; the GTX 675MX has no suggested PSU listed.
  • Bus Interface: The GTX 760 uses PCIe 3.0 x16, while the GTX 675MX uses MXM-B (3.0).
  • Display Outputs: The GTX 760 has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2; the GTX 675MX outputs are listed as "Portable Device Dependent".
  • Dimensions: The GTX 760 has a length of 241 mm (9.5 inches); the GTX 675MX has no listed dimensions.
  • Release Date: The GTX 760 was released on 2013-06-24, while the GTX 675MX was released on 2012-09-30.
  • Predecessor: The GTX 760's predecessor is GeForce 600, while the GTX 675MX's predecessor is GeForce 500M.
  • Successor: The GTX 760's successor is GeForce 900, while the GTX 675MX's successor is GeForce 700M.
  • Launch MSRP: The GTX 760 had a launch MSRP of 249 USD; the GTX 675MX has no listed launch MSRP.
  • Average Benchmark Score: The GTX 760 has an average score of 9458, and the GTX 675MX has an average score of 8427.
  • Percentile vs All GPUs: The GTX 760 is in the 46th percentile, and the GTX 675MX is in the 43rd percentile.

DETAILED SPECIFICATIONS

SPECIFICATION
GTX 675MX
GTX 760
Core Specs
Shading Units
960
1,152 +20.0%
Shaders
960
1,152 +20.0%
TMUs
80
96 +20.0%
ROPs
32
32 0.0%
Clocks
Base Clock
—
980 MHz
Boost Clock
—
1032 MHz
GPU Clock
654 MHz
—
Memory Clock
900 MHz 3.6 Gbps effective
1502 MHz 6 Gbps effective
Memory
Memory Size
2 GB
2 GB
VRAM (MB)
2,048
2,048 0.0%
Memory Type
GDDR5
GDDR5
Memory Bus
256 bit
256 bit
Bandwidth
115.2 GB/s
192.3 GB/s
Cache
L1 Cache
16 KB (per SMX)
16 KB (per SMX)
L2 Cache
512 KB
512 KB
Performance
Pixel Rate
13.08 GPixel/s
24.77 GPixel/s
Texture Rate
52.32 GTexel/s
99.07 GTexel/s
FP32 (TFLOPS)
1,255.7 GFLOPS
2.378 TFLOPS
FP64 (TFLOPS)
52.32 GFLOPS (1:24)
99.07 GFLOPS (1:24)
Power
TDP
100 W
170 W
TDP (W)
100
170 +70.0%
Suggested PSU
—
450 W
Power Connectors
None
2x 6-pin
Architecture
Architecture
Kepler
Kepler
GPU Name
GK104
GK104
Generation
GeForce 600M
GeForce 700
Process Size
28 nm
28 nm
Transistors
3,540 million
3,540 million
Die Size
294 mm²
294 mm²
Foundry
TSMC
TSMC
Density
12.0M / mm²
12.0M / mm²
API Support
DirectX
12 (11_0)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.2.175
1.2.175
OpenCL
3.0
3.0
CUDA
3.0
3.0
Shader Model
6.5 (5.1)
6.5 (5.1)
Physical
Slot Width
MXM Module
Dual-slot
Length
—
241 mm 9.5 inches
Outputs
Portable Device Dependent
2x DVI1x HDMI 1.4a1x DisplayPort 1.2
Bus Interface
MXM-B (3.0)
PCIe 3.0 x16
Other
Launch Price
—
249 USD
Production
End-of-life
End-of-life
Predecessor
GeForce 500M
GeForce 600
Successor
GeForce 700M
GeForce 900
View GeForce GTX 675MX Details View GeForce GTX 760 Details