GEFORCE

NVIDIA GeForce GTX 670MX

NVIDIA graphics card specifications and benchmark scores

3 GB
VRAM
601
MHz Boost
75W
TDP
192
Bus Width

At a Glance

NVIDIA
VRAM 3 GB
Boost Clock 601 MHz
Shaders 960
Bus Width 192-bit
TDP 75W
Memory Type GDDR5
Architecture Kepler
nm
Process 28 nm
Released Oct 2012

NVIDIA GeForce GTX 670MX Specifications

GeForce GTX 670MX GPU Core

Shader units and compute resources

The NVIDIA GeForce GTX 670MX GPU core specifications define its raw processing power for graphics and compute workloads. Shading units (also called CUDA cores, stream processors, or execution units depending on manufacturer) handle the parallel calculations required for rendering. TMUs (Texture Mapping Units) process texture data, while ROPs (Render Output Units) handle final pixel output. Higher shader counts generally translate to better GPU benchmark performance, especially in demanding games and 3D applications.

Shading Units
960
Shaders
960
TMUs
80
ROPs
24

GTX 670MX Clock Speeds

GPU and memory frequencies

Clock speeds directly impact the GeForce GTX 670MX's performance in GPU benchmarks and real-world gaming. The base clock represents the minimum guaranteed frequency, while the boost clock indicates peak performance under optimal thermal conditions. Memory clock speed affects texture loading and frame buffer operations. The GeForce GTX 670MX by NVIDIA dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.

Base Clock
601 MHz
Base Clock
601 MHz
Boost Clock
601 MHz
Boost Clock
601 MHz
Memory Clock
700 MHz 2.8 Gbps effective
GDDR GDDR 6X 6X

NVIDIA's GeForce GTX 670MX Memory

VRAM capacity and bandwidth

VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The GeForce GTX 670MX's memory capacity determines how well it handles high-resolution textures and multiple displays. Memory bandwidth, measured in GB/s, affects how quickly data moves between the GPU and VRAM. Higher bandwidth improves performance in memory-intensive scenarios like 4K gaming. The memory bus width and type (GDDR6, GDDR6X, HBM) significantly influence overall GPU benchmark scores.

Memory Size
3 GB
VRAM
3,072 MB
Memory Type
GDDR5
VRAM Type
GDDR5
Memory Bus
192 bit
Bus Width
192-bit
Bandwidth
67.20 GB/s

GeForce GTX 670MX by NVIDIA Cache

On-chip cache hierarchy

On-chip cache provides ultra-fast data access for the GTX 670MX, reducing the need to fetch data from slower VRAM. L1 and L2 caches store frequently accessed data close to the compute units. AMD's Infinity Cache (L3) dramatically increases effective bandwidth, improving GPU benchmark performance without requiring wider memory buses. Larger cache sizes help maintain high frame rates in memory-bound scenarios and reduce power consumption by minimizing VRAM accesses.

L1 Cache
16 KB (per SMX)
L2 Cache
384 KB

GTX 670MX Theoretical Performance

Compute and fill rates

Theoretical performance metrics provide a baseline for comparing the NVIDIA GeForce GTX 670MX against other graphics cards. FP32 (single-precision) performance, measured in TFLOPS, indicates compute capability for gaming and general GPU workloads. FP64 (double-precision) matters for scientific computing. Pixel and texture fill rates determine how quickly the GPU can render complex scenes. While real-world GPU benchmark results depend on many factors, these specifications help predict relative performance levels.

FP32 (Float)
1,153.9 GFLOPS
FP64 (Double)
48.08 GFLOPS (1:24)
Pixel Rate
12.02 GPixel/s
Texture Rate
48.08 GTexel/s

Kepler Architecture & Process

Manufacturing and design details

The NVIDIA GeForce GTX 670MX is built on NVIDIA's Kepler architecture, which defines how the GPU processes graphics and compute workloads. The manufacturing process node affects power efficiency, thermal characteristics, and maximum clock speeds. Smaller process nodes pack more transistors into the same die area, enabling higher performance per watt. Understanding the architecture helps predict how the GTX 670MX will perform in GPU benchmarks compared to previous generations.

Architecture
Kepler
GPU Name
GK104
Process Node
28 nm
Foundry
TSMC
Transistors
3,540 million
Die Size
294 mm²
Density
12.0M / mm²

NVIDIA's GeForce GTX 670MX Power & Thermal

TDP and power requirements

Power specifications for the NVIDIA GeForce GTX 670MX determine PSU requirements and thermal management needs. TDP (Thermal Design Power) indicates the heat output under typical loads, guiding cooler selection. Power connector requirements ensure adequate power delivery for stable operation during demanding GPU benchmarks. The suggested PSU wattage accounts for the entire system, not just the graphics card. Efficient power delivery enables the GeForce GTX 670MX to maintain boost clocks without throttling.

TDP
75 W
TDP
75W
Power Connectors
None

GeForce GTX 670MX by NVIDIA Physical & Connectivity

Dimensions and outputs

Physical dimensions of the NVIDIA GeForce GTX 670MX are critical for case compatibility. Card length, height, and slot width determine whether it fits in your chassis. The PCIe interface version affects bandwidth for communication with the CPU. Display outputs define monitor connectivity options, with modern cards supporting multiple high-resolution displays simultaneously. Verify these specifications against your case and motherboard before purchasing to ensure a proper fit.

Bus Interface
PCIe 3.0 x16
Display Outputs
Portable Device Dependent
Display Outputs
Portable Device Dependent

NVIDIA API Support

Graphics and compute APIs

API support determines which games and applications can fully utilize the NVIDIA GeForce GTX 670MX. DirectX 12 Ultimate enables advanced features like ray tracing and variable rate shading. Vulkan provides cross-platform graphics capabilities with low-level hardware access. OpenGL remains important for professional applications and older games. CUDA (NVIDIA) and OpenCL enable GPU compute for video editing, 3D rendering, and scientific applications. Higher API versions unlock newer graphical features in GPU benchmarks and games.

DirectX
12 (11_0)
DirectX
12 (11_0)
OpenGL
4.6
OpenGL
4.6
Vulkan
1.2.175
Vulkan
1.2.175
OpenCL
3.0
CUDA
3.0
Shader Model
6.5 (5.1)

GeForce GTX 670MX Product Information

Release and pricing details

The NVIDIA GeForce GTX 670MX is manufactured by NVIDIA as part of their graphics card lineup. Release date and launch pricing provide context for comparing GPU benchmark results with competing products from the same era. Understanding the product lifecycle helps evaluate whether the GeForce GTX 670MX by NVIDIA represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.

Manufacturer
NVIDIA
Release Date
Oct 2012
Production
End-of-life
Predecessor
GeForce 500M
Successor
GeForce 700M

GeForce GTX 670MX Benchmark Scores

geekbench_openclSource

Geekbench OpenCL tests GPU compute performance using the cross-platform OpenCL API. This shows how NVIDIA GeForce GTX 670MX handles parallel computing tasks like video encoding and scientific simulations. OpenCL is widely supported across different GPU vendors and platforms. Higher scores benefit applications that leverage GPU acceleration for non-graphics workloads.

geekbench_opencl #459 of 643
6,125
2%
Max: 388,405
Compare with other GPUs

geekbench_vulkanSource

Geekbench Vulkan tests GPU compute using the modern low-overhead Vulkan API. This shows how NVIDIA GeForce GTX 670MX performs with next-generation graphics and compute workloads.

geekbench_vulkan #381 of 444
5,316
1%
Max: 376,915

About NVIDIA GeForce GTX 670MX

NVIDIA’s GeForce GTX 670MX is a 28 nm Kepler-based mobile GPU (chip GK104) released in late 2012, now end-of-life, with 3 GB of GDDR5 memory on a 192-bit bus and a 75 W TDP. Its benchmark profile places it near the bottom quartile of all GPUs (33rd percentile), with an average benchmark score of 5742. This is not a card for modern high-end gaming; it is a legacy part whose remaining relevance hinges on older titles and light workloads.

Benchmark Performance

The GTX 670MX’s average benchmark score of 5742 is derived from Geekbench OpenCL (6167) and Vulkan (5316) results. The gap between those two scores is notable: the OpenCL result is roughly 16% higher than Vulkan, suggesting the architecture favors compute-style APIs over modern graphics abstraction layers. In practical terms, this means the GPU will perform relatively better in applications that use OpenCL (e.g., some productivity and media tools) than in Vulkan-based games.

Against its nearest rivals, the GTX 670MX sits in a dead heat. It is 0.2% ahead of the NVIDIA GeForce GTX 550 Ti (average score 5731), 0.3% ahead of the NVIDIA Quadro K4000 (5723), and 0.5% ahead of the Intel Iris Pro Graphics P6300 (5712). Conversely, it trails the Intel UHD Graphics P630 (5760) by 0.3%. These deltas are within noise—no rival in this cluster has a meaningful performance advantage. The data shows a tight pack of four GPUs spanning just 0.8% total spread (from 5712 to 5760), meaning the GTX 670MX is essentially interchangeable with any of them in raw benchmark terms.

The 33rd percentile ranking underscores a broader context: this GPU is far below modern entry-level discrete parts. For comparison, a contemporary mid-range card would score several times higher, but the FACT PACK provides no such numbers—only the relative positions of these four rivals. What the percentile does tell you is that 67% of all GPUs in the benchmark database outperform the GTX 670MX. That is a sobering figure for anyone considering this part for current-generation gaming. The FP32 throughput of 1,153.9 GFLOPS, while respectable for 2012, is dwarfed by today’s offerings, and the fixed 601 MHz base and boost clocks mean no dynamic headroom—the card runs flat-out at all times.

Memory Subsystem

The GTX 670MX comes with 3 GB of GDDR5 memory on a 192-bit bus, yielding a bandwidth of 67.20 GB/s. Memory speed is 700 MHz (2.8 Gbps effective). The 192-bit bus is narrower than what many desktop counterparts used at the time, but the 3 GB capacity was generous for a mobile part in its era.

What does this mean for high resolutions? At 1080p, the 67.20 GB/s bandwidth is sufficient for older titles with modest texture budgets, but it becomes a bottleneck in modern games that stream large assets. At 1440p or 4K, the memory bandwidth will constrain performance regardless of the 3 GB capacity—the card simply cannot feed the shading units fast enough to maintain playable frame rates in demanding scenes. The 24 ROPs and 80 TMUs further limit fill-rate-heavy workloads; pixel rate is 12.02 GPixel/s and texture rate is 48.08 GTexel/s. These numbers indicate that the card is best suited to 1080p with reduced settings, not high-resolution gaming.

The 3 GB VRAM is a double-edged sword. It is enough to hold many textures at 1080p, but the bandwidth shortfall means that even when assets fit in memory, the transfer speed will cap performance. In benchmarks, this shows up as a GPU that scores close to the Intel UHD Graphics P630—an integrated part—because memory bandwidth is often the limiting factor in modern workloads. The 192-bit bus width is the primary culprit; a wider bus would have helped, but that is not what the hardware offers.

Who Should Consider It

Benchmark results indicate the GTX 670MX is viable for 1080p gaming at low-to-medium settings in titles from its release era (circa 2012). The 3 GB VRAM allows for moderate texture quality, but the 67.20 GB/s bandwidth and 1,153.9 GFLOPS FP32 throughput will not sustain high frame rates in recent AAA releases. For esports titles with light graphical demands, the card may still deliver acceptable performance, but the 33rd percentile ranking suggests it will struggle with anything beyond that.

Users with older libraries—games released before 2015, for instance—will find the GTX 670MX serviceable. The OpenCL score of 6167 indicates reasonable compute capability for media encoding or light GPU-accelerated tasks. However, for anyone considering this card for current-generation games, the data is clear: it sits within 0.5% of integrated graphics like the Intel UHD Graphics P630, so you would gain little over a modern iGPU in many workloads. The Vulkan score of 5316, being lower than OpenCL, signals that Vulkan-based titles—common in modern engines—will underperform relative to expectations.

How It Compares

vs. NVIDIA GeForce GTX 550 Ti: The GTX 670MX is 0.2% faster in average benchmark score (5742 vs. 5731). These are effectively identical performers. Both are legacy parts, but the GTX 670MX has double the VRAM (3 GB vs. the 550 Ti’s typical 1 GB), which helps with texture-heavy scenes even if raw compute is a wash. In practice, the 670MX’s newer Kepler architecture (vs. Fermi) may handle tessellation better, though the benchmark delta does not reflect that.

vs. Intel UHD Graphics P630: The GTX 670MX trails by 0.3% (5742 vs. 5760). This is a remarkable parity: a discrete mobile GPU from 2012 matches a modern integrated solution in average score. The GTX 670MX still has dedicated VRAM and higher raw bandwidth, but the P630 benefits from newer architecture and driver optimizations. For everyday computing or light gaming, the difference is negligible in benchmarks.

vs. NVIDIA Quadro K4000: The GTX 670MX is 0.3% ahead (5742 vs. 5723). The Quadro K4000 is a workstation card with the same Kepler generation, but the 670MX’s higher memory capacity (3 GB vs. 2 GB) and slightly better average score give it an edge in general benchmarks. For professional applications that use OpenCL, the 670MX’s 6167 score suggests it could hold its own, but driver certification and stability are separate considerations not covered by the data.

vs. Intel Iris Pro Graphics P6300: The GTX 670MX leads by 0.5% (5742 vs. 5712). This is the largest delta among its nearest rivals but still small. The Iris Pro part is an integrated GPU with embedded DRAM on some variants, yet the 670MX’s discrete memory and higher texture rate (48.08 GTexel/s) give it a slight edge. Neither is suitable for modern high-end gaming; this is a contest of legacy hardware.

FAQ

Q: Is the GTX 670MX capable of running modern games?

A: Benchmark results place it at the 33rd percentile of all GPUs, with an average score of 5742—within 0.3% of the Intel UHD Graphics P630. It will run older titles at 1080p with reduced settings, but modern AAA games are likely to be unplayable at acceptable frame rates.

Q: How much VRAM does it have, and is it enough?

A: It has 3 GB of GDDR5 memory on a 192-bit bus with 67.20 GB/s bandwidth. The capacity is sufficient for 1080p textures in many games, but the bandwidth is a bottleneck at higher resolutions or with modern texture streaming.

Q: What is the difference between its OpenCL and Vulkan scores?

A: The Geekbench OpenCL score is 6167, while the Vulkan score is 5316—a 16% gap. This indicates the GPU performs better in compute workloads using OpenCL than in Vulkan-based graphics, which is typical for older architectures without native Vulkan optimizations.

Q: How does it compare to the GTX 550 Ti?

A: The GTX 670MX is 0.2% faster in average benchmark score (5742 vs. 5731). Both are near-identical in raw performance, but the 670MX has 3 GB VRAM compared to the 550 Ti’s typical 1 GB, which can help with texture-heavy workloads.

Q: What API support does it offer?

A: It supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The DirectX 12 support is at the 11_0 feature level, meaning it lacks later DX12 features like ray tracing or mesh shaders.

Q: Is this card better than modern integrated graphics?

A: The data shows it is 0.3% slower than the Intel UHD Graphics P630 and 0.5% faster than the Intel Iris Pro Graphics P6300. In practical terms, it is comparable to modern iGPUs, though it has dedicated VRAM and a higher TDP.

Power and Cooling

The GTX 670MX has a TDP of 75 W, which is modest for a discrete GPU. It requires no external power connectors—the card draws power solely from the PCIe slot (PCIe 3.0 x16 interface). This makes it suitable for systems without auxiliary PCIe power cables, a common scenario for older laptops or small form-factor builds. The lack of a suggested PSU rating in the data indicates that power supply sizing is not a concern for this part; any system that can supply 75 W to a slot will suffice.

Cooling is portable-device dependent, as the display outputs are listed as "Portable Device Dependent"—this is a mobile GPU. The 28 nm process node and 75 W TDP mean that a capable air cooler in a laptop chassis should handle it, but sustained loads will generate heat. The fixed 601 MHz clock (no boost) means thermal throttling is less of a risk than on variable-clock parts, but it also means no performance headroom. For a desktop user retrofitting this into a system, a simple heatsink with a fan would suffice; for laptop users, the original cooling solution is the only option. The 3,540 million transistors on a 294 mm² die (transistor density 12.0M/mm²) are concentrated in a power envelope that is easy to manage, but the card’s age means thermal paste may have degraded in existing units.

The AMD Equivalent of GeForce GTX 670MX

Looking for a similar graphics card from AMD? The AMD Radeon RX 480 offers comparable performance and features in the AMD lineup.

AMD Radeon RX 480

AMD • 8 GB VRAM

View Specs Compare

Popular NVIDIA GeForce GTX 670MX Comparisons

See how the GeForce GTX 670MX stacks up against similar graphics cards from the same generation and competing brands.

Compare GeForce GTX 670MX with Other GPUs

Select another GPU to compare specifications and benchmarks side-by-side.

Browse GPUs