Home / NVIDIA / NVIDIA GeForce GTX 670MX: Performance and Specs

NVIDIA GeForce GTX 670MX

NVIDIA GeForce GTX 670MX: Architectural Relic or Budget Solution in 2025?

Review of an outdated GPU for enthusiasts and owners of old systems

1. Architecture and Key Features

Kepler Architecture: Technologies from 2012

The NVIDIA GeForce GTX 670MX, released in 2012, is built on the Kepler architecture (GK106 chip) using a 28 nm manufacturing process. This is a mobile version designed for laptops, focusing on a balance between performance and energy efficiency. Unlike modern GPUs like the RTX 40 series, it lacks key technologies:

- RTX (ray tracing) and DLSS (artificial intelligence for upscaling) — these didn’t exist during the Kepler era.

- FidelityFX (AMD's optimization package) is also not supported as it's a competitor's solution.

Key features include support for DirectX 11, OpenGL 4.5, and PhysX, which seem archaic in 2025. The card is suited for basic tasks and games from the past decade.

2. Memory: Modest Specifications by Modern Standards

- Type and Volume: 3 GB GDDR5 — in 2025, this is insufficient even for less demanding games. Modern titles require at least 6–8 GB.

- Bus and Bandwidth: The 192-bit bus provides up to 96 GB/s. In comparison, even the budget RTX 3050 (128-bit, GDDR6) offers 224 GB/s.

- Impact on Performance: The lack of memory capacity and low bandwidth limits performance at resolutions above 1080p. In games with HD textures, FPS drops can occur due to VRAM overload.

3. Gaming Performance: Nostalgia for the Past

In 2025, the GTX 670MX is only suitable for older titles and indie games. Examples of average FPS (on low settings, 1080p):

- CS:GO — 60–80 FPS (but in competitive modes on low settings).

- The Witcher 3 — 25–30 FPS (at minimum settings).

- GTA V — 35–40 FPS (medium settings).

Resolution Support:

- 1080p — acceptable for games up to 2015.

- 1440p and 4K — not recommended even for video playback due to lack of hardware decoding for modern codecs (AV1, VP9).

Ray Tracing: Not supported. For comparison, even the GTX 16 series (2019) lacks RT cores.

4. Professional Tasks: Very Limited Potential

- Video Editing: Editing in DaVinci Resolve or Premiere Pro is only feasible for 1080p projects. Rendering will take 3–5 times longer than on modern GPUs.

- 3D Modeling: In Blender or Maya, the card can handle simple scenes, but Cycles on CUDA (supported by 960 cores) will work very slowly.

- Scientific Calculations: CUDA and OpenCL are functional, but less efficient than modern integrated solutions (e.g., Apple M3).

For professionals, the GTX 670MX in 2025 is a relic.

5. Power Consumption and Heat Dissipation

- TDP: 75 W — a modest figure even for 2010s laptops.

- Cooling: Owners of older laptops should regularly clean the fans and replace the thermal paste. Under load, temperatures can reach 85–90°C.

- PC Cases: For the desktop version (if used via an external eGPU enclosure), a power supply of at least 300 W is required.

6. Comparison with Competitors

In its niche (mobile GPUs from 2012-2013), the GTX 670MX competed with the AMD Radeon HD 8970M (2 GB GDDR5):

- Performance: In games, NVIDIA had a 10–15% advantage due to driver optimizations.

- Energy Efficiency: AMD ran hotter (TDP 100 W).

- Modern Analogues: By 2025, both cards are hopelessly outdated. The budget GTX 1650 (4 GB GDDR6) is 3–4 times faster.

7. Practical Advice for Users

- Power Supply: For laptops — use the original adapter. For eGPU — a PSU of 300 W or more.

- Compatibility: Only systems with PCIe 2.0/3.0. Windows 10/11 (drivers updated until 2021).

- Drivers: Use the latest version from NVIDIA (v471.xx) — support has been discontinued.

8. Pros and Cons

Pros:

- Low price on the secondary market ($20–40).

- CUDA support for older professional applications.

Cons:

- Lack of memory and performance for modern tasks.

- No support for new technologies (RTX, DLSS, AV1).

- High risk of overheating in older laptops.

9. Final Conclusion: Who is the GTX 670MX Suitable for in 2025?

This graphics card is a choice for:

1. Owners of old laptops who want to extend the life of their devices for basic tasks (office work, browsing, YouTube).

2. Retro gaming enthusiasts reminiscing about projects from the 2000s to 2010s.

3. Budget-conscious users: If a temporary GPU is needed for a PC without integrated graphics.

Why it’s not worth buying for newcomers? Even budget Intel Arc A380 or AMD RX 6400 (from $120) offer 5–7 times the performance, support for modern APIs and technologies. The GTX 670MX in 2025 is a museum piece, not a working tool.

Basic

Label Name

NVIDIA

Platform

Mobile

Launch Date

October 2012

Model Name

GeForce GTX 670MX

Generation

GeForce 600M

Bus Interface

PCIe 3.0 x16

Transistors

3,540 million

TMUs

Texture Mapping Units (TMUs) serve as components of the GPU, which are capable of rotating, scaling, and distorting binary images, and then placing them as textures onto any plane of a given 3D model. This process is called texture mapping.

Foundry

TSMC

Process Size

28 nm

Architecture

Kepler

Memory Specifications

Memory Size

3GB

Memory Type

GDDR5

Memory Bus

The memory bus width refers to the number of bits of data that the video memory can transfer within a single clock cycle. The larger the bus width, the greater the amount of data that can be transmitted instantaneously, making it one of the crucial parameters of video memory. The memory bandwidth is calculated as: Memory Bandwidth = Memory Frequency x Memory Bus Width / 8. Therefore, when the memory frequencies are similar, the memory bus width will determine the size of the memory bandwidth.

192bit

Memory Clock

700MHz

Bandwidth

Memory bandwidth refers to the data transfer rate between the graphics chip and the video memory. It is measured in bytes per second, and the formula to calculate it is: memory bandwidth = working frequency × memory bus width / 8 bits.

67.20 GB/s

Theoretical Performance

Pixel Rate

Pixel fill rate refers to the number of pixels a graphics processing unit (GPU) can render per second, measured in MPixels/s (million pixels per second) or GPixels/s (billion pixels per second). It is the most commonly used metric to evaluate the pixel processing performance of a graphics card.

12.02 GPixel/s

Texture Rate

Texture fill rate refers to the number of texture map elements (texels) that a GPU can map to pixels in a single second.

48.08 GTexel/s

FP64 (double)

An important metric for measuring GPU performance is floating-point computing capability. Double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy, while single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.

48.08 GFLOPS

FP32 (float)

An important metric for measuring GPU performance is floating-point computing capability. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.

1.177 TFLOPS

Miscellaneous

Shading Units

The most fundamental processing unit is the Streaming Processor (SP), where specific instructions and tasks are executed. GPUs perform parallel computing, which means multiple SPs work simultaneously to process tasks.

960

L1 Cache

16 KB (per SMX)

L2 Cache

384KB

TDP

75W

Vulkan Version

Vulkan is a cross-platform graphics and compute API by Khronos Group, offering high performance and low CPU overhead. It lets developers control the GPU directly, reduces rendering overhead, and supports multi-threading and multi-core processors.

1.1

OpenCL Version

3.0

OpenGL

4.6

DirectX

12 (11_0)

CUDA

3.0

Shader Model

5.1

ROPs

The Raster Operations Pipeline (ROPs) is primarily responsible for handling lighting and reflection calculations in games, as well as managing effects like anti-aliasing (AA), high resolution, smoke, and fire. The more demanding the anti-aliasing and lighting effects in a game, the higher the performance requirements for the ROPs; otherwise, it may result in a sharp drop in frame rate.

Benchmarks

FP32 (float)

Score

1.177 TFLOPS

Hashcat

Score

19727 H/s

Compared to Other GPU

FP32 (float) / TFLOPS

Radeon RX 550

1.235 +4.9%

Quadro P600

1.22 +3.7%

GeForce GTX 670MX

1.177

Quadro P520 Mobile

1.17 -0.6%

FirePro M7820

1.142 -3%

Hashcat / H/s

GeForce GTX 680M

23908 +21.2%

GeForce GTX 675MX

21953 +11.3%

GeForce GTX 670MX

19727

Radeon R9 M270X

18293 -7.3%

GeForce GTX 650 Ti

17544 -11.1%