NVIDIA GeForce GTX 970M

NVIDIA GeForce GTX 970M

NVIDIA GeForce GTX 970M: A Review of an Outdated but Still Relevant Solution for Budget-Conscious Users

April 2025

Despite being released over ten years ago, the NVIDIA GeForce GTX 970M continues to attract attention from users seeking budget-friendly solutions for gaming and basic work tasks. In this article, we will discuss how justified its use is in 2025 and who might still find it useful.


Architecture and Key Features

Foundation: Maxwell and 28 nm

The GTX 970M is built on the Maxwell architecture, which was groundbreaking at the time due to improved energy efficiency. The card is manufactured using a 28-nm process, which was standard in 2014 but now looks archaic compared to 5-nm and 6-nm chips.

Unique Features (and Their Absence)

The GTX 970M does not support modern technologies like RTX ray tracing, DLSS, or FidelityFX. Its strengths lie in stability and tried-and-true features:

- NVIDIA Optimus – automatic switching between integrated and discrete graphics to save energy in laptops.

- ShadowPlay – recording gameplay with minimal impact on performance.

- PhysX – accelerating physical calculations in games (such as in Borderlands 2 or Batman: Arkham Knight).


Memory: Speed and Limitations

Technical Specifications

- Memory Type: GDDR5.

- Size: 3 GB (rarely 6 GB in modified versions).

- Bus: 192-bit.

- Bandwidth: 120 GB/s.

Impact on Performance

For games in 2025, 3 GB of video memory is a serious limitation. For example, in Hogwarts Legacy at medium settings (1080p), FPS drops can occur due to insufficient VRAM. However, in less demanding titles like CS2 or Dota 2, the card performs confidently (60+ FPS).


Gaming Performance: What Can You Run?

1080p (Full HD)

- The Witcher 3: 40–45 FPS at medium settings.

- Grand Theft Auto V: 55–60 FPS at high settings.

- Cyberpunk 2077: 25–30 FPS at low settings (without ray tracing).

1440p and 4K

At 1440p, comfortable gaming is only possible in older titles (Skyrim, Overwatch). 4K is impractical: even in League of Legends, FPS hardly reaches 40.


Professional Tasks: Not for Intensive Projects

Thanks to CUDA cores (1280 in total), the GTX 970M can speed up rendering in Blender or Adobe Premiere Pro, but its 3 GB of memory becomes a bottleneck. Examples include:

- Rendering a 10-minute video at 1080p in Premiere Pro: ~15–20 minutes.

- Simple 3D models in AutoCAD: smooth operation, but complex scenes cause stuttering.

For scientific calculations (e.g., in MATLAB), the card is only suitable for educational tasks.


Power Consumption and Heat Generation

- TDP: 81 W (for the mobile version).

- Cooling Recommendations:

- In laptops: regular cleaning of fans and replacing thermal paste.

- In desktop PCs (when using through an external enclosure): a case with at least two fans.


Comparison with Competitors

Key competitors from 2014–2015:

- AMD Radeon R9 M395X: Comparable performance but higher power consumption (100 W TDP). In Vulkan games (Doom Eternal), AMD shows better results.

- NVIDIA GTX 980M: 15–20% more powerful but more expensive even in 2025 (new units from $300).

In 2025, even budget newcomers like Intel Arc A580 or AMD Radeon RX 6500 XT outperform the GTX 970M by 2–3 times.


Practical Advice

Power Supply

For PCs with the GTX 970M (external chassis or desktop equivalent), a 450 W power supply is sufficient (e.g., Corsair CX450).

Compatibility

- Platforms: Works with motherboards supporting PCIe 3.0.

- Drivers: Official support ended in 2022, but the latest versions (v473.xx) are stable.


Pros and Cons

Pros:

- Low power consumption for the performance level.

- CUDA support for basic work tasks.

- Availability on the secondary market (price $50–80 for used).

Cons:

- Only 3 GB of video memory.

- No support for modern technologies (DLSS, RTX).

- Outdated drivers.


Final Conclusion: Who is the GTX 970M For?

This graphics card is an option for:

1. Owners of old laptops looking to extend their lifespan for office applications and undemanding games.

2. Students studying 3D modeling or basic video editing.

3. Budget-conscious gamers ready to play indie projects or older AAA titles.

However, for modern games at high settings or professional tasks, it’s better to consider newer solutions—such as the NVIDIA RTX 3050 or AMD RX 6600.


In 2025, the GTX 970M is an example of a "workhorse" that, despite its age, can still prove useful. However, it should be chosen only when sticking to a strict budget or for specific use cases.

Basic

Label Name
NVIDIA
Platform
Mobile
Launch Date
October 2014
Model Name
GeForce GTX 970M
Generation
GeForce 900M
Base Clock
924MHz
Boost Clock
1038MHz
Bus Interface
MXM-B (3.0)
Transistors
5,200 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.
80
Foundry
TSMC
Process Size
28 nm
Architecture
Maxwell 2.0

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
1253MHz
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.
120.3 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.
49.82 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.
83.04 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.
83.04 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.
2.71 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.
1280
L1 Cache
48 KB (per SMM)
L2 Cache
1536KB
TDP
Unknown
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.
1.3
OpenCL Version
3.0
OpenGL
4.6
DirectX
12 (12_1)
CUDA
5.2
Power Connectors
None
Shader Model
6.7 (6.4)
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.
48

Benchmarks

FP32 (float)
Score
2.71 TFLOPS
3DMark Time Spy
Score
2237
Blender
Score
172
OctaneBench
Score
53
Vulkan
Score
19677
OpenCL
Score
18130
Hashcat
Score
102283 H/s

Compared to Other GPU

FP32 (float) / TFLOPS
2.81 +3.7%
2.55 -5.9%
3DMark Time Spy
5182 +131.6%
3906 +74.6%
2755 +23.2%
Blender
1497 +770.3%
194 +12.8%
OctaneBench
123 +132.1%
69 +30.2%
Vulkan
98446 +400.3%
69708 +254.3%
40716 +106.9%
5522 -71.9%
OpenCL
62821 +246.5%
38843 +114.2%
21442 +18.3%
884 -95.1%
Hashcat / H/s
112347 +9.8%
105378 +3%
100059 -2.2%
93515 -8.6%