NVIDIA GeForce GTX 965M

NVIDIA GeForce GTX 965M

NVIDIA GeForce GTX 965M: A Retrospective and Relevance in 2025

Understanding who might still find this mobile graphics card useful a decade after its release.

Introduction

Released in 2015, the NVIDIA GeForce GTX 965M became one of the popular solutions for gaming laptops in the mid-2010s. However, by 2025, its capabilities appear modest against the backdrop of modern GPUs. Nevertheless, this card is still found in older devices and on the second-hand market. Let’s explore its strengths today and what tasks it is suitable for.


Architecture and Key Features

Maxwell Architecture: Modest, Yet Effective

The GTX 965M is built on the Maxwell architecture (GM204) and manufactured using a 28nm process. Unlike modern cards with 5nm chips, it does not support ray tracing (RTX), DLSS, or FidelityFX. Its key features include:

- Dynamic Super Resolution (DSR): Enhancing image details by rendering at a high resolution and then scaling down.

- BatteryBoost: Optimizing power consumption for laptops.

- ShadowPlay: Game recording with minimal performance loss.

Despite lacking AI accelerators, Maxwell provides good energy efficiency for its time.


Memory: Limitations for Modern Tasks

GDDR5 and 128-bit Bus

The GTX 965M comes equipped with either 2GB or 4GB of GDDR5 memory with a bandwidth of up to 80GB/s (128-bit bus). For games in 2025, this is insufficient:

- 2GB capacity: Critical even for 1080p in titles like Cyberpunk 2077 or Starfield (minimum requirements — 4GB).

- GDDR5 speed: Lags behind modern GDDR6X standards (up to 1TB/s on the RTX 4090).

This card is suitable only for older games or low settings in less demanding titles.


Gaming Performance: Modest Results

1080p – The Limit of Capabilities

In 2025, the GTX 965M manages only a limited set of tasks:

- CS:2, Dota 2: 50-60 FPS on medium settings (1080p).

- GTA V: 40-45 FPS (high settings, 1080p).

- Fortnite: 30-35 FPS (epic settings, 1080p).

For games with ray tracing support (like Alan Wake 2), the card is unfit. Resolutions of 1440p and 4K are unattainable even at low settings.


Professional Tasks: Basic Capabilities

CUDA at Minimums

With 1024 CUDA cores, the GTX 965M falls short even compared to budget modern GPUs (such as RTX 3050 with 2560 cores). However, it is still relevant for simple tasks:

- Video Editing: Rendering in Adobe Premiere Pro is possible, but slow (3-4 times longer than on RTX 4060).

- 3D Modeling: Working in Blender with simple scenes, but without RTX acceleration.

- Scientific Calculations: Limited support for OpenCL and CUDA in MATLAB or Python libraries.

For professional use, the card is suitable only as a temporary solution.


Power Consumption and Heat Dissipation

TDP 50–60W: A Plus for Laptops

Low power consumption is the main advantage of the GTX 965M in 2025. However, this is a double-edged sword:

- Cooling: In older laptops, overheating may occur due to wear on the cooling system.

- Recommendations:

- Regular dust cleaning.

- Use of cooling pads.

- Change of thermal paste every 1-2 years.

The card is not intended for desktop PCs — it is strictly a mobile solution.


Comparison with Competitors

Against AMD Radeon and Modern Budget GPUs

In its time, the GTX 965M competed with the AMD Radeon R9 M380. In 2025, it is easily overshadowed by even budget newcomers:

- NVIDIA RTX 2050 (laptop): +120% performance, support for DLSS and RTX.

- AMD Radeon RX 6500M: +90% speed, 4GB GDDR6.

- Intel Arc A370M: Superiority in Vulkan projects and support for XeSS.

Even used laptops with GTX 1650 (2019) offer better performance for the same $300–400.


Practical Tips

For Owners of Older Devices

1. Power Supply: An original 90–120W adapter is a must.

2. Compatibility: The card works only in laptops with PCIe 3.0 x16.

3. Drivers: Official driver support from NVIDIA ceased in 2021. Use modified versions (e.g., through the NVCleanstall project) or Windows updates.


Pros and Cons

Pros:

- Low power consumption.

- Support for DirectX 12 (basic features).

- Quiet operation in laptops with functioning cooling.

Cons:

- Obsolete architecture without support for RTX/DLSS.

- Insufficient memory for modern games.

- Lack of official driver support.


Final Conclusion: Who Is the GTX 965M Suitable For in 2025?

This graphics card is a choice for:

1. Owners of Older Laptops wishing to extend their lifespan for basic tasks (office work, web surfing, older games).

2. Retro Gaming Enthusiasts playing titles from the 2010s.

3. Limited Professional Use (e.g., training in graphic editing software).

However, buying a laptop with a GTX 965M new in 2025 is pointless — the same money ($400–500) is better spent on used devices with GTX 1650 or RTX 2050.


Conclusion

The NVIDIA GeForce GTX 965M exemplifies how quickly technology can become outdated. Today, it maintains a niche audience, but for most users, its potential has been exhausted. If you're not ready to upgrade, this card may still serve you… but don't expect miracles from it.

Basic

Label Name
NVIDIA
Platform
Mobile
Launch Date
January 2016
Model Name
GeForce GTX 965M
Generation
GeForce 900M
Base Clock
935MHz
Boost Clock
1150MHz
Bus Interface
MXM-B (3.0)
Transistors
2,940 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.
64
Foundry
TSMC
Process Size
28 nm
Architecture
Maxwell 2.0

Memory Specifications

Memory Size
2GB
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.
128bit
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.
80.19 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.
36.80 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.
73.60 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.
73.60 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.
2.402 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.
1024
L1 Cache
48 KB (per SMM)
L2 Cache
1024KB
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
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.
32

Benchmarks

FP32 (float)
Score
2.402 TFLOPS
3DMark Time Spy
Score
1855
Blender
Score
136
OctaneBench
Score
31
Vulkan
Score
15551
OpenCL
Score
13849
Hashcat
Score
93515 H/s

Compared to Other GPU

FP32 (float) / TFLOPS
2.446 +1.8%
2.35 -2.2%
2.285 -4.9%
3DMark Time Spy
5182 +179.4%
3906 +110.6%
2755 +48.5%
Blender
1506.77 +1007.9%
848 +523.5%
194 +42.6%
OctaneBench
123 +296.8%
69 +122.6%
Vulkan
98446 +533.1%
69708 +348.3%
40716 +161.8%
18660 +20%
OpenCL
62821 +353.6%
38843 +180.5%
21442 +54.8%
884 -93.6%
Hashcat / H/s
102283 +9.4%
100059 +7%
93161 -0.4%
85096 -9%