NVIDIA GeForce GTX 960

NVIDIA GeForce GTX 960

NVIDIA GeForce GTX 960 in 2025: Is It Worth Buying?

Analyzing the legendary graphics card a decade after its release


Introduction

The NVIDIA GeForce GTX 960, released in 2015, became a popular choice for gamers due to its balance between price and performance. But in 2025, when the market is filled with cards that support ray tracing and AI technologies, is it still relevant? Let's determine who might find this model useful today.


Architecture and Key Features

Maxwell Architecture

The GTX 960 is built on Maxwell architecture (GM206) and manufactured using a 28nm process. Compared to modern 5nm chips (like those in the RTX 40 series), this seems archaic, but in its time, Maxwell was renowned for its energy efficiency.

Lack of RTX and DLSS

The card does not support ray tracing (RTX) or DLSS—NVIDIA's key technologies for modern gaming. Instead, it features proprietary technologies from the 2010s:

- MFAA (Multi-Frame Anti-Aliasing, resource-efficient smoothing);

- VXAO (improved shading in games like Rise of the Tomb Raider).

Conclusion: The GTX 960 is a purely rasterization card, aimed at DirectX 12 Feature Level 11_0. Compatibility with Vulkan and OpenGL is limited to outdated versions.


Memory: A Weak Point?

- Type and Size: GDDR5, 2GB (4GB models are rare).

- Bus Width: 128-bit, providing a bandwidth of 112GB/s (compared to 448GB/s for the budget RTX 3050 with GDDR6).

Problems in 2025:

- 2GB of video memory is insufficient even for minimum settings in games like Cyberpunk 2077 or Starfield.

- 4GB models perform slightly better, but experience FPS drops at 1440p and 4K due to memory limitations.

Advice: For comfortable gameplay at 1080p, stick to games released before 2020 (e.g., The Witcher 3, GTA V).


Gaming Performance: The Numbers

Testing in 2025 reveals modest results:

- 1080p / Low Settings:

- Fortnite: 45–55 FPS (without DLSS support);

- Apex Legends: 50–60 FPS;

- Elden Ring: 30–35 FPS (with frequent stutters).

- 1440p and 4K: Not recommended—FPS rarely exceeds 25 frames.

Ray Tracing: No hardware support. Software solutions (like Reshade) reduce performance by 40–50%.


Professional Tasks: Not the Best Choice

- Video Editing: In Premiere Pro or DaVinci Resolve, the card can handle 1080p editing, but rendering 4K will take 3–4 times longer than on the RTX 3050.

- 3D Modeling: In Blender or Maya, basic scenes are processed slowly due to the small number of CUDA cores (1024 vs. 2560 in RTX 3050).

- Scientific Calculations: CUDA and OpenCL support exists, but for serious tasks, it’s better to choose a card with more memory (8GB or more).

Conclusion: The GTX 960 is only suitable for basic tasks. For professionals, it's an outdated option.


Power Consumption and Cooling

- TDP: 120W—modest even for 2025.

- Power Supply Recommendations: A 400W power supply (e.g., EVGA 400 N1).

- Cooling:

- The reference cooler can handle the load, but it becomes noisy at 75–80°C.

- In compact cases (Mini-ITX), overheating may occur—better to use models with 2-3 fans.

Advice: Regularly clean the card of dust—the thermal paste may have dried up over 10 years.


Comparison with Competitors

In 2025, the GTX 960 competes not with the AMD R9 380 (its contemporary), but with modern budget models:

- NVIDIA RTX 3050 (6GB, $199): +200% performance, supports DLSS 3.5.

- AMD RX 6500 XT (4GB, $159): +120% FPS in DX12, but weak driver support.

- Intel Arc A380 (6GB, $129): Better in Vulkan games but issues with optimization.

Conclusion: Even new budget cards outperform the GTX 960. Its only advantage is price (if you find a new one for $80–100).


Practical Advice

1. Power Supply: 400–450W with 80+ Bronze certification.

2. Compatibility:

- Works with PCIe 3.0/4.0;

- Requires a UEFI motherboard for Windows 11 support.

3. Drivers: Official NVIDIA support ended in 2023. The community releases patches (e.g., via NVCleanstall), but stability is not guaranteed.

Important: The card does not support HDMI 2.1 and DisplayPort 2.0—the maximum resolution is 4K@60Hz via HDMI 2.0.


Pros and Cons

Pros:

- Low power consumption;

- Quiet operation (in models with improved cooling);

- Affordable price on the second-hand market ($50–70).

Cons:

- Limited video memory;

- No RTX/DLSS support;

- Outdated drivers.


Final Verdict: Who Is the GTX 960 Suitable For?

This card is suitable for:

1. Budget gamers playing older titles (pre-2018).

2. Owners of PCs with outdated power supplies where a powerful GPU cannot be installed.

3. Enthusiasts building retro systems.

Alternative: If the budget allows $150–200, it’s better to purchase a new RTX 3050 or RX 6600—they will last longer and provide comfort in modern games.

The GTX 960 in 2025 is an example of a "workhorse" of the past that can still be useful, but it no longer meets the demands of the ray tracing and AI rendering era.

Basic

Label Name
NVIDIA
Platform
Desktop
Launch Date
January 2015
Model Name
GeForce GTX 960
Generation
GeForce 900
Base Clock
1127MHz
Boost Clock
1178MHz
Bus Interface
PCIe 3.0 x16
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
1753MHz
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.
112.2 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.
37.70 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.
75.39 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.
75.39 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.
2.365 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
120W
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
1x 6-pin
Shader Model
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.
32
Suggested PSU
300W

Benchmarks

Shadow of the Tomb Raider 2160p
Score
10 fps
Shadow of the Tomb Raider 1440p
Score
24 fps
Shadow of the Tomb Raider 1080p
Score
34 fps
GTA 5 1080p
Score
69 fps
FP32 (float)
Score
2.365 TFLOPS
3DMark Time Spy
Score
2236
Blender
Score
203
OctaneBench
Score
47
Vulkan
Score
20775
OpenCL
Score
18448
Hashcat
Score
112347 H/s

Compared to Other GPU

Shadow of the Tomb Raider 2160p / fps
26 +160%
15 +50%
Shadow of the Tomb Raider 1440p / fps
95 +295.8%
75 +212.5%
54 +125%
Shadow of the Tomb Raider 1080p / fps
141 +314.7%
107 +214.7%
79 +132.4%
46 +35.3%
GTA 5 1080p / fps
213 +208.7%
136 +97.1%
FP32 (float) / TFLOPS
2.497 +5.6%
2.415 +2.1%
2.335 -1.3%
2.243 -5.2%
3DMark Time Spy
5182 +131.8%
3906 +74.7%
2755 +23.2%
Blender
1506.77 +642.3%
848 +317.7%
45.58 -77.5%
OctaneBench
123 +161.7%
69 +46.8%
Vulkan
98446 +373.9%
69708 +235.5%
5522 -73.4%
OpenCL
62821 +240.5%
38843 +110.6%
21442 +16.2%
884 -95.2%
Hashcat / H/s
113870 +1.4%
113137 +0.7%
105378 -6.2%
102283 -9%