NVIDIA GeForce GTX 980 Ti

NVIDIA GeForce GTX 980 Ti

NVIDIA GeForce GTX 980 Ti in 2025: Is the Legend Worth Considering?

Introduction

The NVIDIA GeForce GTX 980 Ti is a legendary graphics card that was released in 2015. Despite its age, it still piques the interest of enthusiasts. But how relevant is this model in 2025? Let’s delve into the details, evaluating its strengths and weaknesses in the context of modern tasks.


1. Architecture and Key Features

Maxwell Architecture: Simplicity and Efficiency

The GTX 980 Ti is built on the Maxwell architecture (GM200), manufactured using a 28nm process. This was a revolutionary solution for its time: 2816 CUDA cores, 176 texture units, and 96 raster units. However, Maxwell does not support modern technologies like Ray Tracing (RTX) or DLSS—these were introduced in later Turing and Ampere architectures.

Unique Features from 2015

The card offered proprietary NVIDIA technologies:

- Dynamic Super Resolution (DSR)—rendering games at a high resolution followed by downscaling to the monitor.

- MFAA—anti-aliasing to improve image quality without significant performance loss.

In 2025, these features seem outdated compared to DLSS 3.5 or AMD FSR 3, which use artificial intelligence to boost FPS.


2. Memory: Volume and Bandwidth

GDDR5: Legacy of the Past

The GTX 980 Ti is equipped with 6GB of GDDR5 memory on a 384-bit bus. The bandwidth is 336 GB/s (effective memory clock of 7GHz). In comparison, modern cards with GDDR6X (such as the RTX 4070) reach 504 GB/s.

Impact on Performance

6GB of VRAM in 2025 is critically low for gaming at 4K or working with heavy textures. For instance, in Cyberpunk 2077: Phantom Liberty at ultra settings in 1440p, the graphics card may face memory shortages, leading to drops in FPS.


3. Gaming Performance: What Can You Run?

1080p: Basic Comfort

In less demanding titles (e.g., CS2, Valorant), the GTX 980 Ti delivers 100–150 FPS on high settings. In AAA games from 2023–2024 (such as Starfield), the average FPS on medium settings is around 40–50.

1440p and 4K: Unfortunately, Not Relevant

In Red Dead Redemption 2 at 1440p (high settings), it achieves 30–35 FPS. For 4K, the card is weak even in older games: The Witcher 3 barely reaches 25–30 FPS.

Ray Tracing: Absent

The GTX 980 Ti lacks hardware support for RT cores. Software emulation (e.g., through Proton on Linux) reduces FPS by 2–3 times, making the technology useless.


4. Professional Tasks: Limited Potential

CUDA Cores: Assistance in Rendering

With 2816 CUDA cores, it can accelerate rendering in Blender or Adobe Premiere Pro, but modern applications (Unreal Engine 5, DaVinci Resolve) require more memory and support for new APIs.

Scientific Calculations: Outdated Drivers

For machine learning or computations based on CUDA/OpenCL, it is better to choose cards that support the latest versions of frameworks (TensorFlow 3.x, PyTorch 3.0). Drivers for the GTX 980 Ti haven’t been updated for a long time, limiting compatibility.


5. Power Consumption and Heat Dissipation

TDP 250W: A Gluttonous "Dinosaur"

The card's power consumption is comparable to modern RTX 4070s (200W), but with lower performance. A power supply of at least 600W is required (recommended: 80+ Gold).

Cooling: Noise vs. Temperature

The reference NVIDIA cooler (turbine) is noisy under load (up to 45 dB). The optimal choice in 2025 would be models with liquid cooling or upgraded coolers (e.g., Arctic Accelero Xtreme IV).

Case Recommendations

- Minimum of 2 expansion slots.

- Good ventilation: 3–4 case fans.

- Avoid compact cases—overheating may occur.


6. Comparison with Competitors

Against Contemporaries (2015):

- AMD Radeon R9 Fury X: 4GB HBM, comparable performance but poorer optimization for DirectX 12.

- NVIDIA GTX 1070: 8GB GDDR5, lower power consumption but weaker in 4K.

Against New Cards (2025):

- NVIDIA RTX 4060: 8GB GDDR6, supports DLSS 3.5, has RT cores, TDP 115W. Price: $299.

- AMD RX 7600 XT: 12GB GDDR6, FSR 3, similar price.

The GTX 980 Ti falls short in energy efficiency and support for new technologies, but it can still be a good value on the second-hand market at a price of $80–$120 (used).


7. Practical Tips

Power Supply

- Minimum 600W (recommended 650–750W for headroom).

- Quality models: Corsair RM650x, EVGA SuperNOVA 750 G6.

Compatibility

- PCIe 3.0 x16—works in PCIe 4.0/5.0 slots, but without speed boost.

- Check the card length (26.7 cm) and case sizes.

Drivers

- The latest version for the GTX 980 Ti is Game Ready Driver 552.44 (2024). NVIDIA support has ended.


8. Pros and Cons

Pros:

- Affordable price on the second-hand market.

- Reliable build (especially in top models from ASUS, MSI).

- Support for DirectX 12 (without features of level 12_2).

Cons:

- No ray tracing or DLSS.

- High power consumption.

- Limited memory for modern games.


9. Final Conclusion: Who is the GTX 980 Ti Suitable For?

This graphics card is a choice for:

1. Budget gamers willing to play on medium settings at 1080p.

2. Retro game enthusiasts (2000s–2010s).

3. Owners of old PCs where upgrading to a modern GPU isn't possible due to lack of PCIe 4.0 or a weak CPU.

However, if you want to comfortably play 2025's new titles or engage in professional tasks, consider the RTX 4060 or RX 7600 XT. The GTX 980 Ti remains a niche solution, relevant only in very limited scenarios.


P.S. New GTX 980 Ti cards are not produced in 2025. If someone offers you a "new" card, it is likely a refurbished or counterfeit unit.

Basic

Label Name
NVIDIA
Platform
Desktop
Launch Date
June 2015
Model Name
GeForce GTX 980 Ti
Generation
GeForce 900
Base Clock
1000MHz
Boost Clock
1076MHz
Bus Interface
PCIe 3.0 x16
Transistors
8,000 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.
176
Foundry
TSMC
Process Size
28 nm
Architecture
Maxwell 2.0

Memory Specifications

Memory Size
6GB
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.
384bit
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.
336.6 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.
103.3 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.
189.4 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.
189.4 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.
6.181 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.
2816
L1 Cache
48 KB (per SMM)
L2 Cache
3MB
TDP
250W
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 + 1x 8-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.
96
Suggested PSU
600W

Benchmarks

Shadow of the Tomb Raider 2160p
Score
24 fps
Shadow of the Tomb Raider 1440p
Score
46 fps
Shadow of the Tomb Raider 1080p
Score
77 fps
GTA 5 2160p
Score
73 fps
GTA 5 1440p
Score
75 fps
GTA 5 1080p
Score
122 fps
FP32 (float)
Score
6.181 TFLOPS
3DMark Time Spy
Score
5663
Blender
Score
552
OctaneBench
Score
130
Vulkan
Score
49482
OpenCL
Score
36927

Compared to Other GPU

Shadow of the Tomb Raider 2160p / fps
39 +62.5%
26 +8.3%
1 -95.8%
Shadow of the Tomb Raider 1440p / fps
95 +106.5%
54 +17.4%
Shadow of the Tomb Raider 1080p / fps
141 +83.1%
107 +39%
79 +2.6%
GTA 5 2160p / fps
167 +128.8%
31 -57.5%
GTA 5 1440p / fps
103 +37.3%
82 +9.3%
29 -61.3%
GTA 5 1080p / fps
213 +74.6%
136 +11.5%
FP32 (float) / TFLOPS
6.61 +6.9%
6.522 +5.5%
5.7 -7.8%
3DMark Time Spy
4243 -25.1%
2958 -47.8%
Blender
2020.49 +266%
1064 +92.8%
292.75 -47%
116 -79%
OctaneBench
627 +382.3%
318 +144.6%
39 -70%
Vulkan
77928 +57.5%
25429 -48.6%
9862 -80.1%
OpenCL
79060 +114.1%
61276 +65.9%
18448 -50%
10722 -71%