NVIDIA GeForce GTX 590

NVIDIA GeForce GTX 590

NVIDIA GeForce GTX 590: A Legend of the Dual-Chip Era in Retrospective 2025

Introduction

The NVIDIA GeForce GTX 590, released in 2011, became a symbol of an era when manufacturers competed to create dual-chip monsters. Despite its age, this card still piques the interest of enthusiasts. In 2025, its relevance is limited, but it remains an important part of GPU history. Let’s look at its features through the lens of modern requirements.


Architecture and Key Features

Fermi Architecture: The Power Backbone

The GTX 590 is built on the Fermi architecture (GF110), combining two GPUs on a single board. The manufacturing process is 40nm, which was cutting-edge in 2011. Each chip contained 512 CUDA cores, totaling 1024 cores.

Unique Features of Its Time

The card supported NVIDIA SLI (for linking multiple GPUs), PhysX for physics calculations in games, and CUDA for parallel computing. Technologies like RTX or DLSS, which emerged after 2018, are absent here—ray tracing and upscaling became available only with the RTX series.

Design Characteristics

The GTX 590 was one of the first dual-chip cards where NVIDIA's engineers placed two GPUs under a single cooling system. This required careful temperature management, which later resulted in overclocking limitations.


Memory: Size and Bandwidth

GDDR5: The Standard of the Era

Each chip was equipped with 1.5 GB of GDDR5 memory, totaling 3 GB. However, due to the AFR (Alternate Frame Rendering) technology, the effective volume for gaming remained 1.5 GB—sufficient for games from 2011–2013, but today even 1080p in modern projects requires a minimum of 4–6 GB.

Bandwidth

The 384-bit bus for each chip and a frequency of 3414 MHz provided a bandwidth of 164 GB/s per GPU. In comparison, modern cards like the RTX 4060 Ti (288 GB/s on GDDR6X) exhibit a speed increase of 1.7 times.


Gaming Performance: Nostalgia in the 4K Era

Retro Games and Old Projects

In games from the 2010s, the GTX 590 delivered impressive results:

- The Witcher 2: 45–55 FPS on Ultra (1080p);

- Battlefield 3: 60–70 FPS (1080p);

- Crysis 2: 50–60 FPS (1080p).

Modern Projects: The Realities of 2025

In 2025, even at 1080p, the card struggles with titles:

- Cyberpunk 2077: 10–15 FPS on Low (without ray tracing);

- Hogwarts Legacy: 8–12 FPS (720p, minimum settings).

Support for resolutions above 1080p (1440p, 4K) is virtually impossible due to lack of memory and computing power.


Professional Tasks: CUDA in the Era of Limitations

Video Editing and Rendering

Thanks to CUDA, the GTX 590 could accelerate rendering in Adobe Premiere Pro or Blender (Cycles) during the 2010s. However, today its 1024 Fermi cores greatly lag behind even budget RTX 3050 cards (2560 Ampere cores). For example, rendering a scene in Blender takes 4–5 times longer.

Scientific Calculations

For tasks like MATLAB or Folding@Home, the card is only suitable for theoretical experiments. Modern CUDA and OpenCL libraries often require support for architectures no lower than Kepler (2012), limiting compatibility.


Power Consumption and Heat Dissipation

TDP and Power Supply Requirements

The GTX 590 has a TDP of 365 Watts. A power supply rated at a minimum of 700 Watts with quality +12V lines was required for stable operation. In 2025, such specifications are considered archaic: even the RTX 4090 (450 Watts) is more efficient in terms of performance.

Overheating Issues

The card is known for its "hot temper": under load, temperatures could reach up to 90°C. Cooling recommendations include:

- A case with good ventilation (at least 3 fans);

- Regular thermal paste replacement;

- Avoid overclocking—the risk of chip overheating is too high.


Comparison with Competitors: The Titan Battle of 2011

AMD Radeon HD 6990: The Main Rival

The HD 6990, like the GTX 590, combined two chips (Cayman XT) with 3072 Stream cores. Its advantages include:

- Larger memory size—4 GB GDDR5;

- Support for Eyefinity to connect 6 monitors.

Disadvantages: higher TDP (375 Watts) and a noisy cooling system.

Summary of the Showdown

Both cards had similar performance, but the GTX 590 excelled in relative energy efficiency and PhysX support. However, by 2025, both solutions are considered outdated both morally and technically.


Practical Tips for Enthusiasts

Power Supply and Compatibility

- Minimum of 700 Watts with 80+ Bronze certification;

- Check the connectors: the GTX 590 requires 2×8-pin PCIe.

Platforms and Drivers

- Compatible with PCIe 2.0, but works in 3.0/4.0 slots;

- Official NVIDIA drivers ceased support in 2020. For Windows 10/11, use modified community versions.

Usage Nuances

- Avoid modern games with Vulkan/DX12—the card is optimized for DX11;

- To reduce heat, limit FPS via NVIDIA Inspector.


Pros and Cons of the GTX 590 in 2025

Pros:

- Legendary status and unique design;

- SLI support for creating multi-GPU systems;

- Sufficient for retro games and older projects.

Cons:

- Limited memory (1.5 GB per chip);

- High power consumption;

- Lack of support for modern APIs and technologies.


Final Conclusion: Who Should Consider the GTX 590?

This graphics card is an artifact of an era that should only be considered for:

1. Collectors—as part of gaming hardware history;

2. Retro PC Enthusiasts—for building systems from the 2010s;

3. Experienced Users—as an experimental project.

For modern gaming, video editing, or calculations, the GTX 590 is unfit. Its value today lies in nostalgia and engineering heritage, rather than practical performance. If you are looking for a workhorse, consider budget models like the RTX 3050 or AMD RX 6600—they are 5–7 times faster while consuming half the power.


P.S. New GTX 590s have not been sold since 2013. On the second-hand market, prices range from $50–100 depending on condition. Remember: buying such a card in 2025 is an investment in history, not in performance.

Basic

Label Name
NVIDIA
Platform
Desktop
Launch Date
March 2011
Model Name
GeForce GTX 590
Generation
GeForce 500
Bus Interface
PCIe 2.0 x16
Transistors
3,000 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.
64
Foundry
TSMC
Process Size
40 nm
Architecture
Fermi 2.0

Memory Specifications

Memory Size
1536MB
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
854MHz
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.
164.0 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.
19.46 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.
38.91 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.
155.5 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.
1.219 TFLOPS

Miscellaneous

SM Count
?
Multiple Streaming Processors (SPs), along with other resources, form a Streaming Multiprocessor (SM), which is also referred to as a GPU's major core. These additional resources include components such as warp schedulers, registers, and shared memory. The SM can be considered the heart of the GPU, similar to a CPU core, with registers and shared memory being scarce resources within the SM.
16
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.
512
L1 Cache
64 KB (per SM)
L2 Cache
768KB
TDP
365W
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.
N/A
OpenCL Version
1.1
OpenGL
4.6
DirectX
12 (11_0)
CUDA
2.0
Power Connectors
2x 8-pin
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.
48
Suggested PSU
750W

Benchmarks

FP32 (float)
Score
1.219 TFLOPS

Compared to Other GPU

FP32 (float) / TFLOPS
1.242 +1.9%
1.176 -3.5%
1.16 -4.8%