NVIDIA GeForce RTX 2070 SUPER

NVIDIA GeForce RTX 2070 SUPER

NVIDIA GeForce RTX 2070 SUPER in 2025: Is It Worth Buying?

Review of an outdated but still relevant GPU for gamers and professionals


Architecture and Key Features: Turing - A Step into the Future

Released in 2019, the RTX 2070 SUPER is based on the Turing architecture, which became revolutionary thanks to the introduction of RT cores for ray tracing and tensor cores for AI algorithms. The card was manufactured using 12nm TSMC technology, which seems archaic in 2025 compared to 5nm and 4nm chips. However, its "features" are still in demand:

- RTX (Real-Time Ray Tracing): Allows for realistic lighting and shadows in games.

- DLSS 2.0: AI enhances image resolution with minimal loss of quality, which is critical for 1440p and 4K.

- FidelityFX Super Resolution (FSR): An AMD technology, but the RTX 2070 SUPER supports it through drivers, expanding the list of optimized games.

Despite its age, the card handles basic ray tracing tasks, but it falls behind the new RTX 40 series GPUs in rendering speed.


Memory: GDDR6 and the Balance Between Speed and Volume

The RTX 2070 SUPER is equipped with 8 GB of GDDR6 memory with a 256-bit bus. The bandwidth is 448 GB/s, which, in 2025, is sufficient for most games at high settings in 1440p, but becomes a bottleneck at 4K or when working with ultra textures in projects like Cyberpunk 2077: Phantom Liberty.

The main downside is the limited VRAM. Modern AAA games (for example, Starfield or GTA VI) at maximum settings can consume up to 10-12 GB, leading to FPS drops. A solution is to reduce texture quality or use DLSS/FSR.


Gaming Performance: 1440p as the Sweet Spot

In 2025, the RTX 2070 SUPER remains relevant for 2560x1440 pixel monitors. Examples of FPS (high settings, without ray tracing):

- Fortnite: 110-130 FPS (1440p, DLSS Quality).

- Hogwarts Legacy: 45-55 FPS (1440p, RTX Medium + DLSS Balanced).

- Call of Duty: Modern Warfare V: 75-90 FPS (1440p, FSR 2.1).

For 4K, the card is only suitable for less demanding projects (CS2, Valorant) or with significant compromises in settings. Ray tracing reduces FPS by 30-40%, so its use without DLSS is not advisable.


Professional Tasks: CUDA Still Holds Up

With 2304 CUDA cores, the RTX 2070 SUPER handles:

- Video Editing: Rendering in DaVinci Resolve or Premiere Pro is accelerated 2-3 times compared to integrated graphics.

- 3D Modeling: In Blender, rendering a mid-level scene takes about 15-20 minutes.

- Machine Learning: Suitable for basic experiments with TensorFlow, but models like Stable Diffusion run slower than on cards with 12+ GB of memory.

For serious professional tasks in 2025, it's better to choose the RTX 4060 Ti (16 GB) or Radeon Pro W7800, but as a budget option, the RTX 2070 SUPER is still viable.


Power Consumption and Heat Dissipation: Modest but Demanding

The card’s TDP is 215W. For a build with this card, you need:

- Power Supply: At least 550W (recommended 650W for headroom).

- Cooling: Reference models heat up to 75-80°C under load. Look for models with 2-3 fans (e.g., ASUS Dual EVO or MSI Gaming X).

- Case: Good airflow (at least 2 intake and 1 exhaust fan).


Comparison with Competitors: Who Does It Compete Against in 2025?

- AMD Radeon RX 6600 XT (8 GB): Cheaper (~$250), but weaker at 1440p and lacks hardware ray tracing.

- NVIDIA RTX 3060 (12 GB): Priced at $300-350, offers comparable performance but more memory and DLSS 3.0 (not available on the RTX 20 series).

- Intel Arc A770 (16 GB): At $300, delivers comparable FPS in DX12 games, but drivers are still immature.

Today, the RTX 2070 SUPER is an option for those looking for a used card priced at $150-200. New models (if still available) are priced between $250-300, which is debatable compared to the RTX 4060.


Practical Tips: How to Avoid Issues

- Power Supply: Don't skimp! It's better to get a Corsair CX650M or Be Quiet! Pure Power 11.

- Compatibility: The card requires PCIe 3.0 x16, but it also works on PCIe 4.0/5.0 motherboards.

- Drivers: NVIDIA supports older GPUs for 5-7 years. In 2025, updates will be less frequent, but critical patches for new games continue to be released.


Pros and Cons

Pros:

- Affordable price on the secondary market.

- Support for DLSS and RTX.

- Better energy efficiency than the RX 5700 XT.

Cons:

- Only 8 GB of memory.

- No support for DLSS 3.0 and Frame Generation.

- Outdated RT cores.


Final Conclusion: Who Is the RTX 2070 SUPER For?

This graphics card is a choice for:

1. Gamers with a 1440p monitor who are willing to make compromises in ultra settings.

2. Streamers who need enough power for gaming and video encoding simultaneously.

3. Budget professionals working in Adobe Suite or Blender.

If you find a new RTX 2070 SUPER for $250 — that’s not a bad deal. But for $300-400, it's better to consider modern AMD RX 7600 XT or NVIDIA RTX 4060. In 2025, the “super” 2070 is still breathing, but its time is running out.

Basic

Label Name
NVIDIA
Platform
Desktop
Launch Date
July 2019
Model Name
GeForce RTX 2070 SUPER
Generation
GeForce 20
Base Clock
1605MHz
Boost Clock
1770MHz
Bus Interface
PCIe 3.0 x16
Transistors
13,600 million
RT Cores
40
Tensor Cores
?
Tensor Cores are specialized processing units designed specifically for deep learning, providing higher training and inference performance compared to FP32 training. They enable rapid computations in areas such as computer vision, natural language processing, speech recognition, text-to-speech conversion, and personalized recommendations. The two most notable applications of Tensor Cores are DLSS (Deep Learning Super Sampling) and AI Denoiser for noise reduction.
320
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.
160
Foundry
TSMC
Process Size
12 nm
Architecture
Turing

Memory Specifications

Memory Size
8GB
Memory Type
GDDR6
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.
256bit
Memory Clock
1750MHz
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.
448.0 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.
113.3 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.
283.2 GTexel/s
FP16 (half)
?
An important metric for measuring GPU performance is floating-point computing capability. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable. 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.
18.12 TFLOPS
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.
283.2 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.
9.243 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.
40
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.
2560
L1 Cache
64 KB (per SM)
L2 Cache
4MB
TDP
215W
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 Ultimate (12_2)
CUDA
7.5
Power Connectors
1x 6-pin + 1x 8-pin
Shader Model
6.6
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.
64
Suggested PSU
550W

Benchmarks

Shadow of the Tomb Raider 2160p
Score
41 fps
Shadow of the Tomb Raider 1440p
Score
78 fps
Shadow of the Tomb Raider 1080p
Score
116 fps
Cyberpunk 2077 2160p
Score
37 fps
Cyberpunk 2077 1440p
Score
44 fps
Cyberpunk 2077 1080p
Score
62 fps
Battlefield 5 2160p
Score
57 fps
Battlefield 5 1440p
Score
99 fps
Battlefield 5 1080p
Score
136 fps
GTA 5 2160p
Score
69 fps
GTA 5 1440p
Score
94 fps
GTA 5 1080p
Score
184 fps
FP32 (float)
Score
9.243 TFLOPS
3DMark Time Spy
Score
10331
Blender
Score
2220.56
Vulkan
Score
94845
OpenCL
Score
103572
Hashcat
Score
528693 H/s

Compared to Other GPU

Shadow of the Tomb Raider 2160p / fps
88 +114.6%
51 +24.4%
30 -26.8%
17 -58.5%
Shadow of the Tomb Raider 1440p / fps
157 +101.3%
59 -24.4%
36 -53.8%
Shadow of the Tomb Raider 1080p / fps
230 +98.3%
165 +42.2%
Cyberpunk 2077 2160p / fps
8 -78.4%
Cyberpunk 2077 1440p / fps
35 -20.5%
11 -75%
Cyberpunk 2077 1080p / fps
127 +104.8%
21 -66.1%
Battlefield 5 2160p / fps
38 -33.3%
Battlefield 5 1440p / fps
165 +66.7%
63 -36.4%
Battlefield 5 1080p / fps
192 +41.2%
GTA 5 2160p / fps
96 +39.1%
55 -20.3%
GTA 5 1440p / fps
191 +103.2%
116 +23.4%
73 -22.3%
GTA 5 1080p / fps
231 +25.5%
156 -15.2%
141 -23.4%
86 -53.3%
FP32 (float) / TFLOPS
10.535 +14%
10.043 +8.7%
8.781 -5%
8.49 -8.1%
3DMark Time Spy
20998 +103.3%
13231 +28.1%
8014 -22.4%
Blender
15026.3 +576.7%
3514.46 +58.3%
1064 -52.1%
552 -75.1%
Vulkan
382809 +303.6%
140875 +48.5%
61331 -35.3%
34688 -63.4%
OpenCL
321810 +210.7%
152485 +47.2%
72786 -29.7%
53439 -48.4%
Hashcat / H/s
530553 +0.4%
529739 +0.2%
521915 -1.3%
521597 -1.3%