NVIDIA GeForce RTX 2070

NVIDIA GeForce RTX 2070

NVIDIA GeForce RTX 2070 in 2025: Is it worth getting a legend of the past?

Review of capabilities, performance, and relevance in the era of new technologies.


1. Architecture and Key Features

Turing Architecture: The Foundation of a Revolution

The RTX 2070 is built on Turing architecture, introduced by NVIDIA in 2018. Despite its age, this technology remains relevant thanks to innovations like RT Core for ray tracing and Tensor Core for AI algorithms. The card was released using a 12nm process technology (TSMC), which seems modest in 2025 compared to 5nm and 4nm chips, but driver optimization compensates for this.

Unique Features

- RTX (Real-Time Ray Tracing): Even in 2025, the RTX 2070 supports basic ray tracing in games like Cyberpunk 2077 or Alan Wake 2, although the activation of DLSS is often required for comfortable performance.

- DLSS 2.0: AI-driven upscaling technology increases FPS by 30-50%, retaining detail. For instance, in Cyberpunk 2077 at 1440p with DLSS, the card delivers around 45-50 FPS with ray tracing on medium settings.

- FidelityFX Super Resolution (FSR): Support for AMD's open technology is available, but its effectiveness is lower than that of DLSS.


2. Memory: Balancing Speed and Capacity

GDDR6: A Proven Standard

The card features 8GB of GDDR6 memory with a bandwidth of 448 GB/s (256-bit bus). This is sufficient for gaming at 1440p, but in 4K or when handling heavy textures (for example, in Microsoft Flight Simulator 2024), there may be stuttering due to insufficient VRAM.

Impact on Performance

- In Hogwarts Legacy (1440p, Ultra), the RTX 2070 achieves a steady 60 FPS without ray tracing, but when RT is enabled, the drop to 35-40 FPS is mitigated by DLSS.

- For editing 4K video in DaVinci Resolve, 8GB is adequate, but rendering complex 3D scenes in Blender may require more substantial solutions.


3. Gaming Performance: Numbers and Realities

1080p: Comfortable Gaming

In Full HD, the card shows excellent results:

- Apex Legends (Ultra): 120-140 FPS.

- Elden Ring (Max, without RT): 55-60 FPS.

1440p: The Optimal Choice

- Call of Duty: Modern Warfare IV (High): 75-90 FPS.

- Starfield (Medium, with DLSS): 60 FPS.

4K: Only with Compromises

- Forza Horizon 5 (High, DLSS Performance): 50-55 FPS.

- Without DLSS, 4K gaming is practically unattainable.

Ray Tracing: Beauty Comes at a Cost

Enabling RT reduces FPS by 30-40%, but DLSS restores playability. In Cyberpunk 2077 (1440p, RT Medium, DLSS Balanced) — 45-50 FPS.


4. Professional Tasks: Not Just Gaming

Video Editing and Rendering

Thanks to 2304 CUDA cores, the RTX 2070 handles rendering in Blender and Premiere Pro. For example, rendering a 5-minute 4K video in Adobe Premiere takes about 12-15 minutes.

Scientific Calculations

With CUDA and OpenCL support, the card is useful for entry-level machine learning or simulations in MATLAB, but for serious tasks, it’s better to choose RTX 3000/4000 series cards with larger memory capacities.


5. Power Consumption and Heat Dissipation

TDP of 175W: Modest but Needs Attention

- Power Supply: Minimum of 550W (recommended 650W for overhead).

- Cooling: Reference models heat up to 75-80°C under load. The optimal choice is versions with 2-3 fans (for example, from ASUS Dual or MSI Gaming Z).

- Case: Good ventilation is essential — at least 2 intake fans and 1 exhaust fan.


6. Comparison with Competitors

AMD Radeon RX 5700 XT

- Pros: Cheaper (~$220), comparable performance in games without RT.

- Cons: No hardware ray tracing, worse compatibility with professional software.

NVIDIA RTX 3060

- Pros: 12 GB of memory, support for DLSS 3.0.

- Cons: Higher price (~$300), in games without DLSS 3.0 RTX 2070 may sometimes be faster.

Conclusion: RTX 2070 surpasses competitors with RTX and DLSS usage but falls short in tasks requiring large amounts of VRAM.


7. Practical Tips

- Power Supply: Don’t skimp — Corsair CX650M or Be Quiet! System Power 10.

- Compatibility: Works with PCIe 3.0 and 4.0, Intel 8th generation and newer / AMD Ryzen 2000+ processors.

- Drivers: Regularly update via GeForce Experience. Avoid “beta versions” for stability.


8. Pros and Cons

Pros:

- Support for DLSS and RTX.

- Optimal for 1440p.

- Good performance in professional tasks.

Cons:

- 8GB of memory is a weak point in 2025.

- High temperatures in reference models.

- Prices of new models (~$250) are close to that of the RTX 3060.


9. Final Conclusion: Who Should Consider RTX 2070?

This graphics card is an excellent choice for:

- Gamers who want to play at 1440p with high settings and moderate use of RTX.

- Budget streaming enthusiasts — NVENC provides smooth transmission.

- Beginner editors and designers who do not require an ultra-powerful GPU.

Why in 2025? Despite its age, the RTX 2070 remains a "workhorse" due to DLSS and NVIDIA support. If you find it for under $250, it's a worthwhile purchase. However, for 4K or complex rendering, it's better to look at newer models.


Prices are valid as of April 2025. Please check availability at local stores!

Basic

Label Name
NVIDIA
Platform
Desktop
Launch Date
October 2018
Model Name
GeForce RTX 2070
Generation
GeForce 20
Base Clock
1410MHz
Boost Clock
1620MHz
Bus Interface
PCIe 3.0 x16
Transistors
10,800 million
RT Cores
36
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.
288
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.
144
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.
103.7 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.
233.3 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.
14.93 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.
233.3 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.
7.316 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.
36
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.
2304
L1 Cache
64 KB (per SM)
L2 Cache
4MB
TDP
175W
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 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
450W

Benchmarks

Shadow of the Tomb Raider 2160p
Score
38 fps
Shadow of the Tomb Raider 1440p
Score
69 fps
Shadow of the Tomb Raider 1080p
Score
96 fps
Cyberpunk 2077 2160p
Score
31 fps
Cyberpunk 2077 1440p
Score
38 fps
Cyberpunk 2077 1080p
Score
56 fps
Battlefield 5 2160p
Score
55 fps
Battlefield 5 1440p
Score
98 fps
Battlefield 5 1080p
Score
125 fps
GTA 5 2160p
Score
88 fps
GTA 5 1440p
Score
92 fps
GTA 5 1080p
Score
174 fps
FP32 (float)
Score
7.316 TFLOPS
3DMark Time Spy
Score
9097
Blender
Score
2020.49
Vulkan
Score
82376
OpenCL
Score
91174
Hashcat
Score
442022 H/s

Compared to Other GPU

Shadow of the Tomb Raider 2160p / fps
77 +102.6%
25 -34.2%
12 -68.4%
Shadow of the Tomb Raider 1440p / fps
131 +89.9%
49 -29%
27 -60.9%
Shadow of the Tomb Raider 1080p / fps
71 -26%
Cyberpunk 2077 2160p / fps
67 +116.1%
37 +19.4%
8 -74.2%
Cyberpunk 2077 1440p / fps
79 +107.9%
11 -71.1%
Cyberpunk 2077 1080p / fps
127 +126.8%
21 -62.5%
Battlefield 5 2160p / fps
Battlefield 5 1440p / fps
150 +53.1%
116 +18.4%
81 -17.3%
Battlefield 5 1080p / fps
186 +48.8%
144 +15.2%
GTA 5 2160p / fps
174 +97.7%
100 +13.6%
GTA 5 1440p / fps
191 +107.6%
116 +26.1%
73 -20.7%
GTA 5 1080p / fps
231 +32.8%
176 +1.1%
141 -19%
86 -50.6%
FP32 (float) / TFLOPS
8.085 +10.5%
7.521 +2.8%
6.592 -9.9%
3DMark Time Spy
16792 +84.6%
4952 -45.6%
Blender
15026.3 +643.7%
3514.46 +73.9%
1064 -47.3%
552 -72.7%
Vulkan
207930 +152.4%
119880 +45.5%
54373 -34%
30994 -62.4%
OpenCL
245925 +169.7%
144289 +58.3%
66428 -27.1%
46137 -49.4%
Hashcat / H/s
505860 +14.4%
452205 +2.3%
406176 -8.1%
403046 -8.8%