Review and Testing of the Asus TUF GeForce RTX 4070 SUPER Gaming OC Graphics Card

Written by mxshow2000 on 17 January 2024. Posted in Test Video Cards

We were given the opportunity to test the Asus TUF GeForce RTX 4070 SUPER Gaming OC graphics card, which uses the advanced Ada Lovelace architecture, which ensures high quality graphics and performance.

APPEARANCE AND CHARACTERISTICS

This subsection of our review highlights the main aspects of the Asus TUF GeForce RTX 4070 SUPER Gaming OC, such as appearance and hardware configuration.

APPEARANCE AND CHARACTERISTICS

The Asus TUF GeForce RTX 4070 SUPER Gaming OC is offered in a compact package that not only protects the video card well during transportation, but also allows it to be conveniently stored. Inside the package there is a video card, installation instructions, and additional components.

The Asus TUF GeForce RTX 4070 SUPER Gaming OC package includes only the video card itself and a screwdriver stand with a 16-pin power adapter.

The backplate of the Asus TUF GeForce RTX 4070 SUPER Gaming OC has a special cutout that helps allow airflow through the heatsink array more freely. This allows for better cooling of the video card and lowers the temperature during use, which in turn can improve its performance and increase stability.

In addition, there is also a BIOS switch on the back of the graphics card, which allows you to quickly and easily switch between quiet mode and performance mode.

The Asus TUF GeForce RTX 4070 SUPER Gaming OC video card occupies 3 PCI-E slots in the system unit, which is currently the standard size for video cards. Before purchasing, make sure that you have enough space inside the case to install the video card...

On the side of the video card from Asus TUF GeForce RTX 4070 SUPER Gaming OC there is a standard 16-pin power connector.

Asus TUF GeForce RTX 4070 SUPER Gaming OC is equipped with 4 video outputs: 1 HDMI and 3 DisplayPort.

ARCHITECTURE

4070 4

The GeForce RTX 4070 SUPER is based on the AD104 GPU and features 7,168 CUDA Cores, delivering 36 FP32 Shader-TFLOPS of rendering power for traditional rasterized graphics, 224 4th Gen Tensor Cores offering 568 sparsity-aware AI TOPS for AI processing and DLSS technology, 56 third-generation Ada RT cores capable of 82 RT-TFLOPS to support next-generation ray-traced graphics, 12 GB GDDR6X memory and 36 MB L2 cache. Like all GeForce RTX 40 series GPUs, the RTX 4070 SUPER is full of Ada innovations including Shader Execution Reordering (SER), new RT cores, and DLSS 3.5, while offering the best combination of performance and power efficiency.

The GeForce RTX 4070 SUPER powers the latest games and is at the center of modern AI-enabled PCs. PC gamers demand the best visual quality, and Deep Learning Super Sampling (DLSS) Super Resolution, Frame Generation and Ray Reconstruction technologies, powered by NVIDIA AI, combine with ray tracing to deliver stunning worlds in games like Alan Wake 2, Diablo IV , Pax Dei and Horizon Forbidden West. With DLSS, seven out of eight pixels can be generated by hardware-accelerated AI, increasing full ray tracing performance by up to 4 times with better image quality. The GeForce RTX 4070 SUPER is an amazing upgrade. The RTX 4070 SUPER supports over 500 RTX-enabled games and apps already available and will prepare users for the wave of generative AI applications coming to PC. The GeForce RTX 4070 SUPER is designed to meet these needs. Technological innovations of the NVIDIA Ada Lovelace architecture include:

New Streaming Multiprocessors (SM) - New SMs provide up to 2x performance and power efficiency.
4th Generation Tensor Cores and Optical Flow - Enable and accelerate transformational AI technologies, including new NVIDIA DLSS 3 frame augmentation technology.
3rd Generation RT Cores - Up to 2x ray tracing performance, creating incredibly detailed virtual worlds like never before.
Innovative Ada Memory Subsystem - Ada provides a significant increase in L2 cache, delivering better performance, reduced memory traffic, and improved power efficiency.
Shader Execution Ordering (SER) - SER improves ray tracing operations by up to 2x.

Features of the NVIDIA Ada architecture include a new shader execution engine (Shader Execution Reordering (SER), new RT cores and an improved version of DLSS, making the RTX 4070 an excellent solution for games at any graphics level.

The GeForce RTX 4070 GPU is ideal for content creation and delivers extreme gaming performance. The RTX experience is accelerated in top content creation applications, along with NVIDIA Studio drivers designed for maximum stability. In addition, exclusive tools are available that leverage the power of RTX to accelerate creative processes using artificial intelligence.

NVIDIA Ada Streaming Multiprocessor

The RTX GPU has three main processors: a programmable shader that was introduced almost 20 years ago, RT cores for accelerating ray and triangle intersections, and an artificial intelligence processing pipeline called Tensor Cores.

The Ada architecture improves all three RTX processors:

Programmable Shader: 29 Shader-TFLOPS, compared to 22 TFLOPS on the RTX 3070 Ti. SM Ada includes new Shader Execution Reordering (SER) technology, which redistributes tasks on the fly, doubling the speed of ray tracing shaders. SER is as significant an innovation for GPUs as out-of-order execution has become for CPUs.
4th Generation Tensor Cores: New Tensor Cores in Ada include the NVIDIA Hopper FP8 Transformer engine, which delivers over 466 tensor terraflops with sparsity in the RTX 4070 for AI workloads. This is double the 174 tensor terraflops of the RTX 3070 Ti. Compared to FP16, FP8 reduces storage requirements and doubles AI performance.
3rd Generation RT Cores: New Opacity Micromap Engine speeds up alpha-tested geometry ray tracing by 2x, and new Micro-Mesh Engine allows additional geometry to be created on the fly, allowing for increased geometric complexity without the expense of building and storing BVH . Ray-triangle intersection performance on Ada reaches 67 RT-TFLOPS, compared to 42 RT-TFLOPS on the RTX 3070 Ti.

4th Generation Tensor Cores

Tensor cores are high-performance computing cores specialized in matrix multiplication and accumulation that are used in artificial intelligence and high-performance computing applications. Tensor cores provide breakthrough performance for matrix computations, which are critical for training and inference of deep learning neural networks. Example output applications include NVIDIA DLSS 2.1 technology for gamers, which introduced a new 9x upscaling mode with 1440p internal rendering and upscaling to 8K resolution. In addition, many creative applications have begun to use artificial intelligence features to speed up the work of artists. NVIDIA Broadcast also offers tools for noise removal, virtual backgrounds, and other capabilities for video broadcasting and video conferencing.

The fourth-generation Tensor Cores of the Ada architecture build on the capabilities of previous Ampere GPUs, which supported many new data types and introduced sparsity that doubles the throughput of Tensor Cores compared to Turing Tensor Cores. The fourth generation Ada Tensor Cores support the new FP8 data format, which was first introduced in the NVIDIA Hopper GPU architecture. Compared with FP16, FP8 data format reduces data storage requirements and doubles AI performance. With the new FP8 format and sparsity feature, the GeForce RTX 4070 delivers 466 terraflops of performance for AI inference workloads.

3rd generation RT cores

The third generation of RT cores in the Ada architecture are specialized hardware units for accelerating BVH traversal and performing ray and triangle intersection calculations, which are critical to accelerating ray tracing.

RT cores free up the resources of programmable SM shader cores, allowing them to perform other pixel, vertex, and computation tasks. Ada's RT cores enable ray and triangle intersection testing acceleration up to 2x faster than Ampere GPUs, allowing developers to add more detail to their virtual worlds. The Ada architecture also features the Opacity Micromap Engine, which speeds up ray tracing of alpha-tested geometries by a factor of 2, and the Micro-Mesh Engine, which generates micro-meshes on the fly to generate additional geometry, allowing for increased geometric complexity without the cost of BVH generation and storage.

AV1 video encoding

The GeForce RTX 4070 offers hardware support for video encoding using the AV1 codec. AV1 provides improved image quality at the same bitrates as H.265/H.264, which is a big advantage for game streaming. When using AV1, users can achieve the same image quality with lower bitrates, resulting in smaller file sizes and faster loading videos. Ada's AV1 encoder is 40% more efficient than Ampere's H.264 encoder.

So, with advanced technologies and Ada architecture, the GeForce RTX 4070 is an excellent choice for content creation professionals and gamers who demand high performance and visual quality. With support for the latest technologies such as ray tracing, DLSS and AV1, the RTX 4070 delivers superior visual quality and performance in today's and tomorrow's games and applications.

GPU Tweak III OPERATING MODES

It is important to note that the Asus TUF GeForce RTX 4070 SUPER Gaming OC video card is fully compatible with the GPU Tweak utility, which is one of the main analogues of MSI Afterburner in the field of adjusting clock frequencies, creating operating profiles and monitoring various parameters of graphics adapters from NVIDIA and AMD.

GPU Tweak offers an updated interface that provides greater accessibility by bringing core features to the forefront, and VF Tuner has been redesigned to make manual overclocking much easier. In addition, automatic profile replacement and monitoring session logging are added to exclusive features such as 0dB fan technology.

Thus, users of the Asus TUF GeForce RTX 4070 SUPER Gaming OC video card can use a powerful tool to configure and monitor the performance of their video card, which will allow them to optimize its performance as much as possible and obtain maximum image quality.

In standard mode, the card frequency is specified at 2532 MHz with a power limit of 100%, while the frequency in games reaches 2890 MHz.

In automatic overclocking mode, the card receives a frequency of 2595 MHz, but the power limit is already at 110%.

In silent mode, the processor frequency is reduced to 2535 MHz, and the power limit is also reduced to 90%.

HEATING AND POWER CONSUMPTION

Energy efficiency testing of the Asus TUF GeForce RTX 4070 SUPER Gaming OC video card was carried out in an assembly that included an Intel Core i9 14900K processor and an Asus Z790Pro WIFI motherboard, mounted in a closed DeepCool CH510 WH case.

FurMark 2024 01 11 17 46 15 587

As a result of tests performed using the FurMark program, the Asus TUF GeForce RTX 4070 SUPER Gaming OC video card showed excellent thermal efficiency and energy efficiency. At full load, the device reached a maximum temperature of 62 degrees Celsius, while its power consumption was 225 watts.

TUF Gaming 1000W Gold

We tested the video card using a TUF Gaming 1000W Gold power supply.

The ASUS TUF Gaming 1000W Gold power supply is an efficient and reliable power source designed for long-term use.

Main characteristics:

Capacitors and chokes in power supplies meet the requirements of military-industrial standards.
Two ball bearings provide twice the service life of traditional sliding bearings.
Special board coating to protect against moisture, dust and extreme temperatures.
80 Plus Gold certified thanks to high quality Japanese capacitors and other premium components.
Axial-tech fans: extended blades and limiter ring for increased air pressure.
Fully modular cables with etched surface.
10 year warranty.

The ASUS TUF Gaming 1000W Gold power supply uses the best components that meet military-industrial reliability standards, including capacitors with a mean time between failures of 170 thousand hours (at 105°C).

The supplied modular cables have an etched surface, so they look as stylish as braided cables, but are more flexible - this will be useful when routing them in a compact housing.

The TUF Gaming 1000W Gold PSU uses high-quality, low-impedance Japanese capacitors to deliver excellent power efficiency that meets the 80 PLUS® Gold standard. It's up to 92% efficient, so it runs cooler, is quieter, and is more reliable than lower-efficiency power supplies.

The new ATX 3.0 standard introduces stricter voltage and current regulation requirements for next-generation hardware, and the TUF Gaming 1000W Gold power supply is fully compliant with this standard. The 16-pin PCIe cable included with the power supply delivers up to 600W of power to the latest PCIe 5.0 graphics cards.

As we can see, the 16-pin PCIe cable fits perfectly into the Asus TUF GeForce RTX 4070 SUPER Gaming OC...

TEST PART

Test configuration
Test benches	Asus Z790Pro WIFI 32 GB DDR5 7200@7800 CL30 KingstonFURY Renegade Kingston FURY Renegade PCIe 4.0 NVMe M.2 SSD Case DeepCool CH510 WH
Multimedia equipment	Monitor Philips 326M6VJRMB/00 Power supply TUF Gaming 1000W Gold
Software configuration
operating system	Windows 10 21H2/ Windows 11(Alder Lake)
Graphics driver	Nvidia GeForce/ION Driver Release 546.52 WHQL
Monitoring programs	MSI Afterburner

GAMES

All games were taken at the discretion of the editors, from the point of view of why this new product could be purchased now. We haven’t taken synthetics seriously for a long time; we’re only interested in games:

Hogwarts Legacy