Nvidia Ampere - new features, better gaming

The list of new features and enhancements brought by the Nvidia Ampere GPU is truly impressive. Numerous technical explanations are often incomprehensible to gamers and users, so in this text, we will address the specific benefits you can expect by switching to one of the GeForce RTX 30 graphics card models. There are a lot of them, but we have chosen the most important ones that will bring a qualitative shift in everyday work and play...

We have previously explained in detail all the important changes in the field of the improved microarchitecture of the Nvidia Ampera graphics processor. High power allows the introduction of some new and improved proven functionalities, which improve the speed and quality of gameplay in games.

DLSS, 4K & 8K Gaming

One of the most important functions is certainly the mention of DLSS, which is performed on Tensor processors. What exactly is the essence of this technology. Gaming in Ray Tracing (RT) mode is a computationally extremely demanding operation for a graphics processor. That's why we mostly play in 1080p resolution with RT effects included, to feel some minimal fluid movement of the actors in the game. Each increase in resolution further slows down the animation in the game, making it unusually slow and ultimately unplayable. Nvidia Ampere is changing the previously described practice. On the GeForce RTX 30 series, DLSS is especially useful for playing in high resolutions when Ray Tracing effects are included. What does DLSS do? 

This technique renders frames at a lower resolution (1440p) than predicted (4K). Simply put, it tracks changes between the last displayed frame and the one to be displayed, analyzes the entire scene by processing RayTracing effects in parallel with classic raster rendering, merging them at the end, and then rendering the final frame displayed in 4K or 8K ultra-high resolutions.

This is most simply explained because Tensor cores are capable of performing RT calculations and graphical operations simultaneously, combining details from multiple rendered frames into final, high-resolution ones. All this in the end before your eyes, results in a perfectly sharp image, fluid animation, and rich graphic details at high resolutions. Until now, something like this has been a mission impossible for older generations of graphics cards.

This is the famous asynchronous mode of operation of the graphics processor, which allows it to perform several different tasks simultaneously - performing multiple calculations and graphics operations at the same time.

 

What does that mean in a specific example? We will try to explain it to you pictorially. When playing Wolfenstein with demanding light effects and shading, in the case of work and calculations exclusively on the shader, ie CUDA processors, the average value of "frame rate" would be 20. With RT processors included, "frame rate" increases to 50 fps, while in the situation engagement of the Tensor core and active DLSS function, reaches a value of 83 fps. So it was with the introduction of the first generation of GeForce RTX graphics cards.

Nvidia Ampere goes a step further, thanks to better architecture and parallel operation of several operations at the same time, it brings a huge improvement by delivering as much as 150 frames per second in the same situation. Now it is at least a little clearer to you why the GeForce RTX 30 is the first generation of graphics cards with which gaming in ultra high resolution and 60 fps becomes possible!

 

Just remember previous generations were such attempts to scale from lower to higher resolutions usually ended in the effect of blurring and a serious drop in quality, especially in those fine graphic details. With the RTX 30 series, this is no longer the case, because, with DLSS 2.0 and the new generation of AI algorithm, the frame display has become practically sharper than the "native" UHD resolution.

Nvidia RTX I / O

Since it is a huge number of calculations and calculations that the graphics processor has to perform in a situation of playing at high resolutions, with high set graphic details, and because all this needs to be done in real-time, a new problem has opened up regarding the delivery of large amounts of information. which must be processed. The traditional way has become a bottleneck and a limiting factor in the further progress of performance and the improvement of graphics in games. Today's latest gaming titles take up as much as 200 GB of space. Just imagine that flow of information while playing. Practically with classic hard drives, playing with the new generation of games becomes painfully slow.

What has it looked like so far? The game reads the necessary data from the hard disk, using a central processor (CPU) to decompress various images, materials, textures, and everything that makes up an individual frame in the animation. To make the games more beautiful, the developers use huge uncompressed images, and at the same time, you need to provide playful gameplay, which is in direct contrast. As the efforts to make the games more beautiful with richer graphic details, the amount of information increased, which was reflected in a higher disk space usage. Transferring that huge amount of data has become an aggravating circumstance that slows down the operation of the entire PC.

If you use a standard hard drive with a maximum transfer of about 100 MB / s, you will be able to employ only a few cores on your CPU, but by using high-speed M.2 PCIe SSDs with a transfer of over 7 GB / s, you will "charge" data and fully load all 24 cores on the job of decompressing the textures of one heavyweight such as the AMD Threadripper 3960X processor. This increases the occupancy of system resources, which certainly affects the performance of the entire PC. On the other hand, the power of the GeForce RTX 30 Ampere processor is so great that in terms of data decompression, it far exceeds the capabilities of the currently fastest M.2 PCIe x4 Gen4 SSD models.

The Nvidia RTX I / O system is significantly faster and will be built into the Windows Direct Storage API. It takes full advantage of the presence of fast NVMe SSDs in the context of the high demands of modern games. Their request to load data is now processed by a graphics processor that directly pulls data from the SSD into the local video memory while completely freeing up the CPU and RAM. The complete job of decompressing materials, textures, and data transfer are done by the GPU, leaving the CPU completely free to perform system jobs.

A lot of work has been avoided around transferring an enormous amount of data, which solves the problem of traffic congestion on the PCIe bus, improves the performance of the entire computer, which the end-user feels in the almost instantaneous response in games when loading. Specifically, for everything you waited for to load for one minute, RTX I / O will load in 2-5 seconds. Simply, you spend more precious time playing, not waiting for the game and the level to load!

Nvidia Performance Tuning

Another detail is important for your ultimate gaming pleasure - fine-tuning performance. With the release of the GeForce RTX 30 series graphics cards, Nvidia has enabled the GPU Tuning option within the GeForce Experience tool, which allows the user to adjust the GPU voltage and operating frequency, temperature maximum, fan curves and all without the use of special tools. . If you are not experienced enough, you can use automatic settings at first and thus leave overclocking to Nvidia automation.

The first thing to do is to turn on the “Enable experimental features” option and then select “Performance” in the GeForce Experience “overlay” menu. Increase the value of the “Power Limit” slider to the maximum and turn on the “Enable Automatic Tuning” option. Automation will get the most out of your graphics card, and control over what's going on and how much your GPU is straining during gaming, you can track in real-time with "overlay" telemetry, which you call by pressing the "Alt + R" key.

Nvidia Reflex and Broadcast

Competitive players often talk about how it feels in the game when they talk about performance. In duel games and fast-paced ones such as Battle Royal titles or tactical FPS, the situation is a little different because not everything is worth the average "frame rate" which is practically the primary benchmark of performance. However, not everything is in the "raw" transfer, the finesse in the mentioned games is hidden in the speed of response between your action and the reaction of your computer!

Perhaps the best way to express the response speed and response category while playing on your PC is the system delay category, which is the time that elapses between your mouse or keyboard click and the action displayed on the screen. The shorter the time, the better the response, and the more successful you are in a direct duel with your opponents. Nvidia Reflex is a suite of technologies designed to improve system response and reduce its latency in highly competitive gaming titles. It does this by currently increasing the GPU frequency and sending the frame to the display screen a little earlier in situations where the CPU is under heavy load.

The result is that you will spot the target faster, react before the enemy, and at the same time increase the accuracy of aiming. The combination of Nvidia Reflex and GSync technology will give you an advantage and superiority in direct duels with opponents.

If you want to share your gaming experience with others, NVIDIA GeForce RTX 30 is the right tool for gaming "live streaming" and transmitting your comments "live" while you play. This is a very hardware-intensive work scenario. Thanks to the best hardware GPU for video encoding, the new series of GeForce RTX 30 graphics cards enables video transmission at the maximum number of frames and the best image quality.

And that's not all, because to allow you to make the broadcast look "professional", you can now turn your study into a real, virtual studio, raising the quality of your live production to a whole new level of quality, but also a visually better environment. The whole work includes a Tensor core and an AI algorithm that neutralizes noise in the sound, enables work and effects with a virtual studio background, combining a part of the frame with the one delivered by the webcam.

Thanks to the acceleration achieved by the new generation Tensor core, the AI network can create and reproduce high-quality visual effects during live streaming in real-time without penalties in terms of image quality or performance. In a word, GeForce RTX 30 models are your perfect choice for live streaming.

Studio Content Creation

And last but not least, the users who are engaged in creative work and the creation of demanding, professional content in GPU accelerated applications are not forgotten. A large number of applications are supported by the Nvidia Studio platform, which includes: special drivers, then development tools, and applications that know how to recognize and use the computing power of GeForce RTX 30 graphics cards. This type of application will enable faster rendering of static scenes and 3D animations, as well as video rendering, thanks to the fastest hardware processor for encoding video content. Special Nvidia Studio drivers will help you with acceleration that you have never had the opportunity to feel at work before.

Let's be specific:

• - 2nd generation Ray Tracing processor brings 2x acceleration in the most popular rendering programs such as Blender Cycles, Chaos V-Ray, and Autodesk Arnold.

• - The new option of hardware acceleration of motion blur during the Ray Tracing rendering technique has been accelerated as much as 5 times!

• - The most powerful GeForce RTX 30 models arrive equipped with 24 GB of GDDRX video memory to process complex 3D projects and multi-running applications and their processes faster.

Everything previously said and explained speaks in favor of the fastest generation of GeForce graphics cards in the form of RTX 30 series, which through numerous improvements of Ampera graphics processors, brings the biggest jump in performance so far, but also the quality of 3D animation. Superior gaming is not the only quality of the GeForce RTX 30 series, as it provides so far the biggest shift in the speed of applications for content creation and professional 3D modeling, video production, audio and image processing. In a word, in front of you is the best series of graphics cards.

We hope that we have explained to you a part of the great features and concrete benefits that you will have in your hands when switching to the new generation Nvidia GeForce RTX 30 series.