Why are we facing the end of the graphic driver problem?

The biggest problem that exists in the world of computer games is the issue of graphics card drivers . This is because unlike consoles where it is optimized for a specific configuration on computers, where there are thousands of different configurations. However, using RISC-V processors in GPUs could end this issue forever. Why? If you are interested, then keep reading, we will explain it to you

We have been launching for several years in which graphics cards that promised a lot have been less than promised due to bad graphics drivers. However, this problem could come to an end with the implementation of certain hardware changes that we are going to explain in detail and that could be the end of one of the biggest scourges suffered when playing on a computer.

the end of the graphic driver problem

The eternal problem of graphics drivers on PC

The GPUs in graphics cards have a much higher number of cores than what we see in the central processor, so they are much simpler and this makes their control unit much easier. Since it is an accessory element, the driver is left in charge of managing the resources and creating the lists of commands that the graphics card will execute.

Elden Ring PC Games

Ideally, the code that generates the graphics should be optimized as much as possible with the aim of occupying the greatest amount of resources possible at any time in the most efficient way possible. That is, without overloading each of the resources and without leaving some unused. However, this requires optimization not only for each architecture, but also for any graphics card configuration. Taking into account the number of models available, you can already understand that it is a work of Sisyphus for developers.

Well, a few days ago, we already told you how NVIDIA plans to use AI in its drivers, however, they cannot use any function that is not previously implemented in terms of hardware. This is where the other part of our story comes in, which has to do with the use of RISC-V on GPUs.

RISC-V has been in use for a long time on GPUs

We must assume that RISC-V is a completely free ISA and, therefore, unlike ARM, it not only means that those who use it do not have to pay a royalty, but also does not limit how to use it. This means that specific purpose processors can be used, which do not have all the instructions of a conventional CPU despite working the same and this allows them to be used in specific scenarios.

RISC-V

And it is that in the case of NVIDIA we find that the brand has RISC-V processors within its graphics cards. Since when? Well, from the GTX 1000 with Pascal architecture onwards. And on specific elements, like now:

  • The GPU command processor itself has been RISC-V for at least 5 years.
  • The second level cache management processes are carried out by one or several processors of this type.
  • If we are talking about consumption, voltage and clock speed management mechanisms, they are not only carried by temperature sensors, but also by cores of this type.

These are, therefore, specific tasks in which this type of core is used. So the use of RISC V on GPUs is not new.

And how can RISC-V on GPUs overcome the driver problem?

Well, with the simple fact of downloading the work to the CPU of having to manage the different processes that have to be carried out in parallel on the GPU. Specifically, it would be an evolution of the current command processor that would use inference algorithms through deep learning, one of the disciplines of artificial intelligence and, therefore, it would be using evolutionary algorithms, with the aim of being able to better manage the GPU resources.

NVIDIA RISC-V GPU

In any case, entrusting the work of the drivers to a RISC-V processor for the GPU placed in the command processor of the same has the following advantages:

  • It relieves the CPU of having to handle the events itself, causing its time in each frame to decrease and thus increasing the FPS rate.
  • They are in charge of precompiling shaders, an element of which they also offload to the central processor.
  • It allows adding new features much more easily, as well as more efficient bug handling.
  • For programmers it gives them a more accurate debugging ability of programs executed on the graphics chip cores.
  • It facilitates communication with the accessory blocks of the GPU without the processor having to act. That is, it makes the use of video codecs, the use of the DMA unit or the display controller more efficient. In specific terms, it allows you to reduce the different latencies more efficiently.
  • They can learn how each game works on our PC and manage the graphic settings automatically.

So, and in conclusion, it all comes down to the use of AI through RISC-V cores on the GPU.

The great common problem of Intel, NVIDIA and AMD

The problem with graphics cards is the large amount of resources in personnel and capital just for the creation of controllers or graphics drivers. Precisely if these do not work out in the end, this ends up affecting the performance of the hardware and with it its price, since you cannot sell a system with less performance at an equal or higher price. That is, a bad driver can make you lose a lot of money.

precio-gpus

This type of optimization, however, requires changes in the hardware and, therefore, the creation of new chips. Everything indicates that this is the great advantage that NVIDIA has over its greatest rival and that they are going to exploit it to the full in order to win in the performance tests and further consolidate their enormous market share. On the other hand, it is the demonstration of the consequences by AMD of having ignored the artificial intelligence market and having underestimated its application in gaming graphics hardware.

Fortunately, RISC-V is a completely free ISA to use and such solutions will not be carried out solely by NVIDIA. What’s more, other companies like Imagination Technologies are already implementing them in their mobile GPUs, where the problems are similar to those on PC.