HardwareNews

Intel shows the dedicated GPUs Raja Koduri is working on

Intel shows the first diagrams of the graphics processors, in which Raja Koduri is already working, to develop a dedicated graphics card, with a GPU that competes with AMD and NVIDIA.

It was something sung and it was something we expected. When Raja Koduri joined Intel, we were already very clear that he was not going to enhance the company's iGPUs, but that it was for something else, specifically to develop a dedicated graphics card that could compete with AMD and NVIDIA. Intel has not managed, at any time, to develop a graphics solution for the gaming market, you just have to see Larrabee, a resounding failure, which was nothing more than a ridiculous sound for the company.

We do not know the conditions of Koduri, but what is clear is that he will not lack money and Intel will surely give him everything he asks for. The company has already published some images of what is a GPU prototype, which is still in development, but looks pretty good. It seems that this new attempt by the company to get into such a lucrative market, such as that of GPUs, and more so at the moment, with the mining of cryptocurrencies, is going well and is based on Intel HD Graphics iGPUs, which They are a chestnut, but with the hand of Raja Koduri, they can become quite an interesting GPU.

The first of the dGPU prototypes is under development and is based on the 14nm process, at least for now, as shown in the image. This first chip is divided into two key elements: one is the GPU itself and the other is the system controller. The second version of this chip or the second chip is an FPGA that is combined with the system through a bus. The GPU, for the moment, is based on Intel's Gen 9 architecture, the same one that is integrated in the Skylake and in the image you can see three clusters of Execution Units.

There is not much more data regarding the Execution Units, integrated in each of the clusters, but, for the moment, it is information that is not relevant at this time, since they are different development tests. These clusters are connected by means of a special system for managing consumption and working frequencies, which manages the power and frequencies of each of the Execution Units, to be adjusted efficiently.

A double clock system has been implemented, to double the working frequencies, which increases performance and that is much more than what the Gen 9 iGPUs of Intel processors can say. We assume that, once performance is optimal and consumption is well controlled, a new architecture will be moved and the Execution Unit counts will be improved in each of the manufacturing processes, which will allow having a product that competes with NVIDIA. and maybe end up destroying AMD in the graphics market, where it is not particularly good.

Source: TPU

Show more

Robert Sole

Director of Contents and Writing of this same website, technician in renewable energy generation systems and low voltage electrical technician. I work in front of a PC, in my free time I am in front of a PC and when I leave the house I am glued to the screen of my smartphone. Every morning when I wake up I walk across the Stargate to make some coffee and start watching YouTube videos. I once saw a dragon ... or was it a Dragonite?

Related publications

Leave your comment

Your email address will not be published. Required fields are marked with *

Button back to top
Close

Ad blocker detected

This site is financed through the use of advertising, we always ensure that it is not too intrusive for the reader and we prioritize the reader's experience on the web. But if you block ads, some of our funding will be reduced.