Under the threat of home-made chips from Apple, processor giants AMD, Intel and Nvidia are fighting back. The three companies took advantage of CES to announce new versions of their flagship products responding to the need to provide more performance in a more efficient way. Among the techniques they covered, they all aim to increase performance and efficiency by harnessing the versatility of the Windows ecosystem for new ways for CPUs and GPUs to work together. These measures are aimed in part at countering Apple, which has achieved impressive performance with its integrated GPU, but has not (yet) developed its own discrete GPU.
As AMD is the oldest supplier of processors and GPUs, it’s no surprise the company is embracing smarter power transfers between the two. The company has upped its SmartShift technology for routing the compute load between CPU and GPU to SmartShift Max. The company says this technology can now speed up a wider range of games and workflows, citing examples of games where the acceleration went from zero to a 13% improvement.
Discrete GPU newcomer Intel has showcased a wider range of integration benefits between its CPUs and Arc discrete graphics processors under the Deep Link banner. While the company briefly discussed a series of power routing tips that can improve gaming performance, it also showed how its discrete GPUs can work in conjunction with the integrated GPU of its Core architecture. Using this approach, for example, the video editing application DaVinci Resolve could distribute the encoding of alternating frames of a video between each graphics processing option, which Intel said could produce a 40% improvement in time. rendering.
Nvidia is also making a strong comeback
Finally, Nvidia’s absence in the area of PC CPUs has not prevented it from developing its own approach to the partnership between its discrete GPUs and advanced CPUs. As part of the fourth generation of its AI-powered Max-Q approach to system design, the company presented in detail CPU Optimizer, a low-level framework that allows the company’s GPUs to regulate performance. , the temperature and the power of the new generation CPUs.
This results in more efficient performance of the CPU, which allows more power to be transferred to the GPUs. The company also showcased its second-generation Battery Boost technology, which takes into account factors such as CPU and GPU power usage, battery drain, image quality, and performance in real time. number of frames per second, to improve battery life by up to 70% when playing games.
The top three PC silicon vendors showcased their approach to closer collaboration between CPU and GPU, as well as advancements in AI-powered upsampling, such as Nvidia’s Deep Learning Super Sampling, in the context of improving the performance of games. Such improvements go hand in hand with improvements for many creative software applications. Over time, these improvements could spill over into more common applications, especially as augmented reality and virtual reality applications evolve.
Source: ZDNet.com