Intel launches datacentre GPUs for AI and gaming, while Google shuts cloud computing platform

Intel launches datacentre GPUs discretely which are designed to power Android-based cloud computing. Intel said it is rolling out for Datacentre GPUs and its $329 Intel Arc 770 for mid-range gaming computers as well. The timing might not be great as there are many GPU chips available in the market.

On the other hand, the cloud gaming market has not been great this year as Google shuts cloud gaming service Stadia by January 2023. After three years of launching, Stadia is finally put down to rest. The company said it has proven over the years and will continue offering that technology and cloud computing. Additionally, other competitors like Amazon Luna, Microsoft's Xbox Cloud Gaming, and Nvidia's GeForce Now all continue to provide their services. The platform had originally planned to go big on launching integrations and other in-house studios exclusively but now that all isn't happening anymore, shutting down.

Cloud gaming is a promising concept

For gamers who want to enjoy games without having to build a specked-out system that can cost a lot of money and, in certain circumstances, have high maintenance and energy costs as well, cloud gaming solves several problems. Instead, if your internet connection is fast enough, your computer can play games over the air, allowing other people to handle the processing while you play it on your monitor.

The concept gained traction over the past few years, which is why there are now several cloud gaming platforms available in addition to Google Stadia. Amazon's Luna and Nvidia's GeForce Now are more cloud gaming options in addition to Microsoft's Xbox Cloud Gaming, Sony's PlayStation Plus, and PlayStation Now from Sony. While not all of the aforementioned are supported in every country, you might find more users than you expect in those that are.

More gamers now have a lot faster connection to the web than they did just a few years ago thanks to the introduction of 5G in more areas and the rapid acceleration of internet speeds. This allows a lot more people access to cloud gaming. Other benefits of the service include the ability to play on practically any compatible device and restart your in-game progress from virtually any other device. Without having to physically transport any of your progress data, you could finish one boss battle on a game on your home PC and then play the next battle on your phone at a friend's house.

What went wrong with Google Stadia?

Despite some early issues, Google Stadia got off to a good start when it launched back in November 2019, allowing users to join up and begin playing games without the necessary hardware. Additionally, Stadia's server speeds at the time were faster than those of competitors like the PlayStation Pro program. However, barely a year after launch, the platform's fortunes began to fast deteriorate due to technological problems, a lack of features, and simply not enough users.

Google reportedly stated that the program "hasn't gotten the traction with users" that the business had hoped for.

But this has been the situation for a while Google also introduced "Immersive Stream for Games," a version of the service that third parties could license, in March of this year. Additionally, Google is said to have stated that it sees "potential to use this technology across other sections of Google," such as YouTube, Google Play, and the business's augmented reality (AR) initiatives.

Intel's discrete launch

Intel's Arctic Sound-M discrete processing units for data centers have already begun selling for approximately a month. Intel formally unveiled its Data Center GPU Flex series. When the products reach the necessary levels of maturity, they will be made available via a variety of Intel partners. The new graphics boards are based on the company's Arc Alchemist graphics processors and are targeted at a variety of data center applications.

Two base graphics card models are available from Intel's Data Center GPU Flex family of products: one is intended for performance-intensive workloads, and the other is made for extremely dense installations. The single-chip Flex Series 170 is designed for tasks that demand maximum performance and is based on one ACM-G10 GPU with up to 32 Xe cores (equivalent to up to 4,096 stream processors) and 16GB of memory. The dual-chip Flex Series 140 card is designed for high-density PCs and has two ACM-G11 GPUs with 16 Xe cores and 12GB of memory.

Server-grade graphics cards from Intel's Data Center GPU Flex family are fully supported by the company's cutting-edge application programming interfaces and tools, such as one API, OpenVINO, oneVPL, and VTune Profiler.