Intel Formally Announces Iris Xe MAX Graphics, Deep Link

Laptop vendors have recently released Xe MAX graphics in the form of discrete Intel graphics that will appear in laptops in the coming weeks. The announcement was somewhat unexpected and Intel did not inform the media in advance, while today – in an unusual announcement made on Saturday (Intel says it coincides with the appearance of the system, apparently for the first time in China) – the company officially announces the release of Iris Xe MAX.

The Iris Xe MAX graphics are based on Xe-LP, as in Tiger Lake, while it is the first discrete GPU based on Xe on the market. Xe MAX will be launched in the 4th quarter. They can be found in several notebooks in the second quarter. It is interesting to note that Intel has also indicated that it will introduce discrete Xe-LP graphics in its desktop prices in the first half of 2021.

The Iris Xe MAX graphics include 96 run blocks, a 1.65 GHz clock speed and 4 GB of video memory, while the rest of the Xe-LP resembles the Gen12 graphics currently found in Tiger Lake.

Besides Xe MAX hardware, Intel is also talking about Deep Link as a new software platform for its CPUs and GPUs. Deep Link offers a structure that can cover multiple computer systems in one way. Of the few information disclosed so far, it mainly enables the use of media engines and executive blocks of discrete Xe MAX and integrated Xe hardware in single and parallel modes.

Intel is preparing Deep Link for OBS, HandBrake and other software, while they expect it to appear soon in both Blender and CyberLink, Fluendo and Magix. When I was asked about Linux support for Intel Deep Link, I was told it would be supported by VTune and OpenVINO because I hadn’t seen Deep Link in their different source code repositories. In the coming weeks you will have more news about the support of Linux software.

With Xe MAX and Deep Link, Intel proves this concept by offering twice as much video encoding performance on the H1’2021 as on the RTX 2080. The RTX 2080 SUPER is equipped with NVIDIA NVENC’s own video codec SDK. They are a bit ahead of current Xe graphics, but after using Deep Link they project it about twice as fast as RTX 2080 encoding.

In its Windows gaming benchmarks, Intel uses the Iris Xe MAX like NVIDIA’s competitive and sometimes superior GeForce MX350.

Xe MAX models include the Acer Swift 3x, ASUS VivoBook Flip TP470 and Dell Inspiron 15 7000 2-in-1. These laptops with Intel Tiger Lake processors and Xe MAX graphics should launch this quarter – or, as reported today, in selected markets.

Considering the excellent performance of Gen12 Xe Graphics under Linux for OpenCL / oneAPI Level Zero / Vulcan, we are very excited to see Xe MAX graphics on more laptops and possibly desktops starting next year. According to our tests, Tiger Lake has experienced a significant increase in the number of Gen9 (and Gen11) images a long time ago. Linux driver support for Gen12 Xe graphics was implemented immediately, and new optimizations were also released in Mesa and in the Linux kernel. It will also be interesting to see what happens with deep link support for open source/Linux.

If you liked this article, remember to join Phoronix Premium to see this site without advertising, multi-page articles on one page and other benefits. Tips from PayPal are also accepted. Thank you for your support.

Related Tags:

intel iris xe max vs mx350,intel deep link,intel iris xe max graphics vs gtx 1650,iris xe max specs,iris xe max benchmark,intel iris xe vs nvidia