Today could also be Halloween, however what Intel is as much as isn’t any trick. Almost a 12 months after exhibiting off their alpha silicon, Intel’s first discrete GPU in over twenty years has been launched and is now transport in OEM laptops. The first of a number of deliberate merchandise utilizing the DG1 GPU, Intel’s preliminary outing of their new period of discrete graphics is within the laptop computer area, the place in the present day they’re launching their Iris Xe MAX graphics answer. Designed to enhance Intel’s Xe-LP built-in graphics of their new Tiger Lake CPUs, Xe MAX will probably be exhibiting up in thin-and-light laptops as an upgraded graphics choice, and with a deal with cellular creation.
We’ve been speaking about DG1 on and off since CES 2020, the place Intel first confirmed off the chip in laptops and in a stand-alone growth card. The firm has repeatedly been coy concerning the product, however at a excessive stage it’s been clear for a while that this was going to be an entry-level graphics answer appropriate to be used in smaller laptops. Based closely on the built-in graphics in Intel’s Tiger Lake-U CPU, the Xe-LP structure GPU is a decidedly entry-level affair. None the much less, it’s an necessary milestone for Intel: by launching their first DG1-based product, Intel has accomplished first step of their plans to determine themselves as a serious competitor within the discrete GPU area.
Sizing up Intel’s first dGPU in a era, Intel has actually made some fascinating selections with the chip and what markets to pursue. As beforehand talked about, the chip relies closely on Tiger Lake-U’s iGPU – a lot in order that it has just about the identical {hardware}, from EUs to media encoder blocks. As a outcome, Xe MAX isn’t as a lot a much bigger Xe-LP graphics answer a lot as it’s a further discrete model of the Tiger Lake iGPU. Which in flip has important ramifications within the efficiency expectations for the chip, and the way Intel goes about positioning it.
To minimize proper to the chase on an necessary query for our extra technical readers, Intel has not developed any type of multi-GPU rendering know-how that permits for a number of GPUs for use collectively for a single graphics process (ala NVIDIA’s SLI or AMD’s CrossFire). So there isn’t a technique to mix a Tiger Lake-U iGPU with Xe MAX and double your DOTA framerate, for instance. Functionally, Xe MAX is nearer to a graphics co-processor – actually a second GPU within the system.
As a outcome, Intel isn’t significantly positioning Xe MAX as a gaming answer – in truth I’m a bit of hesitant to even connect the phrase “graphics” to Xe MAX, since Intel’s superb use circumstances don’t contain conventional rendering duties. Instead, intel is primarily pitching Xe MAX as an improve choice for cellular content material creation; a further processor to assist with video encoding and different duties that leverage GPU-accelerated computing. This can be issues like Handbrake, Topaz’s Gigapixel AI picture upsampling software program, and different such productiveness/creation duties. This is a really totally different tack than I believe lots of people have been envisioning, however as we’ll see, it’s the route that makes essentially the most sense for Intel given what Xe MAX can (and may’t) do.
At any fee, as an entry-level answer Xe MAX is being setup to compete with NVIDIA’s last-generation entry-level answer, the MX350. Competing with the MX350 is a decidedly unglamorous process for Intel’s first discrete graphics accelerator, but it surely’s an correct reflection of Xe MAX’s efficiency capabilities as an entry-level half, in addition to a ripe goal since MX350 relies on last-generation NVIDIA know-how. NVIDIA shouldn’t really feel too threatened since in addition they have the extra highly effective MX450, however Xe MAX has an opportunity to at the least dent NVIDIA’s near-absolute cellular marketshare by going…
