Intel Shares Details on 5 Different Dedicated Arc GPUs for Laptops

You may have heard about it: after years of rumors and occasional false starts, Intel is using its expertise in integrated GPUs to enter the dedicated graphics market. The company’s Arc GPUs will be arriving throughout 2022, and Intel’s stated goal is to shake up a market that has been dominated by Nvidia (and AMD, but let’s be honest, mostly Nvidia) products for years now.

That process officially begins today with the launch of the first Arc GPUs for laptops. These mobile Arc GPUs first appeared back in January, when they were indirectly mentioned (without model numbers, specifications, or other technical information) in product announcements from some laptop manufacturers. Now that these PCs are close to shipping, Intel has a little more to say about what you can expect from its first serious effort to manufacture modern dedicated GPUs.

Arc mobile GPU branding reflects the division between Intel Core i3, i5, and i7 processors; the company is now releasing a pair of Arc 3 GPUs, with one Arc 5 and two Arc 7 models coming “early summer.”These five products are based on two different stamps. The smaller ACM-G11 die is the basis for two Arc 3 GPUs, while the Arc 5 and Arc 7 are based on the larger ACM-G10.

As we said last year, the main building block of every Arc GPU is the “Xe core”, each of which integrates 16 vector engines, 16 matrix engines, L1 cache and some other elements. The four Xe cores, combined with dedicated ray tracing modules and other fixed-function hardware, form a “render slice”, and each Arc GPU is built by stacking a set of render slices on top of each other. The full-featured ACM-G11 uses two render fragments, while the full-featured ACM-G10 uses eight.

The Arc 3 is clearly positioned as a competitor to low-cost dedicated laptop GPUs such as the Nvidia GeForce MX 550 and MX 570, which also use a 64-bit memory bus and a similar power budget (25W to 45W). It will also likely end up competing with the Radeon 660M and 680M GPUs included in the Ryzen 6000 series APUs, which are technically integrated system memory for graphics sharing but benefit from DDR5 support and the RDNA2 GPU architecture.

However, Intel’s performance comparison cut these competing GPUs out of the picture, opting instead to focus on how the Arc A370M GPU stacks up against the integrated Iris Xe GPU.

While Intel’s charts don’t include actual numbers for Iris Xe, they mostly show 1080p gameplay for slightly older titles like Doom Eternal or Destiny 2 running in the 30 to 40 fps range at medium to high settings on Iris Xe but smashes 60fps. on Ark A370M. For lighter competitive games like Fortnite and Valorant, 1080p gameplay above 90fps is possible on the A370M at high or medium settings.

The amount of boost you’ll see is highly dependent on the game; frame rates at least double for Doom Eternal or Fortnite, while Final Fantasy XIV and Rocket League improve by a much smaller margin. As with CPUs, Intel will also allow laptop manufacturers to use different TDPs for each of their GPUs, so an A370M in one laptop might look slightly different than the same A370M in a laptop with a larger heatsink and fan. It can also be assumed that the gap between the Iris Xe GPU and the Arc A350M GPU will be even smaller, since the A350M has two fewer Xe cores and a significantly lower clock speed.

All of the games that Intel highlights are primarily playable on integrated graphics at 1080p (using an average of 30fps as our “playability”threshold), and we don’t know how more demanding games like Cyberpunk 2077 or Microsoft Flight Simulator. 2020 will run on the A370M. But it’s clear that Arc 3 GPUs are positioned as a step up from integrated graphics for mid-range laptops, rather than a solution designed to power a dedicated gaming rig.