Arm ships its neural upscaler in a Xiaomi foldable a year ahead of its own frame generation

The silicon is on sale ^[5], and Android rendering teams face a calendar problem before a technical one. Of the three techniques Arm quotes its headline number against, one is available to the first titles, frame generation lands in early 2027, and denoising follows that ^[4][10]. The plumbing work is now; most of the feature set is later.

Arm claims up to four times the framerate with neural super sampling, frame rate upscaling and denoising running together ^[4]. The Verge reports that half of that comes from frames the phone imagines between real ones, and that frame generation as shipped by Nvidia, AMD and Intel normally costs latency ^[8]. Strip the invented frames out and the portion that reduces actual rendering work is closer to two times ^[14]. Two times is still a lot of pixels you no longer have to draw, and it is the figure a schedule can survive.

The conventional part of the chip is having an ordinary year. The 85 percent uplift Arm quotes is measured against the GPU Xiaomi shipped previously; against Arm's own last flagship, non-AI gaming improves by 14 percent ^[2]. The neural path is the product here, not the raster gain.

Heat is where this either works or does not. Arm's Chris Bergey puts a whole phone running these chips at no more than 2.5 watts, with 1.5 of that for the GPU ^[11], which leaves a single watt for the display, the radios and everything else ^[16]. He describes the practical result as 1080p at 60fps for three hours where it used to be 30 minutes ^[12], a sixfold increase in session length ^[15]. Xiaomi is separately showing the Pad 9 Pro Max, which pairs the same GPU with a larger battery and more surface area for cooling, at 3.2K and 120fps ^[9].

Every one of those figures is Arm's or Xiaomi's. The Verge's report names no shipping game using the neural path and carries no third-party measurement of the claimed framerates or the power budget ^[7]. Bergey's own framing is that this is mainstream, not overclocking, and that AI will replace traditional rendering anywhere it can ^[13], which is a statement about intent rather than about install base.

Super sampling reduces the pixels you render, which converts directly into watts and session length, and it is the tier that exists today ^[10]. Frame generation shortens the interval between displayed frames without reducing the work behind them, and pays for that in latency ^[8], so in a competitive shooter it is a cost wearing a framerate number. Denoising only pays if you already ship ray-traced lighting. A title that throttles thermally after twenty minutes has a reason to do the super sampling integration for one device family this year and treat the rest as future work. A title where input latency is the product should read Arm's four times as roughly two times ^[14] and price the integration against a device population that, for now, is one foldable and one tablet in one market ^[5][9].