Hacker News Reader

Hacker News

Comment by pedro_hab | Hacker News Reader

Parent Full thread

pedro_hab

As a developer I am bit ashamed of this question, but I gotta ask:

What consumer apps actually use Neural Engines?

I think something like Photoshop, maybe. But wouldn't it just train a model and use it as regular code?

I'm interested in AI, but it's a joke to me about startups and jargon more often than not.

I feels weird to add this to all chips when I can't see that much usage.

View on HN

Pixelmator Pro uses it for some of its ML functionality. Image scaling can use it, and it provides a cleaner image when upscaling, removing some compression artifacts and just smoothing it out more naturally. I’ve found it can work well downsizing too, although less of an effect. They also have an ML auto-crop tool and ML denoiser. All of these will hit the Neural Engine pretty good.

Anything "predictive" probably uses Neural Engine. These are iOS features, but a lot of them apply to MacOS too

- Visual Lookup - Animoji - Face ID - recognizing "accidental" palm input while using Apple Pencil - monitoring users' usage habits to optimize device battery life and charging - app recommendations - Siri - curating photos into galleries, selecting "good" photos to show in the photos widget - identifying people's faces in photos - creating "good" photos with input from tiny camera lenses and sensors - portrait mode - language translation - on-device dictation - AR plane detection

Core ML API allows third party developers to use Neural Engine to run models

But which consumer apps use it? I know of a handful of photo apps that use it to enhance photos, but I'm not aware of any other types of apps.

I think there may be more indirect usage than direct usage. Little bits of neural engine usage are peppered through the native APIs.

Of course if you’re using third party libraries that don’t rely on macOS APIs this won’t be happening in your app.

could the M1 itself be using it for branch prediction?

Unlikely, the latency would be much too high. If it is doing neural branch prediction it's inside the CPU core itself.

Maybe the neural engine is ideal to scan your local image library to find kiddy porn:

https://www.theverge.com/2021/12/15/22837631/apple-csam-dete...

Reply on news.ycombinator.com