Also, if they make a neo 2 from the A19 with 12 GB that'd be cool.
It’s just a software restriction. I’m sure they have macOS running on iPhones and iPads internally, probably both through a VM and natively. But it’s very Apple to block this use case.
Small, memory efficient programs run better everywhere. It feels like theft to make consumers pay for our inefficient software. Electron / react on the desktop is such a rort.
Sure. Right along with Intel Core 2 Duo and Windows Vista.
Slim the fat!
I am not sure single digit tk/s on weak model like ChatGPT OSS 20b @ Q3_K_M is going to be worthwhile.
Not to mention this has to be <0.01% of users who "want" to do this. The iPhone tooling would make it hard to even configure and spin up a local LLM.
You can't possibly expect Apple to take that market segment seriously.
According to what? (I don't disagree, just curious)
How about avoiding that scenario again? Is that a good enough reason? Being a diligent, informed consumer?
I have no idea how much ram my phone has. So long as apps run well, why would I care?
> the most important property of compute devices right now
Memory didn't matter when iPhone 12 was released, but these days it determines which local LLMs you can run on device. That's the point of that comment.
The net price for a new phone then can be considered a relatively inexpensive hobby when looked at from a per week basis.
> iPhone
Pick one.
Sure, but this situation where "all else being equal" doesn’t exist.
especially for Apple, I imagine if I were buying a phone knowing they historically throttle processors on older models, I'd try to at least control the one aspect that wouldn't make sense to throttle
The A19 Pro had 12 GB and was around 77GB/sec memory bandwidth for reference. If I was a betting man, I'd say the memory bandwidth on the A20 Pro is the same.
Since 1984, Apple has targeted regular people, not spec nerds. You're not the market.
Threadcount, Watts, cubic inches: different products use different misleading measures.