There is no physics based reason why it couldn't work. If the industry really wanted to do it they could. But they don't. The primary reason is LPDDR just has too many pins. A DDR5 SODIMM has 262 pins and is an unwieldy beast. LPDDR5 has 644 pins.
LPCAMM2 really shows the trade-offs. It adds a lot of bulk and cost, and repairability hasn't been valued high enough by the market to cover that overhead for most consumers. That's why Micron exited the market they played a big part in founding.
They should share the battery life numbers of default shipping configuration while running Linux with whatever settings they want. Then publish the configuration and settings. Same as every other manufacturer.
The expansion card system seems like something I would actually really like, especially as a hardware engineer. But the more I thought about it I couldn't really think of any compelling expansion cards that were worth the effort. So I figured I would look at what was in their store to see what other people thought up, and there isn't really any 3rd party store that I could find.
I think a lot of the disconnect in the programming world is we treat all programming as equivalent and it's not.
There really are many programming jobs that are rote and I have no problem believing that an LLM based tool can learn the pattern and regurgitate with the tweak de jour. In those jobs LLMs probably do increase productivity.
But there are other programming jobs that are not rote and there is no pattern to learn because you haven't done the thing yet. LLMs aren't any more useful than a normal base library would be, and if you're already good at using a library of code, they're not a productivity booster and often, in my experience, a hinderance.
I think another point is the prompt actually forces the engineer to spend a moment to actually think about what they're doing and make some kind of plan. Pre-AI tools way too many programmers just jumped straight into problems without thinking what they were doing figuring they could code their way out of anything and ending up stuck in some cul de sac and having to back track. And if they just stopped and made a basic plan they wouldn't have that issue. Forcing engineers to make a plan, who wouldn't otherwise do so, before they start, could definitely be a productivity booster for them.
> forces the engineer to spend a moment to actually think about what they're doing and make some kind of plan
i thought so too, until i observed the exact opposite at $job
basically i see people just jump into claude and tell it to do x, then x is wrong or not quite right, so they go down a rabbit-hole fixing up things again and again burning tokens like a bonfire till it sort of works... i see very little upfront planning from where im at but maybe this is just my bubble...
If AI just means automation, then sure. We absolutely need more automation and if LLMs are not the mechanism then something else better be. More automation is the life blood of our industry. But are LLMs a game changer or today's fuzzy logic? [1] Time will tell...
P.S. I'm not saying fuzzy logic doesn't have applications, I know rice cookers are a thing, but I think it's safe to say we have other options for controlling non-linear systems these days.
And I've even heard rumors of software engineers that don't even write apps or write code that runs on the internet at all. They say some of them don't even use javascript or python! The horror.
"AI-insiders" are trying to market their tools to you. See Anthropic's continuous lithany of "all programmers will be replaced in 6 months" while they struggle to make their TUI API wrapper consume less than 2-4 GB of RAM (they brought it down from 68 GB[1]), or have a decent uptime.
I have heard many replace their phones due to dropping them and becoming unusable. But everyone uses a case now and the build quality is generally better that one mishap does trash the phone. Most people I know getting new phones now did so bc their old phone "got too slow to be usable." I believe that's a matter of new OS versions really are much heavier. Both my last 2 phones I had upgraded bc I went one version too far and had a nearly bricked phone.
Any phone that gets more support than it should have, such that the only OS you can install is too slow to make using the device enjoyable, makes it more likely for the device owner to throw that device out, and then it becomes e-waste.
It also harms software preservation. Sure, we have IPSWs for every single public build of iOS that exists (and if you dig around, probably a ton of betas and even internal builds). But you can't really do anything with any of them once you get to the point in the iOS product line where things were sufficiently hardened
LPCAMM2 really shows the trade-offs. It adds a lot of bulk and cost, and repairability hasn't been valued high enough by the market to cover that overhead for most consumers. That's why Micron exited the market they played a big part in founding.
https://www.ifixit.com/News/95078/lpcamm2-memory-is-finally-...
reply