apatheticonion

The OLED screen is actually a nice development given the original MBP screens had noticeable latency while gaming.

If only you could run Linux on the damn thing so you could play games now.

Where is the reverse engineering army when you need them lol

Agreed. I get the best outcomes out of agents when I use them as essentially autocomplete.

"Render what I was thinking"

I just double check to make sure it aligns

I see your point. In this case I'm intentionally reinventing the wheel for the sake of correctness and quality.

For the end user, the outcome will likely be the same when compared to opencode or similar, but the journey will be nicer

I expect no one to actually use it though, however nice it might be to hope for that

Haha, in my case, it only renders when frames are needed. Otherwise it's basically an image at 0fps.

Here's an anecdote. I have repurposed my 6 year old 32gb M1 MacBook Pro to run Qwen 3.6 35b a3b and I used to give it nightly maintenance tasks that span my OSS github repos.

I gave it mundane tasks like updating dependencies, writing changelogs, etc. It ran about 2 hours a night.

My MBP consumes 0.1w idle and about 10w running inference. I pay $0.33/kwh. If it runs for 2 hours a day over 365 days - it will cost $2.41/year (excluding idle time).

Excluding the cost of hardware - the running costs alone are more expensive than DeepSeek's API and the inference quality is much lower. The economics are even worse on AMD AI Max mini PCs and worse still with a dedicated GPU setup.

That said - my Apple Silicon example is orders of magnitude cheaper than western inference providers and offers unrivaled privacy.

Perhaps a local AI setup would be compelling for small/medium business users... and it would be practical if the hyperscalers didn't buy all of the potatoes on Earth.

It's just a subtle reminder to learn mandarin, haha

Me, watching the AI world from the sidelines while using DeepSeek at $0.15($0.003 cached)/$0.60 - still bitter about the pricing changes.

It might not be as smart, though my employer gives me access to Claude, so I tried giving the same task to Opus 5.5. Both DS4.1 Flash and Opus5.5 produced nearly identical outputs.

After that, my company switched the entire engineering department to Azure hosted DeepSeek4.1 and costs were cut by 3/4. So far, no one has really noticed the change.

> nor do I want that from them.

This perspective I can't wrap my head around. Like I get that you wouldn't use it, and that's fine - but if Linux was available and we made it incredible, that would push Apple (and Google) to do better. So you'd get a better MacOS / iOS by proxy.

Like, don't we want competition? What motivates people to desire an unchallenged status quo?

Do you get that on the Apple App Store?

I hope so. We haven't yet seen LLM-led reverse engineered firmware, but here's hoping!

Haha, shots fired.

It's better than Windows 11 at least - though we probably can't credit that to improvements to the Linux DE