Codex and I were trying to infer why Jev gave a certain parameter a given score. So I asked codex to produce a list of like 30 plausible reasons Jev might've selected that and then presented Jev with the initial prompt, followed by "You scored this with XYZ. What was your reasoning for doing this?" And then allowed it to do a noul value for each of the reasons codex generated. Felt like those people that give their dogs the buttons to push.
I also signed up for a new account and it's right back to working how it used to. They absolutely do not consume tokens equally across accounts. I did TONS of work on the new account and barely made a dent, even on Astra. Old account chews through 20% like it's nothing
That sounds like large amounts of context (maybe from memory or history?) are chewing through your quota, context that hadn't yet had a chance to build up in your newer account. That's just a guess, I'm not an account holder there or anything so I can't try to re-create it on my end. You could try to erase history on your old account too, too see if it makes a difference. 'Course, then you'd lose all that history!
Apple Silicon was autopilot. Or more specifically the insanely cool decision happened under Jobs, and everything else is just coasting (i.e. repeating the same business success again and again without taking very many risks). Their first custom core design was the A6 in 2012, a year after his death, but they bought the team that made the design (P.A. Semi) in 2008. It took the team about a decade and change to scale up to something that could power a laptop, but the decision was made very early on and was specifically part of the same set of decisions that Apple has been coasting off of ever since. If Apple hadn't bought its own chip design firm they likely wouldn't have remained competitive long-term, in the same way that their computers got really uncompetitive for a few years before every major architectural transition.
Unless by "Apple Silicon" you specifically meant "the decision to transition from x86 to ARM". There is actual work there, but not a lot of risk. Apple famously maintained an internal x86 port of Mac OS X for years after Rhapsody was shitcanned and renamed "Mac OS X Server", so doing the same with ARM isn't that worse. iOS was derived from OSX; and just having iOS already up and running on ARM is already a good deal of the way towards a macOS port to ARM. In fact, it got easier as iOS got more technology hand-me-downs from macOS.
Over the course of about a month, this went from being vaguely detectable on my mom to very overpowering (to me). I have a particularly sensitive nose, but now I can be across the room and if she comes over to our house I can smell her by the time I enter the room, even if she's just been let in. It's been about 1 year now and it's just getting stronger.
Most providers do what's called "prefix caching", where each turn in a session is cached such that sending new messages with the exact same "prefix" (set of previous messages) gives you the cache read price on that input instead of the full price. As long as you're not changing your system prompt, available tools, etc mid-session, you automatically benefit from this.
Cursor is like one of the least economical ways to use Claude and OpenAI, virtually nobody does it. You use Cursor for their autocomplete/composer + grok. For using Claude and Codex, you get immeasurably more bang for your buck just using them directly.
reply