cylon13

It’s never bad advice to consider something.

Is your assertion that it takes more time for a CPU to read values out of a 30 byte struct and do a couple shifts and branches than to parse a JSON representation?

I think that goomba bumped into the squished goomba Mario had just squished. Mario was just a bit to the left so the flat goombas hit box stuck out to the right a bit and the other goomba hit it.

Yeah it doesn't really land well as a joke because it's the very first thing. Joking hot takes at least need a bit of warm up so readers know what's going on, whereas this just opens with a no-context dis at women where it's not even clear what the joking read is supposed to be talking about. What hard work? What's happening?

A notable improvement over the GLHF strategy for interacting with GPT models.

There's not much data in the performance section of the front-page readme of the repo, but the one bullet point there seems promising:

Coremark 1.0: ~7% slower than native [0]

[0] https://github.com/turbolent/w2c2#performance

Do you never find yourself interested in the results of a code change you have just made?

Might be fun to use diffusion to make adversarial/false answers to the pixelated scenes, depending on how pixelated they are. If they're quite pixelated you could probably come up with some crazy alternatives.

This is incorrect

Why would you expect a law professor to be an expert in avoiding being framed?

I use arch and I don’t like fiddling with my system, I just wanted something light. After the painful initial install a few years ago it’s been really stable on the laptop I used to run xubuntu on. I switched because I was having problems with ubuntu running stuff I didn’t care about in the background and eating up CPU time.

There are fundamental issues with x86 that make it impossible to match the efficiency of ARM. Variable length instruction coding for instance, which means a surprising amount of power is dedicated to circuitry which is just to find where the instruction boundaries are for speculative execution. Made sense in the 80s when memory was scarce and execution was straightforward, but now it’s a barrier to efficiency that’s baked right into the ISA.

It's not solving a technical problem. It's trying to avoid points where the user has to make an active decision to continue looking at whatever is being served and thus have a chance to decide to leave.

You obviously live in a much nicer place. Sounds like the GP lives in a rougher neighborhood in a larger city, which if you’ve never experienced, can definitely leave you wanting to avoid having to shop in public.

Ah yes, the goes-to operator -->

I want more freedoms than just the ability to complain about not having freedom though.

With shared array buffers it’s not actually a big problem. We’re building a multithreaded game that targets wasm and we just spin up workers up front which all use the same shared memory, and then just use standard concurrency primitives to schedule work on them. Just have to be careful not to block on the main thread, and you have to serve all your assets from the same domain for browsers to allow you to make shared array buffers.

Even knowing what they all are it’s annoying to read compared to having them spelled out. I’m constantly pausing to translate. If you’re talking to someone IRL you just say “hardware vertex shader,” not “HW VS”.

You can, but the docs for Deref beg you not to use Deref for types that aren’t smart pointers because of how confusing and surprising the results can be to an unfamiliar reader of the code.

Writing tests and doing TDD aren't the same thing. The commenter is talking about TDD in particular I think, not saying "unit testing has no success stories".

There’s only two possibilities: I’ll either win the lottery or I won’t, so it’s a 50/50 shot!

Unless you have a weird skull

Any noise strong enough to have a good chance of hiding the signal would completely defeat the benefit of having dynamic frequency scaling in the first place, I think.

You can’t just slap “perhaps” on arbitrary statements and expect to not be corrected if they’re inaccurate.

Scrollbar blocksize wouldn’t usually be useful for estimating content length anyway because of comment sections.

Diffs aren't impossible over UDP. The client should be sending to the server which states it has locally, along with inputs. Then since the server has a (potentially non-exhaustive) list of recent states it knows the client has seen, it can choose to make a diff against the most recent of those and tell the client which state the diff is against. Then the client and server just need to keep a small buffer of pre-diff-encoded packets sent/received.

Cool, when should I expect this fix to be deployed globally? Looking forward to it!

I definitely read the title as sarcasm.

Language is dynamic and flowing and changing, and it's going to be whatever it is no matter how angry it makes you. Better to go with the flow than to scream into the wind. Is "fuckwits" proper English? Who cares?

I agree with the sentiment about Slack being slow and bad. It’s not exactly the same as IRC though. There’s voice chat, screensharing, I can drag and drop video files without worrying about hosting, etc. People choose Slack over IRC for a reason, it’s just too bad it’s a private centralized thing written as a resource hogging electron behemoth.