WXLCKNO

Wow lol. I just posted the exact same comment, there are dozens of us! I literally cannot play rocket league without the steam controller for this reason.

Also set rotate left and right to the grip triggers (roll in aviation terms I guess).

I love my OG steam controller still. I can't tell if this new one has the dual stage triggers like the og (like if there's an additional click on full trigger pull).

I used that to set things like boost in rocket League and it felt super intuitive.

Bit beside the point but Windows 11 is the first version since Windows 3.1 that I haven't used.

Nuked my Windows 10 install and put Pop OS on it + a MacBook separately.

I hate ChatGPTs writing style so much and as you said, here it's chilling.

As someone who doesn't really drink and never has alcohol at home, is it just that people buy beer/wine and drink on a daily basis for fun? The wine I feel like you can pair with food and feel classy as an excuse but beer that ain't the case.

I realize how completely dumb this question might sound.

I hate casino games but I enjoy poker. I don't play often anymore but I used to play online a lot, multi tabling.

There's something very satisfying about making thousands of decisions that have a positive expected value and seeing the math win throughout the variance, resulting in a proven edge.

Like anything in life, it's fun to get good at something and at some point it's beyond the math and theory, it feels like the cards become transparent at times, you just have a gut feeling that this hand, this person is trying to bluff. It can be based on what you know about them or just how fast they clicked or their bet size, but this guy wants you to fold so you'll call with almost nothing and still win.

But overall as a profession fuck poker, you don't contribute anything to the world.

Now verifiably useless in real life

I find it less offensive on YouTube honestly. Those playlists you mention specifically are like a novelty that you can explore for fun.

But on Spotify when it gets shoved onto your playlists without warning..

I've for Plex + PlexAmp for music , does jellyfin have a good mobile app for audio?

Agreed, on YouTube the AI slop music has been trending even more. I hadn't experienced it on Spotify however and only last month they had stated that AI music would be tagged as such.

I agree with this and actually Claude Code agrees with it too. I've had Codex cli (gpt-5-codex high) and claude code 4.5 sonnet (and sometimes opus 4.1) do the same lengthier task with the same prompt in cloned folders about 10x now and then I ask them to review the work in the other folder and determine who did the best job.

100% of the time Codex has done a far better job according to both Codex and Claude Code when reviewing. Meeting all the requirements where Claude would leave things out, do them lazily or badly and lose track overall.

Codex high just feels much smarter and more capable than Claude currently and even though it's quite a bit slower, it's work that I don't have to go over again and again to get it to the standards I want.

You know you can use CMD+backtick (CMD+`) to cycle between windows of the same app? Add shift to go in reverse.

Or otherwise you can enable the app exposé feature to swipe down with three fingers and it will show you only windows of the same app.

Pretty funny how they're actively targeting M1 users with their marketing copy with this release.

Sorry you made your first gen chip so good that I don't feel the need to upgrade lol.

And I love it

> I now feel weird posting Nice Photos to social media. Who are those for, really?

I have a private Instagram account with ~100 followers of people I actually know.

And I still agree with this although I'm trying hard not to. I'm just trying not to be cynical.

But I hate this feeling of my life experiences seeming like they require external validation.

Am I doing things just so that others know I did them? Is it just a curated feed I can look back on by myself? Social media feels like it steals our lives away from us.

"Write an extremely cursed piece of Python"

text-davinci-001

Python has been known to be a cursed language

Clearly AI peaked early on.

Jokes aside I realize they skipped models like 4o and others but the gap between the early gpt 4 and going immediately to gpt 5 feels a bit disingenuous.

It's all so exhausting to see.

Also baffled to see Zhu Su not in prison and tweeting.

It doesn't even have that, knowledge cutoff is in 2024.

It's very doubtful that they'd have any kind of magical breakthrough that makes the model anything other than incrementally better right now.

o3 feels pretty good to me as well but o3-pro has consistently one shotted problems other LLMs got stuck on.

I'm talking multiple tries of claude 4 opus, Gemini 2.5 pro, o3 etc resulting in sometimes hundreds of lines of code.

Versus o3-pro (very slowly) analyzing and then fixing something that seemed completely unrelated in a one or two line change and truly fixing the root cause.

o3-pro level LLMs at reduced cost and increased speed will already be amazing..

Glad I could help :)

I think they have like a one time pop up to explain which is easy to miss.

You're gonna have to explain that one.

Do you mean the new One UI update that made the notification pull down split into left and right swipes instead swipe down and then swipe down again? Because if that's what you mean, you can configure it to be the way it used to be again.

Little pencil button, then panel settings and choose together instead of separate.

Not typing hg is the best part of my day

I pulled the trigger on a full Firefox migration a few months ago because of ublock.

Google Chrome had browser change inertia going for it, nothing else.

What did you use before Grok 4 came out for architecture consulting, Gemini 2.5 Pro?

In the system I'm building the main agent doesn't have access to tools and must call scoped down subagents who have one or two tools at most and always in the same category (so no mixed fetch and calendar tools). They must also return structured data to the main agent.

I think that kind of isolation is necessary even though it's a bit more costly. However since the subagents have simple tasks I can use super cheap models.

Same for me.

Except I'm never gonna give Elon money, I don't care how good his model is.

"We’ll be using FastMCP, a lightweight framework for building model-control-plane servers"

Article written by AI (and not reviewed by humans) that doesn't know MCP is model context protocol?

Or author being intentional for some weird reason?

Right but it probably takes like 8-10 H100s to run Claude Opus for inference just memory wise? I'm far from an expert just asking.

Does "one" Claude Opus instance count as the full model being loaded onto however many GPUs it takes ?