This is an option.

The problem is, while faster on paper, its not really any faster than my 3090, especially in spots where I need its performance most. It’d be a downgrade.

I could get a second 3090, I guess, and have some redundancy in case one fails. But I’m on an ITX board with 1 slot.

I’d be WAY more interested in a roughly equivalent 9000 series or Battlemage card, but alas, there is none.

Oh, I know. I bought my previous GPUs in the peak of cypto busts.

But I need 24GB… My plan was to eventually buy a used 4090 when they get cheaper, or 24GB Intel/AMD card. But it doesn’t feel like AMD/Intel are interested in big GPUs anymore, and the 4090 just keeps going up in price.

So I FOMO bought an SSD and SD card when all this started.

Not regretting it.

…But now I’m considering hoarding a spare GPU. Is that nuts? I don’t need it, I don’t want to hoard, but If my old Ampere card suddenly dies, is it gonna take $2000 to replace?

Start posting. Please. I am begging you.

• An American tired of wallowing.

I am late to this argument, but data center imagegen is typically batched so that many images are made in parallel. And (from the providers that aren’t idiots), the models likely use more sparsity or “tricks” to reduce compute.

Task energy per image is waaay less than a desktop GPU. We probably burnt more energy in this thread than in an image, or a few.

And this is getting exponentially more efficient with time, in spite of what morons like Sam Altman preach.

There’s about a billion reasons image slop is awful, but the “energy use” one is way overblown.

Valve’s margins are almost criminal as-is. They make cash hand over fist, and they’re basically a monopoly already.

…So not much might actually change, at least at first?

But this why they don’t need to go public. They have plenty of cash. And it’d just shrink the slice of profit Valve’s owners get.

Still, even private, I’m afraid of what Steam might look like if big competitors like EGS and Microsoft footgun themselves out of the market, and GoG fades away due to their DRM-free policy.

Oh, sweet summer child.

You just wait.

It’s especially bizzare when git repos for “open source corporate alternative” type software uses it over their own repo’s issue trackers and forums.

WTF

They want the social media engagement, I guess. But still. It’s ridiculous how much of a blind spot folks have for it.

It also makes giving or getting any kind of support a hellish time sink, but that’s almost besides the point…

I mean… I think it’s worrying that the norm is “who cares if it’s AI? It’ll fool plenty of people, spread it!” If that’s what you mean.

Social media is going to kill us all well before AI does.

I don’t want to leap into your throat, but have you tried a clean install of a different distro on a USB? And I mean clean; no reusing your home partition, no weird configs until you test out-of-the-box settings.

One thing I’ve come to realize is that I have tons of cruft, workarounds, and configurations in my system that, to be blunt, screw up Nvidia + Wayland. And my install isn’t even that old.

Hunting them all down would take so long that I mind as well clean install CachyOS.

I haven’t bitten the bullet yet (as I just run Linux off my AMD IGP, which frees up CUDA VRAM anyway), but it’s feeling more urgent by the day.

It’s not so much about English as it is about writing patterns. Like others said, it has a “stilted college essay prompt” feel because that’s what instruct-finetuned LLMs are trained to do.

Another quirk of LLMs is that they overuse specific phrases, which stems from technical issues (training on their output, training on other LLM’s output, training on human SEO junk, artifacts of whole-word tokenization, inheriting style from its own previous output as it writes the prompt, just to start).

“Slop” is an overused term, but this is precisely what people in the LLM tinkerer/self hosting community mean by it. It’s also what the “temperature” setting you may see in some UIs is supposed to combat, though that crude an ineffective if you ask me.

Anyway, if you stare at these LLMs long enough, you learn to see a lot of individual model’s signatures. Some of it is… hard to convey in words. But “Embodies” “landmark achievement” and such just set off alarm bells in my head, specifically for ChatGPT/Claude. If you ask an LLM to write a story, “shivers down the spine” is another phrase so common its a meme, as are specific names they tend to choose for characters.

If you ask an LLM to write in your native language, you’d run into similar issues, though the translation should soften them some. Hence when I use Chinese open weights models, I get them to “think” in Chinese and answer in English, and get a MUCH better result.

All this is quantifiable, by the way. Check out EQBench’s slop profiles for individual models:

https://eqbench.com/creative_writing_longform.html

https://eqbench.com/creative_writing.html

And it’s best guess at inbreeding “family trees” for models:

Did y’all read the email?

embodies the elegance of simplicity - proving that

another landmark achievement

showcase your philosophy of powerful, minimal design

That is one sloppy email. Man, Claude has gotten worse at writing.

I’m not sure Rob even realizes this, but the email is from some kind of automated agent: https://agentvillage.org/

So it’s not even an actual thank you from a human, I think. It’s random spam.

To be fair, relying on OneDrive was like waiting for a bomb to go off. This was bound to happen once they got everyone stuck using it.

“let’s save today’s photos from my phone onto my machine” bullshit?

Once it’s set up, I don’t see how that’s a hassle? Especially if its just syncing over WiFi.

Come to CachyOS!

It’s like everything I’ve ever wanted from an OS served on a silver platter.

It’s probably less for OEMs, right? Most people don’t install their own OS, much less pay full price for a license.

And yeah, consumer Windows could disappear and MS wouldn’t care, as long as office computers are still stuck with it. Which they are.

Yeah, see, that’s an “old Windows” kind of problem! “Oh, the printer works fine as long as I don’t leave it plugged in too long.” Users learn to deal with that.

Outlook + the start menu changing overnight, and ceasing to work, and the younger tech scratching their head because they haven’t gotten the update? Or a “wait, it’s screenshotting my stuff?” That’s different. It’s completely out of the user’s control.

The difference is more dramatic now. Linux isn’t so finicky, largely thanks to hard work but also to Windows’ feature stagnation, meaning Linux isn’t a “few years behind” like before. Imagine the situation if, say, Microsoft hadn’t completely screwed up UWP.

Meanwhile, users subconsciously ignored a lot of junk with 95, Vista, 8, whatever. But that’s much harder to ignore in 11.

If Windows crosses a the threshold of “major OEMs start shipping Linux,” what happens to windowscentral?

Do they split the staff/site to a linuxcentral? Winecentral?

That’s kind of a blessing in disguise; otherwise basically all web traffic would be Chrome.

Apparenty this is softening some: https://www.techspot.com/news/108965-japan-gives-apple-december-deadline-drop-ios-browser.html

And Safari is quite performant on iOS.

Maybe I’m too cynical, but I wouldn’t mind if that continues, just so there’s some chunk of traffic that isn’t Chrome and that web development doesn’t turn into a complete monoculture. A smidge of Firefox and Safari alone isn’t enough for that.

(EDIT: My assumption is that if Apple allows Chrome on iOS, you can bet they are going to funnel basically everyone into it).

There are some good iOS browsers.

At the moment, I use Orion (from Kagi) and Narrow32. Quiche Browser is good, DuckDuckGo is fine.

Discoverability on iOS is awful though. The store is just packed with SEO spam and corporate slop on top of all the passion projects or “benevolent” ones.