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?

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 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:

I used to write a few things for HardOCP, and we always made fun of Tom’s for clickbait, chum, and straight up groveling to Nvidia. We straight up weren’t allowed to cite them, and had no desire to.

That was like a decade ago.

…Maybe they were alright when Tom was still with them, but that was a loooong time ago. Too long for me to even know.

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.

Even down to the questionable marketing.

To be fair, the 9000 series has the same architecture as Intel’s 6000 series Skylake CPUs, from 2015.

There is no 9500K, but I assume they’re referencing this one: https://www.intel.com/content/www/us/en/products/sku/134895/intel-core-i59500-processor-9m-cache-up-to-4-40-ghz/specifications.html

Which is 6C/6T.

Fair enough, heh.

There are other game streaming services, like: https://shadow.tech/

Just to name one.

Dare I say it, but if you are really that excited, consider GeForce Now? It’s cheap, as long as you drop the subscription like a rock when you’re done.

Clickbait headline:

It will run on 16GB; the original announcement even mentions the Switch 2. There’s no reasoning would need 32GB for 60 FPS, given low settings and a CPU that can push it.

That headline’s a bit disappointing, as I really love Notebookcheck. Their reviews and databases are incredible.

I mean… It functioned as a CPU.

But a Phenom II X6 outperformed it sometimes, single thread and multithreaded. That’s crazy given Pildriver’s two generation jump and huge process/transistor count advantage. Power consumption was awful in any form factor.

Look. I am an AMD simp. I will praise my 7800X3D all day. But there were a whole bunch of internet apologist for Bulldozer back then, so I don’t want to mince words:

It was bad.

Objectively bad, a few software niches aside. Between cheaper Phenoms and the reasonably priced 2500K/4670K, it made zero financial sense 99% of the time.

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.

Perhaps the plan is to drive up demand with war, non-petro power rollbacks, cutting emissions regulation, and datacenters?

Even the 6-core Phenom IIs from 2010 were great value.

But to be fair, Sandy Bridge ended up aging a lot better than those Phenom IIs or Bulldozer/Piledriver.

AMD almost always had the better price/performance

Except anything Bulldozer-derived, heh. Those were more expensive and less performant than the Phenom II CPUs and Llano APUs.