I haven’t played a skating game since the old Tony Hawk games. Honestly, this is tempting me to buy Skate.

Is anyone here actually playing the mod?

I’m pretty sure delivery made it obsolete?

Curbside delivery is awesome, in many situations.

And they can afford to leave a few checkers in there for the situations that call for in-store shopping. That’s where self checkout would fail anyway.

People discovering their apps watch them.

Amazon is definitely going to plug that hole.

I know.

One can’t just bolt HBM onto something else, no matter how skilled the hacker is. It needs an interposer to connect it because the pin pitch is so small, and it needs a memory controller that supports it.

EDIT: And if you’re talking about AMD using it in gaming GPUs or some other hardware, unfortunately the design pipeline is very long. If AMD decided to make a gaming HBM GPU right this second, and rushed it, it would take years to design it and get it in production.

The trailer is so wholesome.

I will share some impulse thoughts:

• I love the SDV-style NPCs. Personality, their style, their lives; that’s such a draw.

• Love the look of resource gathering. Yes, I will mine some tin and chop some trees. Though I might suggest making these activities “interesting” with physics (like how trees will fall down and tumble in Valheim).

• All the animations are gorgeous.

• Love the variety of monsters. It feels alien.

• I have more mixed feelings about the style of traits. I know many are fans of ‘oldschool’ RPG systems, but I’m personally not.

• I’d prefer more of a ‘psyche’ and skill system like Rimworld or something.

• Combat is down to user taste.

• I’m sure many have told you this, but TBH, I’m unlikely to play a survival game like this if it’s just solo. But I’m not asking for networking; even if it had couch split screen, that’d be great. I could even stream the screen remotely.

Unfortunately, memory production capacity changes vary slowly.

In spite of the drama, memory production hasn’t increased all that much, which is why there is such a RAM crisis. Maybe it will grow some before the bubble pops, but Micron, SK Hynix, and Samsung aren’t stupid:

https://en.sedaily.com/finance/2026/10/01/microns-long-term-chip-deals-jump-63-percent-as-ai-memory

They are getting companies to sign long term (5+ year) pricing contracts. In other words, they clearly know the bubble could pop. They are already hedging their bets, and they aren’t going to put themselves in the position of massive overexpansion.

Datacenter GPUs can’t game :(. As in, it’s literally impossible, as they are missing ROPs.

The last generation that could is basically the Nvidia V100, and the datacenter equivalent of your Radeon VII. AMD removed the ROPs after the Radeon VII generation. The RTX 3000-generation Nvidia A100 can technically run a game if hacked to do so, but it performs extremely poorly because of a lopsided compute config.

Now, if you are interested in GPU compute and self-hosting, maybe we can use them.

But Nvidia learned to buy back and destroy GPUs during the last crypto bust, so I’m afraid they may do it again.

If I had a few billion for a TSMC factory, I would…

This is what I’m referencing. The memory sits on the package with the processor, connected with thousands of microscopic traces, so fine it needs an expensive substrate rather than a typical PCB, and expensive equipment to assemble:

The connectors are tiny, and made with immense precision. There are about a thousand reasons you can’t just transfer HBM once its been used, and besides; other memory controllers wouldnt’ support it.

HBM is only usable with chips co-designed for it, packaged on an interposer with microscopic traces. An enthusiast is not doing anything useful with it unless they own a few TSMC assembly plants.

The CPU modules can be tricky, too.

Now, some enthusiasts are already hacking. Go to the Level 1 Tech forums, and you’ll find a group trying to reverse engineer AMD Instinct GPUs designed for server motherboards, dealing with locked firmware, soldering stuff, you name it.

…But you probably won’t like the end goal of the hacking: using them to run LLMs and other GPGPU stuff locally.

Unfortunately, not really. RAM made as HBM or put in weird data center modules is stuck in that form factor, and when the bubble does pop, there is loads of pent up demand for memory makers to fill with new production.

It’s not like crypto where they used consumer stuff.

That’s true.

I keep reading that Meta employee morale is in the gutter, and think ‘that cannot be sustainable.’ Leaks may be a symptom.

I don’t think any “LLM enthusiast,” in their right mind, would feel okay about:

Meta asks users to let Muse control everything from their bank accounts to their emails.

If you use these things at a lower level, that is horrifying. Especially with how rushed the implementation seems to be, even assuming the VMs don’t leak.

I contest the food part.

The last sandwich I had there was… fine. Squished, plain bun, thin breading, yet outrageously expensive.

A Raising Cain’s sandwich is way better. Even if price wasn’t a factor, and I couldn’t go local for some reason, I’d much, much rather have Steak and Shake, P Terry, Chicken Express, Rosa’s, Fuzzy’s, PDQ…

As the business? Like I said, I’d start with a text classifier, a small model quickly categorizes the input prompt. If it’s something off topic, I’d return some generic refusal reply, or maybe send the score to the LLM itself if it’s somewhat ambiguous.

Chatbot UIs that give very quick refusals are using “filtering” models in just that way. There’s a whole world of language modeling that existed before LLMs, just for that sort of thing.

Now, if I was the LLM provider? I dunno, but I would try to integrate it into the weights. There are some really interesting papers and hacks out there for prefiltering prompts that have nothing to do with system prompts.

It’s wild that a system prompt is still the major guardrail in most chatbots like this. Even if the LLM provider won’t implement anything, the user could add a classifier, some kind of preprocessor…

I think they’re referencing the app/engagement model.

Being pornography is… tangential, sort of. But part of what makes OF and all those other services so addictive is the way they’re structured. Living on your phone, pinging you, targeted personalized offers, FOMO, and engineering the app to form parasocial relationships in particular.

That’s different than randomly watching porn on the Internet, IMO.

One other common factor is the “primal” addictiveness hook, e.g. gambling, food craving, or the draw of porn.

I know scientifically minded people who view it both as ‘hysteria,’ and for what remains, something that geoengineering can solve in the future.

In other words, look up the WSJ article advocating for fossil fuel production maximization as a net economic optimization in the long term, basically ‘better to maximize efficiency and minimize suffering now, to better geoengineer solutions in the future.’

…I read geoengineering papers. They are horrific. As in “our state-of-the-art-solution would… pour acid rain over the ice caps. And it’s still economically unfeasible, but less unfeasible. And maybe it will blunt things some?”

I’m offended such arguments are being presented.

But I can see right where those opinions come from, as with every twist of climate change denialism.

The environmental conspiracists are right. It really is Big Oil and such groups, from what I can see.

Sad, but makes sense. I always wondered how they could fund the dev effort for Linux/Windows.

These comments though, whoa… it seems everyone either loves or hates Orion.

What are they built on outside the US?

IIRC Google researchers have publicly used it for machine learning experiments. For example, for grabbing a sample video as part of some demo code.