I really want a mod that deletes Silverhand and his whole plotline. I've played it, gotten all of the endings, and now I want to never see or hear him ever again while I explore entirely different stories in this world. I'd love for expansions where you just play as someone else who isn't V. It's true that V is connected to basically all of the interesting stories, so I'm asking for a lot of work, or a really empty night city with only the ambient missions left. But that sounds fine to me.

Cheaper than expected given the economic hellscape, still more than the Index. Oh well. Glad they are attempting to prevent scalping, sad that it means that there's a random draw to be able to buy them. Still trying to buy one. I figure the 256GB model will be entirely fine, especially with the SD card expansion slot. Pity that you can't upgrade the storage like the Deck, but oh well.

Is it illegal to stop in a bike lane in London? A quick search suggests it is in fact legal. That sucks, they should change that. It's illegal to drive a car in a protected bike lane in CA, and Waymo follows the law. When it's not a protected bike like, they put their stupid huge SUV into the lane and sit for as long as legally allowable and it sucks.

Spiciness isn't related to taste buds. Related; you don't have taste buds on your eyes or up your butt, but I'll bet you can feel pepper oil there! So don't worry, you can still have plenty of ability to taste, and still have high heat tolerance.

Portal 2 has moving surfaces, with the cop-out that it's accelerating surfaces that cause the portal to break.

::: spoiler spoiler No matter that you literally put a portal on a different planet... :::

I played hundreds and hundreds of hours of Tremulous back in the day, and at least a few of its successor Unvanquished -- before mostly falling out of playing online games.

This is exactly why I am so sad that I didn't buy more DDR4 back when it was reasonable. I run Unsloth's Qwen 3.5 122B A10B UD_Q4_k_XL and while it works great, I really wish I had enough ram for Q6 or even Q8. The speed difference won't be wildly worse, but the quality of output is noticeable. I'm just glad that it works as well as it does in Q4. It's mostly limited by my main ram bandwidth, the GPU helps but I only barely hit 15t/s decode with MTP hitting >80%.

I'm not the person who you're responding to, but at least in my case I am just not comfortable with having the ability to switch, since I would like to connect devices which could be seriously damaged if the switch toggled for any reason. I have modified a couple of switchable power monitoring plugs already, but I would rather pay slightly less and not have a hassle.

Interesting idea, but if you didn't bother to write this, why would you expect others to read it? Rather than slop out an essay, just describe the part that matters. I'm interested in seeing evals, until then it's a cool theory.

I run a 1600X as a Proxmox host, I have used TLP to set all of the configurable that are exposed, and it idles at 70W. There are a few other things consuming power though - a spinning HDD, a discrete GPU (quadro K2200) doing zero work, and four DIMMs.

This is helpful to heat my apartment in the winter, but regrettable overall.

RIP RAD, but good news for cell harvesters. I'll see how many of these recalled batteries I can pick up for pennies.

I still have my PSP on my desk, on who knows how many battery replacements. I come back for:

• Burnout Legends
• Locoroco
• Lumines & Lumines II
• N+
• Echochrome
• Guilty Gear XX

This has got to be an LLM. The post history is just more of the same. But I don't get it, why would anyone use an LLM for lemmy responses?

Ryzen 5950X with a slight underclock
2x32+2x16 DDR4 2666
Radeon 7900XTX 24GB with a 250W power limit
15tps is entirely usable as an agent, but I can't go above 131k ctx and must set --parallel 1 to fit in available RAM+VRAM. It actually still OOMs periodically, as even with nothing else running it only has a couple GB to work with. If that happens with context mostly filled then you're looking at ten minutes of prompt processing before next token.

I'm not about to buy another 64GB pair to replace the 16s, but... I wish I had done that years ago.

Wendell and Serve The Home in general have totally lost the "home" focus. Sure, I love reading about cutting edge 800GbE transceivers... but they hardly ever cover actually practical home stuff anymore. Their network switch reviews are the only thing vaguely applicable left.

Yeah, I'd love a recumbent trike style commuter ride, but there's no way I'm spending used-car-level-money on one. I assume they're expensive because they're niche... but that's not going to change if they cost so much.

I've had HA hang while running wyoming-openwakeword, but my setup is pretty different from yours -- I was trying to get wakeword detection off the audio track in an RTSP stream. I didn't solve it, and gave up.

I sat in one well over a decade ago, and it was pretty cool! I'm sad it never really got beyond cool prototyping.