They had that “hey, you’re on adblocker, you don’t get to use this unless you disable it” screen. I don’t know if Google stopped using it or if my adblocker just got good at preventing it, but I was getting it for a while, then it stopped.
Definitely issues. For a while I was getting the ban screen until they figured things out. Now I am clearly getting worse performance than I do with ads. I also noticed a tendency to get a frozen image when stepping over an ad break after using Firefox for a long time, but closing it down and opening it again seems to clear that up, so that’s what I do.
But it still works and I still get no ads once the daily key situation got cleared up. At least for now. Not seamless, though.
I guess that depends on the use case and how frequently both machines are running simultaneously. Like I said, that reasoning makes a lot of sense if you have a bunch of users coming and going, but the OP is saying it’s two instances at most, so… I don’t know if the math makes virtualization more efficient. It’d pobably be more efficient by the dollar, if the server is constantly rendering something in the background and you’re only sapping whatever performance you need to run games when you’re playing.
But the physical space thing is debatable, I think. This sounds like a chonker of a setup either way, and nothing is keeping you from stacking or rack-mounting two PCs, either. Plus if that’s the concern you can go with very space-efficient alternatives, including gaming laptops. I’ve done that before for that reason.
I suppose it’s why PC building as a hobbyist is fun, there are a lot of balance points and you can tweak a lot of knobs to balance many different things between power/price/performance/power consumption/whatever else.
OK, yeah, that makes sense. And it IS pretty unique, to have a multi-GPU system available at home but just idling when not at work. I think I’d still try to build a standalone second machine for that second user, though. You can then focus on making the big boy accessible from wherever you want to use it for gaming, which seems like a much more manageable, much less finicky challenge. That second computer would probably end up being relatively inexpensive to match the average use case for half of the big server thing. Definitely much less of a hassle. I’ve even had a gaming laptop serve that kind of purpose just because I needed a portable workstation with a GPU anyway, so it could double as a desktop replacement for gaming with someone else at home, but of course that depends on your needs.
And in that scenario you could also just run all that LLM/SD stuff in the background and make it accessible across your network, I think that’s pretty trivial whether it’s inside a VM or running directly on the same environment as everything else as a background process. Trivial compared to a fully virtualized gaming computer sharing a pool of GPUs, anyway.
Feel free to tell us where you land, it certainly seems like a fun, quirky setup etiher way.
Yeah, but if you’re this deep into the self hosting rabbit hole what circumstances lead to having an extra GPU laying around without an extra everything else, even if it’s relartively underpowered? You’ll probably be able to upgrade it later by recycling whatever is in your nice PC next time you upgrade something.
At this point most of my household is running some frankenstein of phased out parts just to justify my main build. It’s a bit of a problem, actually.
OK, but why?
Well, for fun and as a cool hobby project, I get that. That is enough to justify it, like any other crazy hobbyist project. Don’t let me stop you.
But in the spirit of practicality and speaking hypothetically: Why set it up that way?
For self-hosting why not build a few standalone machines and run off that instead? The reason to do this large scale is optimizing resources so you can assign a smaller pool of hardware to users as they need it, right? For a home set of two or three users you’d probably notice the fluctuations in performance caused by sharing the resources on the gaming VMs and it would cost you the same or more than building a couple reasonable gaming systems and a home server/NAS for the rest. Way less, I bet, if you’re smart about upgrades and hand-me-downs.
Yeah, on that I’m gonna say it’s unnecessary. I don’t know what “integration with the desktop” gets you that you can’t get from having a web app open or a separate window open. If you need some multimodal goodness you can just take a screenshot and paste it in.
I’d be more concerned about model performance and having a well integrated multimodal assistant that can do image generation, image analysis and text all at once. We have individual models but nothing like that that is open and free, that I know of.
Look, the naming ship has sailed and sunk somewhere in the middle of the ocean. I think it’s time to accept that “AI” just means “generative model” and what we would have called “AI” is now more narrowly “AGI”.
People call videogame enemies “AI”, too, and it’s not the end of the world, it’s just imprecise.
That is a stretch. If you try to download and host a local model, which is fairly easy to do these days, the text input and output may be semi-random, but you definitely have control over how to plug it into any other software.
I, for one, think that fuzzy, imprecise outputs have lots of valid uses. I don’t use LLMs to search for factual data, but they’re great to remind you of names of things you know but have forgotten, or provide verifiable context to things you have heard but don’t fully understand. That type of stuff.
I think the AI shills have done a great disservice by presenting this stuff as a search killer or a human replacement for tasks, which it is not, but there’s a difference between not being the next Google and being useless. So no, Apple and MS, I don’t want it monitoring everything I do at all times and becoming my primary interface… but I don’t mind a little search window where I can go “hey, what was that movie from the 50s about the two old ladies that were serial killers? Was that Cary Grant or Jimmy Stewart?”.
Seems to go to the OPs point of having open software alternatives, though. I am in a fairly unusual place regarding the practical usage of some of these things, but I do agree that if the entire concept is fundamentally rejected among proponents of open software that delays the possibility of developing viable alternatives to work around those issues.
Yeah, for sure. If you just drop Ubuntu or Fedora or whatever on a machine where everything works for you out of the box the experience is not hard to wrap your head around. Even if one thing needs you to write something in a terminal following a tutorial, that’s also frequent in Windows troubleshooting.
The problem is that all those conversations about concurrent standards for desktop environments, display protocols, software distribution methods and whatnot are hard to grasp across the board. If and when you hit an issue that requires wrapping your head around those that’s where the familiarity with Winddows’ messy-but-straightforward approach becomes relevant.
In my experience it’s not going through the motions while everything works or using the system itself, it’s the first time you try to go off the guardrails or you encounter a technical issue. At that point is when the hidden complexity becomes noticeable again. Not because the commands are text, but because the underlying concepts are complex and have deep interdependencies that don’t map well to other systems and are full of caveats and little differences depending on what combination of desktop and distro you’re trying to use.
That’s the speed bump. It really, really isn’t the terminal.
Well, the good news is that of course you can use Linux with only as much command line interaction as you get in Windows.
The bad news is that the command line REALLY isn’t what’s keeping people away from Linux.
Hell, in that whole list, the most discouraging thing for a new user isn’t the actually fairly simple and straightforward terminal commands, it’s this:
Here’s where it gets a little trickier: Scrolling on Firefox is rough, cause the preinstalled old version doesn’t have Wayland support enabled. So you either have to enable Wayland support or install the Flatpak version of Firefox.
This is a completely inscrutable sentence. It is a ridiculous notion, it brings up so many questions and answers none. It relates to concepts that have no direct equivalent in other platforms and even a new user that successfully follows this post and gets everything working would come out the other end without understanding why they had to do what they did or what the alternative was.
I’ve been saying it for literal decades.
It’s not the terminal, it’s not the UX not looking like Windows.
Hey. Hey? Hey. Hello friend.
You just got good advice. Remember that the world is not only for you.
Like, it’s super not for you. It’s mostly not for you at all. If you ask me whether I care about things for people with disabilities or things for you, you don’t even chart. That’s only two options and you’re not even second on that list.
So yeah. Good advice.
Local and secure image recognition is fairly trivial in terms of power consumption, but hey, there’s likely going to be some option to turn it off, just like hardware acceleration for video and image rendering, which uses the same GPU in similar ways. The power consumption argument is not invalid, but the way people deploy it is baffling to me, and is often based on worst-case estimates that are not realistic by design.
To be clear, Apple is building CPUs that can parse these queries in seconds into iPads now, running at a few tens of watts. Each time I boot up Tekken on my 1000W gaming PC for five minutes I’m burning up more power than my share of AI queries for weeks, if not months.
On the second point I absolutely disagree. There is no practical advantage to making accessibility annoying to implement. Accessibility should be structural, mandatory and automatic, not a nice thing people do for you. Eff that.
As for the third part, every alt text I’ve seen deployed is not adding much of value beyond a description of the content. What is measurable and factual is that the coverage of alt-text, even in places where it’s disproportionately popular like Mastodon, is spotty at best and residual at worst. There is no question that automated alt-text is better than no alt-text, and most content has no alt-text.
That is only the tip of the iceberg for ML applied to accessibility, too. You could do active queries, you could have users be able to ask for additional context or clarification, you could have much smoother, automated voice reading of text, including visual description on demand… This tech is powerful in many areas, and this is clearly one. In fact, this is a much better application than search, by a lot. It’s frustrating that search and factual queries, where this stuff is pretty bad at being reliable, are the thing everybody is thinking about.
Same but for tab groups. I can’t believe it took this long and every extension-based alternative is busted in some fundamental way.
AI generated alt-text running locally is actually a fantastic accessibility feature. It’s reliable, it provides a service, it can absolutely be deployed securely.
It’s fine to be critical of technology, it’s not fine to become as irrational about it as the tech bros trying to make a buck.
You guys know that MS rolls out updates through a system of multi-layered opt-in channels and deploy to the wider channels in chunks over longer periods of time, right?
They may not be the most competent or scrupulous megacorp, but they do hire actual software engineers and stuff.
Quest Link yes, I was referring to the alternative Steam Link app that is available on Quest. That’s maintained by Valve (and honestly works better than the wireless version of Quest Link, IMO). I was wondering if that works as an alternative, but I wouldn’t be surprised if there are still dependencies for controller inputs and head tracking that need Oculus software installed to work on the server side.
It’s roughly consistent with Steam survey data, and given the current context, Linux Steam users are heavily incentivized to contribute to the survey. The numbers are what they are.
Plus the guy’s argument is that relevant data sources separate ChromeOS out because it’s substantially different, which is a fair point, regardless of the accuracy of the data.
It’s the PC version with controller support patched into the old Sidewinder support. It works super smoothly, although these days you can get the PS1 version to look a lot better with emulation tweaks.
Still, go get it. It’s so cool to have this DRM free. The installer is safely in my library now.