Depends how familiar with disk partitioning you already are, but it’s generally not that bad. Shrink your windows partition, create a new ntfs partition in windows, and install Linux. Pretty much every major installer has an option to preserve Windows.
I guess this is the thing with Linux, at various levels it skews DIY and more time intense. And honestly the market seems to be bifurcating between that and the opposite extreme: Android/iOS, which many use as their main PCs now.
Consider this: why do you have 2TB of games installed?
IDK what your gaming habits are, but I find myself bouncing between a few at most, 1 big AAA at most, and found keeping the rest… to be blunt, data hoarding.
I suppose the only exception is modded setups, it even then I always redo them anyway since the base games, modding frameworks and mods get updated over time.
You can redownload what you use, progressively.
This is not always true. I measured that the Proton version of Rimworld and Stellaris, for instance, are massively faster than the Linux one.
What I do is dual boot Windows/Linux and have them share big NTFS partitions for bulk data they can both access.
You could try getting autocad running?
https://github.com/cryinkfly/Autodesk-Fusion-360-for-Linux#fusion-360---linux-wine-version
But TBH, neutering Windows down to a “software tool” to the point I could wipe it and not really care, and spending most of my desktop time on Linux, has been wonderful.
Depends what you do on your computer.
Macs are amazing and unique and, at the same time, extremely infuriating and very expensive.
It is if you’re a billionaire and want to be an even bigger billionaire.
Not that most of America will know :/
It’s ironic how conservative the spending actually is.
Awesome ML papers and ideas come out every week. Low power training/inference optimizations, fundamental changes in the math like bitnet, new attention mechanisms, cool tools to make models more controllable and steerable and grounded. This is all getting funded, right?
No.
Universities and such are seeding and putting out all this research, but the big model trainers holding the purse strings/GPU clusters are not using them. They just keep releasing very similar, mostly bog standard transformers models over and over again, bar a tiny expense for a little experiment here and there. In other words, it’s full corporate: tiny, guaranteed incremental improvements without changing much, and no sharing with each other. It’s hilariously inefficient. And it relies on lies and jawboning from people like Sam Altman.
Deepseek is what happens when a company is smart but resource constrained. An order of magnitude more efficient, and even their architecture was very conservative.
Saw it here, as new as January: https://en.wikipedia.org/wiki/Movements_for_the_annexation_of_Canada_to_the_United_States#Opinion_polling
Seems some newer polls may not follow this trend, thankfully: https://www.ctvnews.ca/canada/article/large-majority-of-canadians-reject-trumps-annexation-overtures-poll-suggests/
As for general American sentiment on Trump, I don’t know if most Americans approve of annexation, but that may not matter if they approve in general: https://www.axios.com/2025/03/16/trump-high-dems-low-new-poll
Plenty of the US military seems to adore Trump whatever he proposes.
Canadian polls are rising too.
Don’t discount the effect of propaganda, and of decision makers in Washington drinking some of their own kool-aid.
Heh, that’s a good future.
The one we’re speeding towards is where they all collude, and all the service is shitty and the same (or worse) as its ever been. If its not just monopolized. That’s why Night City is a relatively good place to live in CP2077: There’s still competition.
Meanwhile, Rust punches you in the face for the mere suggestion. Again. And again.
Python happily nods, runs it one page at a time, very slowly, randomly handing things off to a C person standing to the side, then returns a long poem. You wanted a number.
Assembly does no checking, and reality around you tears from an access violation.
EDIT: Oh, and the CUDA/PyTorch person is holding a vacuum sucking money from your wallet, with a long hose running to Jensen Huang’s kitchen.
Also, welcome to Lemmy!
I had to think about it too, lol. This is an equation for DC/instantaneous power, and if you want to get into AC math, this is more like a square wave. Averaging the power out over time doesn’t necessarily work with the equation, as you figured out, as it doesn’t when you try to measure AC (sinusoidal) power by average voltage or whatever.
Sadly not always true. For example, AT&T’s board chair is also its CEO.
Shoudn’t it be 25%?
Current is not controlled here, resistance (aka the soldering iron) and voltage are.
Power = Voltage ^ 2 / Resistance. Double the voltage, that quadruples the power. So you only want to plug in 25% of the time to get the equivalent power of 120V.
But it might not melt at double power? Maybe the extra heat helps, I can’t find a resistance/temperature curve for a soldering iron…
Source: EE dropout.
So?
I guess I’m not drawing lines between the volume of death threats, any exaggeration of the article, the culprit(s), and what any of that has to do with an LLM. They all seem like independent factors. Sure someone could have used an LLM to spam her, or theoretically the article could be editorialized, but… She still got death threats, maybe a lot.
Not sure what an LLM has to do with this. Anyone can make a death threat? And anyone can use an LLM to do it I guess.
That’s the bottom line, isn’t it? Everyone deserves one’s respect, no one deserves one’s worship.
Even worse, TPS. Days chug along in Linux compared to Windows, along the lines of a 1.5x slowdown.
FPS spikes are much worse, too. I tested both extensively on the same saves, using their native performance tools (the perf commands in stellaris and DPA in Rimworld).