• 0 Posts
  • 60 Comments
Joined 3 years ago
cake
Cake day: June 13th, 2023

help-circle




  • I understand being intimidated. Under normal circumstances I’d say maybe try building a cheap computer for your first time, but there kinda aren’t any right now. That said, my first one was a reasonably high-end machine for the time, and it turned out fine.

    I feel like the biggest sticking point is actually #2, not #1. I was the kind of kid who was obsessively reading tech magazines/websites/newspaper ad flyers and knew exactly what I wanted and where to get the best deal on it. So yeah, I definitely recommend doing your research first.

    But good news: you’ve got it easy these days with YouTube showing you what to get and how to assemble it instead of having to read. The parts themselves are easier these days, too: no jumpers to set, and almost everything is keyed to only fit one way. Short of being really inept/careless and bending CPU pins or snapping the edge connector off a PCB or something, you’re not going to hurt anything.

    The bottom line is, if you want to do it eventually you might as well do it now, because there really isn’t that much to it.


  • The only real reasons to get a Steam Machine are if:

    1. You really want your computer to be a 6" cube
    2. You really want HDMI CEC support
    3. You really want to give your money to Valve instead of some other company

    Otherwise, something else will be cheaper and/or faster, at the cost of being a more normal desktop size and shape.

    I haven’t had a prebuilt desktop since I was 15, so I would definitely build my own. (In other words, if the thing stopping you is being nervous about screwing it up, don’t be because it isn’t actually hard.)

    Otherwise, I don’t have any specific prebuilt brand recommendations for you, but I’ll echo the advice to get something with a Radeon 9060 XT (or better). Alternatively, if you think you might want to do AI things with it too instead of just gaming, consider a small-form-factor PC with a fast APU and a lot of unified memory (e.g. like the Framework Desktop or Minisforum MS-S1 Max) or a Radeon RX 7900 XTX (with 24GB RAM), but be aware that those would be a lot more expensive right now.






  • I’m just saying it takes a while to get far enough from an event for historians to be able to look at it in retrospect, consider all its context and consequences, and come to a reasonable consensus about which parts were important and what to write about them.

    For example I saw this video earlier about the levee failures in New Orleans during Katrina, which was 21 years ago. It made the good point that the repercussions of the event (in terms of engineering and otherwise) are still unfolding today: the historical record of what happened then is incomplete without the context that’s happening now.

    (The youtuber was also hawking his new book on the history of engineering failures, but never mind that part.)

    Or for a more prosaic example, that’s why my history textbooks in school in the '90s stopped around 1970 or so. They weren’t 20 years old; they just didn’t know what to say about the '80s yet. (And if they had been 20 years old, they probably would have barely talked about anything after WWII.)