• 0 Posts
  • 36 Comments
Joined 1 year ago
cake
Cake day: August 30th, 2025

help-circle

  • That’s tight fit. If you really need the two capture cards then that’s the room you’ve got to work with. If you can bring it down to just one card at the top that would be a lot better. I personally use a capture card that’s connected using USB3, which works very well and doesn’t require room inside the computer (even though I would have room).

    Most videocards have two or three fans on the bottom these days, with such a tight fit one of the fans can easily collide with the capture card. And even when it doesn’t it will have a hard time sucking in air from there. You can try and find a blower type card, but that would be a bit harder to find.

    And you wouldn’t happen to have a GPU inside the CPU? Because the iGPUs from AMD are very solid if you don’t need much in terms of performance (easily faster than the 1030 you mentioned). Otherwise two capture cards and a GPU is a big ask in such a small case.


  • The A series is pretty limited, I would go for a B series if you can find it. But keep in mind drivers for Intel on Windows are terrible and on Linux even worse. So that would not be a good time.

    There’s plenty of cards that would fit in 40mm, but your airflow would be pretty bad. Do you have a pic of the inside of your system? There has to be a way to make something more fit.

    But most 6600, 7600 and 9060 cards should be able to fit in 40mm. And with AMD the drivers on Linux are very good.







  • Reminds me of a ray tracing obsession I had around 1990. I created ray tracing “games” on every weird system I had in any programming language I knew at the time. I even had one on my old 1984 MSX machine, that ran at 5 fps, which wasn’t bad at all. The one I made in C on my DOS machine was pretty good actually. Except I never got very far with the actual game part, as soon as it rendered and you could move around I got bored and moved on to doing the same thing on the next weird machine or language.

    These were inspired by a lot of demos going around at the time and papers being passed around. I think Wolfenstein itself was inspired by the same demos.






  • Yeah those refurb drives from eBay were the last good source. I got a bunch of them last year, 2 of them had issues but were replaced under warranty by the manufacturer. All of those seem to be either gone or not priced very well.

    Doing anything PC related these days is very rough with prices being sky high. And even if you are willing to pay, there isn’t a lot of good stuff to get. It sucks ass.


  • You are the only one who can answer that. Things will change, but if you qualify that as a loss that’s up to you.

    In my experience with Linux, where there is a will there is a way. If you want something to run badly enough it will simply work.

    Personally I still play Dark Souls and Dark Souls 3 a lot, including mods, randomizer etc. I’ve even used cheat engine on those, for example to spawn items at the start of the game to do challenge runs. That’s old software and primarily designed for Windows, it runs just fine on Linux tho.


  • Yeah a friend of mine got a new job last year and they finish off every sprint by creating a new release and pushing it to prod on Friday afternoon. She was shocked when she found out about that.

    The company doesn’t have a definition of done for the devs and doesn’t include QA in the release cycle. So the devs just push broken unfinished shit into releases. Then when QA points out all of the issues, the devs go crying to management about how they can’t keep to the schedule for the new sprint if QA makes them work “endlessly” on tasks that were “finished” long ago. Basically blaming QA for the issues the devs themselves caused. Management and sales over promise to customers, so they want the devs to work on the new stuff, as they promised it was already in place when development wasn’t even planned let alone started. QA and support need to deal with the fallout, trying to handle customers that got non-working broken shit and are pissed about it. Management just tells support to forward any “troubled” customers to them, as they will “handle” it. But what management does is just over promise again, till the customer is happy, setting up an endless loop of failure and disappointment.



  • Usually the previous registrar will hold on to the domain for 6-18 months in case the customer wants the domain back. After that it’s automatically sold in bulk auctions. The companies, sometimes called domain squatters, buying those domains in bulk hold on to them often for years asking a premium price. The idea is once someone thought that domain name was valuable, it might be valuable again. They buy 1000 domains and maybe sell a dozen or so, but as operating costs are extremely low it’s still somehow worth it. It’s all highly automated at this point, so just a fact of life. I’ve seen domains being held for over 10 years.

    See https://en.wikipedia.org/wiki/Cybersquatting


  • I don’t think you have any idea how hard EUV actually is. ASML was told for decades it could never be done and they were throwing money away by trying to make it happen. Even inside the company a lot of folk were against the whole thing, stating it was not possible. If it weren’t for the leadership having stuck by it during development, it would never have been done.

    It took advances in physics, math and engineering to even create the technology, let alone make it reliable, fast and cheap enough to make it usable for mass production purposes. It’s a huge advancement and has a good few years in optimizations and improvements ahead.

    What’s next after EUV? I don’t think anyone really knows, this might be the end of the line as far as shrinking node sizes goes. And we’ll need to look into novel structures and materials. Or who knows, ASML might have something else cooking in a top secret project.