• 0 Posts
  • 6 Comments
Joined 2 years ago
cake
Cake day: June 18th, 2023

help-circle

  • You can use sync in terminal. But it’s tricky because it sometimes returns even when the writing isn’t finished.
    My method is to use sync multiple time, if it returns immediately 2 times it should be clear,
    Only then do i dismount the stick, because I don’t like to dismount a device with pending operations. But when the dismount says the stick is ready to be removed, you should be clear.




  • File transfer progress bars generally aren’t worth shit, I have no idea how this is still a problem. It dates back to the early 90’s, and it’s still not done properly.
    Almost everything to do with file transfer report it to be finished before it actually is, because it doesn’t consider write back cache. So it’s only the reading part that is finished, not the writing part. Meaning the transfer isn’t actually finished. Never never never trust it is my motto. Personally I’d rather not have the dysfunctional progress bar, if I could have an actually accurate completion notification.

    PS: I just switched back to KDE/Plasma after many years. Overall I’m pretty pleased, but also a bit puzzled about a lot of the simplifications.
    like inability to disable caps lock, and the inability to change double click speed for the mouse. So now I need a startup command for the caps lock, and I needed to edit an ini file for the mouse for such simple things, that used to be accessible through settings???
    But they finally added the ability to use numpad for hotkeys, which I found out, and was what made me switch back to KDE. Now I only miss the ability to use the scroll wheel on hotkeys too. To finally have similar functionality like I had 10-15 years ago with Compiz. 😜


  • A completely redesigned PC based on free and open standards and low latency technologies throughout to maximize real time capabilities.
    As it is, the busses are too slow to enable for instance a good PC based oscilloscope. And there is not an open bus like for instance on Raspberry Pi.
    It would need to abandon X86, maybe a modern Alpha could do it. Alpha is basically a better API than X86, and they were way faster, until they were cancelled to promote Itanium instead. Risc V would be another option.

    Such a project however, would probably have little chance of succeeding no matter how good. Because software would be needed for success, and to attract software it needs to be successful.

    But it should be a system that invites creativity through openness and capabilities. A bit like the Amiga allowed by being open in the 80’s, but not owned by any single company. Amiga did all sorts of stuff no other computers were capable of, because it was open and better at real time tasks than anything Intel had.

    The price estimate for the project including designing new CPU and Custom chips, would be about $10-50 billion. And probably closer to the 50!!!