• 0 Posts
  • 7 Comments
Joined 3 years ago
cake
Cake day: June 12th, 2023

help-circle

  • The exact quote is:

    And this was a debug session from hell, enormously helped by an AI doing much of the grunt-work.

    I’d like to call it my tireless helper, but the AI several times stated flat out that this was impossible and unsolvable and that we should just write a report about it.

    I suspect those things have been trained by people who may not be quite as stubborn as I am.

    But while the AI was ready to give up several times, it did keep adding debug code and analyzing it faithfully when I pushed. So credit where credit is due and I let the AI write the commit message above.

    This is basically a one-liner fixing a bogus “round_up()” to a “round_down()”, but there were 24 patches adding more and more debug information to this, and 18 kernel boot to finally narrow it down to this. - Linus

    That’s a long way short of finding the bug and pointing it out, and instead is just the tedious work of adding checks and then deciding where to add more checks when you have the result of the previous checks. I suspect if we’d spent a trillion dollars on a tool for that, we could have come up with something much more efficient than an LLM.



  • No, don’t use CLAMP, use CLAMP_TO_EDGE. CLAMP is a dumb mode for niche applications that all switched to using shaders around the turn of the millennium. No GPUs have direct hardware support for CLAMP and so they have to emulate it instead. You are throwing performance away for no reason.

    (also, technically, the image is using CLAMP_TO_EDGE as CLAMP is a 50% mix of the border colour and the edge pixel colour, so it would be greyer if the border colour is grey like the first image suggests.)



  • In the UK, it’s because it’s the defence’s job to notice this kind of thing, but paying for a good lawyer is life-ruiningly expensive, and public defenders via legal aid are paid hourly, but have a cap on the number of hours that’s never enough to deal with anything properly, so are either rushing everything and miss things, or are massively overworked and earning well below minimum wage when their unpaid work is included.

    I don’t know if the same problems extend to Canada, but it’s fairly likely, as the adversarial approach just inherently means the defence has to do loads of things that would be someone else’s job or not need doing at all under alternative systems, so you have to have a trained professional doing lots of work, and that’s inherently expensive, but taxpayers don’t like spending money on defending people who probably did crimes.