site banner
ArjinFerman  Tinfoil Gigachad  2d ago (text post)   347 thread views

Tinker Tuesday for May 27, 2025

This thread is for anyone working on personal projects to share their progress, and hold themselves somewhat accountable to a group of peers.

Post your project, your progress from last week, and what you hope to accomplish this week.

If you want to be pinged with a reminder asking about your project, let me know, and I'll harass you each week until you cancel the service

2
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Homies: Ride or Die the 3d game.

Hacking away at stuff. Make is missing .hpp file changes, probably something about generating the .d files properly via clang.

Aside from that I added proper bounds checking to the racetrack by constructing a bunch of quadrilaterals to paper over it and simply iterating them and asking "is car inside quad?" and rejecting forward vector if it is not found in any. Can change this to a quadtree once I'm confident this is the right approach.

But weirdly the starting point where I spawn the car (700, 0, 0) doesn't fall inside of any quad even though I'm creating the quads procedurally and they should be sharing neighboring line segments. Wondering if this is some floating point precision issue creating some subtle gaps.

EDIT: I pasted the 1500 quadrilateral coordinates into ChatGPT and said why doesn't 700, 0, 0 lie inside of any of these and it responded that it's probably because it sits exactly on the edge of one line and my bounds checking is not inclusive of the actual boundary line itself. E.g. that sits exactly on the edge of one of the quads and my comparison function may be excluding that. Plausible!

EDIT2: yeah, that was it. I had this

bool Quad::is_point_inside_xz(const bx::Vec3& pos) const {
  const auto sign = [](const bx::Vec3& a, const bx::Vec3& b,
                       const bx::Vec3& c) {
    return (b.x - a.x) * (c.z - a.z) - (b.z - a.z) * (c.x - a.x);
  };

  bool b1 = sign(pos, p0, p1) >= 0;
  bool b2 = sign(pos, p1, p2) >= 0;
  bool b3 = sign(pos, p2, p3) >= 0;
  bool b4 = sign(pos, p3, p0) >= 0;

  bool result = (b1 == b2) && (b2 == b3) && (b3 == b4);
  return result;
}

and ChatGPT o3 told me to try this. Basically using an epsilon value to check for float equality than exact because of precision, and also being more tolerant to point ordering, I'm not sure I'm definitely doing it clockwise in all cases.

This works!

bool Quad::is_point_inside_xz(const bx::Vec3& pos) const {
  // 2-D cross product (signed area *2) in the X-Z plane
  auto cross = [](const bx::Vec3& a, const bx::Vec3& b,
                  const bx::Vec3& c) -> float {
    return (b.x - a.x) * (c.z - a.z) - (b.z - a.z) * (c.x - a.x);
  };

  constexpr float EPS = 1e-6f; // tolerance for “on the edge”

  // Determine polygon orientation (+1 = CCW, −1 = CW)
  // Any non-degenerate triple works; p0-p1-p2 is simplest.
  float orient_sign = (cross(p0, p1, p2) >= 0.0f) ? 1.0f : -1.0f;

  auto same_side = [orient_sign](float v) // interior ⇒ v*orient ≥ 0
  {
    return v * orient_sign >= -EPS; // accept 0 within EPS
  };

  // Point must be on the “same side” of every edge.
  // Use edges p0→p1, p1→p2, p2→p3, p3→p0.
  return same_side(cross(p0, p1, pos)) && same_side(cross(p1, p2, pos)) &&
         same_side(cross(p2, p3, pos)) && same_side(cross(p3, p0, pos));
}

What am I even doing here. Feels like I'm one of those people playing chess still even though computers totally crush humans at it.

I mean, not really though. I had Claude Code write a lot of boilerplate for me w.r.t. get to triangle but it mostly produced buggy shader code and I had to sit down and think pretty hard a lot of the time. Even the model loading code was buggy and model files are pretty standard.

LLMs can collect answers from StackOverflow and sometimes get the right one. How nice.

Meanwhile, I'm switching off CoPilot because it keeps suggesting that I use classes and methods that plainly don't exist.

And people claim LLMs aren't useful for coding..

They're useful, except when they're not. It would be amazing if they could actually say "I don't know" instead of hallucinating 200 lines of code that are wrong that had no chance of being right.

I've had Claude code waste an hour of my time before and bill me $20 in tokens.

Was it 3.7 Sonnet? That model was far too gung-ho for its own good. I've heard better things about the 4 series.

It was. I haven't tried 4 yet. Kind of anxious it might waste my time and send me a bill again. I'll report back if I get proper fucked and try it out of desperation.

Post repo, if gpt's solution doesn't work. Might take a stab at it when I have a moment.

Edit: oops, didn't see your edits when posting.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats