Відео

Shoutout to the homie Computation, doing the heavy lifting

yes it's all math, generated on GPU in real time using 100% my own code, I added some more info in the description now

That's right

@@metadaat5791 Ah sorry I thought you meant you hooked it up to a MIDI music player, not a controller 😅

@@j-maffe next time I'll add WebAudio to play muffled squeaks when I turn the knobs 🤣

@TheCodingTrain, thanks so much!! I'm a huge fan of your channel! :D You've been super inspirational in my creative coding journey, especially your enthusiasm, "go for it" attitude and I especially learned about not worrying about smaller details getting in the way :-) BTW if you're short on time, feel free to watch my video at 1.5x speed, it's not edited and I don't talk that quickly :-) Very much looking forward to see your video on SDFs :-) It's a broad subject to give a short intro to :-) My approach was to stick to 2D SDFs (so I don't need to also explain a raymarcher) and hopefully provide people with intuition how to recognize they might already be using 2D SDFs (like checking if a point's inside a margin frame == box SDF, or say, circle packing is full of circle SDFs and their union) and then show how they give you super powers :-)

@@metadaat5791 Thanks for the reply, this video is so terrific! I especially love the simplicity of rendering by just drawing lots of random dots over time (rather than filling in every pixel every frame!)

Yep, y is an argument to the function representing the y coordinate of the vector.

Beginners can also come in several flavours :-) I know I would've eaten that up, studied and peeled it apart, when I just got into SDFs. I also want to give that part of my audience some meat to chew on. But it doesn't really matter for the rest, it's just to show that the box SDF as constructed from the max() function is not 100% correct. There's many more tricks you can do with the k/edge function which deserve their own 30 minute video (and I don't think these tricks are in any Iq article, either). I will, however, firmly disagree on one letter variable names adding unnecessary difficulty. This is a math topic with math formulas, and formulas get increasingly harder to read, the longer variable names you use. They quickly can start spanning multiple lines. There's a reason math and physics papers use single letters in their formulas. I find it much harder to spot a mistake in a wide formula with long variable names. Also I think the letters I used are pretty self explanatory: x, y, d for distance, p for point. And it's not like if you're a beginner coder, one letter variable names are any harder to read, in fact many beginner coders use single letter variable names themselves :-) You may know I also enjoy size optimizing, so I could be a _little bit_ biased :-) But I wasn't writing size optimized code here (except the `k` formula, which I copypasted from one of my projects), and was using short variable names honestly to aid with readability. My hope is that people learn to get an intuition for when a piece of their program logic is actually using an SDF, and quickly seeing a formula's "shape" helps with that.

Really? I couldn't stand the constant "and uuuhmm yeah uuuhhmmm".