Pol and I will tonight.

We will start opening some of the attack vectors we closed, gradually admnd with more secueity. Bit some will remain closed. Send me an DM or better an email at info@shadertoy.com, and we can coordinate.

Yeah, we only had 1 week of break between AI crawler attacks. This one is even hungrier than the last one.

It's not real, it's digital art (visual effects, like all the digital cities and creatures you'd see in an Avengers film)

This is an improved design, with paddles at 100 degree or so, which improves throughput by almost 4x (can't do 90 degrees because paddles collide), and make a fun wavy pattern:

For those who didn't peep the comments, f(x) is:

∣x∣ˣ

or in code

pow( abs(x), x )

Beautiful, but unfortunately the derivative f'(x) = (1+ln ∣x∣)⋅∣x∣ˣ shoots to -∞ at the origin.

Nope! Even if you adjusted the exponential exp( 3(x-1)/2 ) and the gaussian 1.3*exp(-2(x+0.5)^2), you wouldn't create that level of asymmetry in the bump.

No. But if you try synthetizing it, you'd probably want to replace the sine with a gaussian or something that falls-off rapidly from its center

I'll share the solution later; I was surprised when I found it.

Oh wow, very nice try! But it's even simpler than that.

Nope. But the feature you detected on the right side is a good clue indeed.

No, it's even simpler.

We did no ipads/screens for our kid till age 8. Now it's 1.5 hours a week IF creative/craft based. No internet, "digital literacy" is overrated imho. For social media, so far we have enough critical mass of school parents willing to delay it till post-teenage. But we'll see, fingers crossed.

You can expand the cos directly (1 - x^2/2! + x^4/4! + ...) and apply the said eigen decompoaition. Taylor convereges slowly though, better use some other polynomial.

I think it's 1 :71 (1 black every 72 tiles)?