作业5
作业描述
参考维基百科，写了一个视角持续拉近的谢尔宾斯基地毯。
效果展示
上图的最初版本里，最小的那一层方块在视角拉近的时候显得很不自然，然后我尝试了一些奇怪的办法，都没有什么好效果，最后试了下加了一个渐变效果，效果如下图，感觉自然多了，而且很有助眠效果哈哈。
小感想
程序动画真有趣，之后准备再写一个能生成类似STG游戏里的弹幕的taichi程序。
代码
# reference ==>
# https://zh.wikipedia.org/wiki/%E8%B0%A2%E5%B0%94%E5%AE%BE%E6%96%AF%E5%9F%BA%E5%9C%B0%E6%AF%AF
import taichi as ti
import handy_shader_functions as hsf
ti.init(arch=ti.cuda)
deepth = 5
res_x = 3**deepth
res_y = 3**deepth
pixels_1 = ti.Vector.field(3, ti.f32, shape=(res_x, res_y))
pixels_2 = ti.Vector.field(3, ti.f32, shape=(res_x, res_y))
flags_1 = ti.field(ti.i32, shape=(res_x, res_y))
flags_2 = ti.field(ti.i32, shape=(res_x, res_y))
@ti.func
def fill(x, y, W, H, deepth):
c = 0
for i in range(deepth):
if x < 1 or W < 1 or H < 1:
break
x_ = (3 * x) // W
y_ = (3 * y) // H
if x_ == 1 and y_ == 1:
c = 1 + i
break
x -= (x_ * W) // 3
y -= (y_ * H) // 3
W = W // 3
H = H // 3
return c
@ti.func
def zoom_in(t: ti.f32):
scale = ti.pow(1 / 3, (t % 120) / 120)
for i, j in pixels_2:
flags_2[i, j] = flags_1[i, j]
if scale < 1:
i_ = ti.floor(scale * i)
j_ = ti.floor(scale * j)
flags_2[i, j] = flags_1[i_, j_]
@ti.kernel
def render(t: ti.f32):
# draw something on your canvas
white = ti.Vector([1.0, 1.0, 1.0])
black = ti.Vector([0.0, 0.0, 0.0])
for i, j in flags_1:
flags_1[i, j] = fill(i, j, res_x, res_y, deepth)
pixels_1[i, j] = white
pixels_2[i, j] = white
zoom_in(t)
for i, j in flags_2:
if flags_2[i, j] > 0:
pixels_2[i, j] = black
if flags_2[i, j] == deepth:
gray = 2 - ti.pow(2, (t % 120) / 120)
pixels_2[i, j] = ti.Vector([gray, gray, gray])
gui = ti.GUI("Canvas", res=(res_x, res_y))
if __name__ == '__main__':
for i in range(120):
t = i * 1.0
render(t)
gui.set_image(pixels_2)
filename = f'./results/frame_{i:05d}.png' # create filename with suffix png
print(f'Frame {i} is recorded in {filename}')
gui.show(filename) # export and show in GUI
