LG SmartTV "loading" animation (kind of)

[FellippeHeitor]

This was written just for fun and the inspiration for it is as stated in the title of this thread. It was supposed to be something similar to Newton's cradle, but it was easier to allow the circles to coexist in the z-axis so that I only needed to check collision between the currently moving circle and the next in the array (without having to calculate where they would rest and whatnot). The circles have an intentional gravitational tendency to remain at 90 degrees, even after having been hit.

This would likely have been much more efficient if Terry's library had been used, but I'm admittedly still to learn it.

Code: QB64: [Select]

1. CONST false = 0, true = NOT false
2.  
3. SCREEN _NEWIMAGE(200, 200, 32)
4.  
5. TYPE ball
6.     x AS SINGLE
7.     y AS SINGLE
8.     angle AS SINGLE
9.     color AS _UNSIGNED LONG
10.     moving AS _BYTE
11. END TYPE
12.  
13. DIM ball(1 TO 4) AS ball
14.  
15. DATA 78,194,155,61
16. DATA 90,194,89,83
17. DATA 102,89,188,150
18. DATA 138,83,128,188
19.  
20. FOR i = 1 TO 4
21.     READ ball(i).angle
22.     READ r, g, b
23.     ball(i).color = _RGB32(r, g, b)
24. NEXT
25. ball(4).moving = true
26.  
27. x = _WIDTH / 2
28. y = _WIDTH / 2
29.  
30. DO
31.     CLS
32.     CircleFill x, y, 60, _RGB32(216, 216, 216)
33.     FOR i = 1 TO UBOUND(ball)
34.         nextBall = i + 1
35.         IF nextBall > UBOUND(ball) THEN nextBall = 1
36.  
37.         IF ball(i).moving THEN
38.             ball(i).angle = (ball(i).angle + 6) MOD 360
39.             ball(i).x = x + COS(_D2R(ball(i).angle)) * 50
40.             ball(i).y = y + SIN(_D2R(ball(i).angle)) * 50
41.             IF dist(ball(i).x, ball(i).y, ball(nextBall).x, ball(nextBall).y) <= 20 THEN
42.                 ball(nextBall).moving = true
43.                 ball(i).moving = false
44.             END IF
45.         ELSE
46.             IF ball(i).angle < 90 THEN ball(i).angle = ball(i).angle + .5
47.             ball(i).x = x + COS(_D2R(ball(i).angle)) * 50
48.             ball(i).y = y + SIN(_D2R(ball(i).angle)) * 50
49.         END IF
50.     NEXT
51.  
52.     FOR i = 1 TO UBOUND(ball)
53.         CircleFill ball(i).x, ball(i).y, 10, ball(i).color
54.     NEXT
55.  
56.     _DISPLAY
57.     _LIMIT 60
58. LOOP
59.  
60. FUNCTION dist! (x1!, y1!, x2!, y2!)
61.     dist! = _HYPOT((x2! - x1!), (y2! - y1!))
62. END FUNCTION
63.  
64. SUB CircleFill (x AS LONG, y AS LONG, R AS LONG, C AS _UNSIGNED LONG)
65.     DIM x0 AS SINGLE, y0 AS SINGLE
66.     DIM e AS SINGLE
67.  
68.     x0 = R
69.     y0 = 0
70.     e = -R
71.     DO WHILE y0 < x0
72.         IF e <= 0 THEN
73.             y0 = y0 + 1
74.             LINE (x - x0, y + y0)-(x + x0, y + y0), C, BF
75.             LINE (x - x0, y - y0)-(x + x0, y - y0), C, BF
76.             e = e + 2 * y0
77.         ELSE
78.             LINE (x - y0, y - x0)-(x + y0, y - x0), C, BF
79.             LINE (x - y0, y + x0)-(x + y0, y + x0), C, BF
80.             x0 = x0 - 1
81.             e = e - 2 * x0
82.         END IF
83.     LOOP
84.     LINE (x - R, y)-(x + R, y), C, BF
85. END SUB
86.  

[TempodiBasic]

I find it cool!
I'm ready to watch cartoons.

[FellippeHeitor]

Thanks, Tempodi!

[bplus]

Here is the first vision I had for a mod of this:

Code: QB64: [Select]

1. _TITLE "Dang balls again! by bplus 2018-10-03     press spacebar for different color setting..."
2. 'QB64 X 64 version 1.2 20180228/86  from git b301f92
3. CONST WW = 800
4. CONST WH = 600
5. SCREEN _NEWIMAGE(WW, WH, 32)
6. _SCREENMOVE (1280 - WW) / 2 + 30, (760 - WH) / 2
7. RANDOMIZE TIMER
8. CONST PI = 3.141492653589793
9. CONST PI2 = 1.570796326794897
10. CONST bigr = 100
11. CONST x0 = 400
12. CONST y0 = 300
13. TYPE ballType
14.     x AS SINGLE
15.     y AS SINGLE
16.     c AS _UNSIGNED LONG
17. END TYPE
18. DIM SHARED cN, pR, pG, pB
19. DIM p5(4) AS ballType
20. DIM p4(3) AS ballType
21.  
22. angle = 2 * PI / 20 ' 20 degrees
23. smallr = bigr * SIN(.5 * angle)
24.  
25. resetPlasma
26.  
27. FOR i = 4 TO 0 STEP -1
28.     p5(i).x = x0 + bigr * COS((i + 3) * angle)
29.     p5(i).y = y0 + bigr * SIN((i + 3) * angle)
30.     p5(i).c = changePlasma~&
31.     'fcirc p5(i).x, p5(i).y, smallr
32. NEXT
33. bc~& = p5(1).c
34. cN = cN - 4
35. FOR i = 3 TO 0 STEP -1
36.     p4(i).x = x0 + bigr * COS((i + 3.5) * angle)
37.     p4(i).y = y0 + bigr * SIN((i + 3.5) * angle)
38.     p4(i).c = changePlasma~&
39.     'fcirc p4(i).x, p4(i).y, smallr
40. NEXT
41. d = .05
42. WHILE _KEYDOWN(27) = 0
43.     k$ = INKEY$
44.     IF k$ = " " THEN resetPlasma
45.     FOR a = (4 + 3) * angle TO 2 * PI + 3 * angle STEP PI / 45
46.         CLS
47.         COLOR _RGB32(200, 200, 200)
48.         PRINT "R,G,B: ";
49.         PRINT USING "#.####  "; pR, pG, pB
50.         FOR i = 0 TO 3
51.             fcirc p4(i).x, p4(i).y, smallr, p4(i).c
52.         NEXT
53.         fcirc x0 + bigr * COS(a), y0 + bigr * SIN(a), smallr, bc~&
54.         _DISPLAY
55.         _LIMIT 120
56.     NEXT
57.     bc~& = p5(4).c
58.     FOR i = 4 TO 0 STEP -1
59.         IF i <> 0 THEN p5(i).c = p5(i - 1).c ELSE p5(i).c = changePlasma~&
60.         fcirc p5(i).x, p5(i).y, smallr, p5(i).c
61.     NEXT
62.     FOR i = 3 TO 0 STEP -1
63.         p4(i).c = p5(i + 1).c
64.     NEXT
65.     _DISPLAY
66.     _LIMIT 30
67. WEND
68.  
69. FUNCTION changePlasma~& ()
70.     cN = cN + 1
71.     changePlasma~& = _RGB32(127 + 127 * SIN(pR * cN), 127 + 127 * SIN(pG * cN), 127 + 127 * SIN(pB * cN))
72. END FUNCTION
73.  
74. SUB resetPlasma ()
75.     pR = RND * RND: pG = RND * RND: pB = RND * RND
76. END SUB
77.  
78.  
79. 'Steve McNeil's  copied from his forum   note: Radius is too common a name
80. SUB fcirc (CX AS LONG, CY AS LONG, R AS LONG, K AS _UNSIGNED LONG)
81.     DIM subRadius AS LONG, RadiusError AS LONG
82.     DIM X AS LONG, Y AS LONG
83.  
84.     subRadius = ABS(R)
85.     RadiusError = -subRadius
86.     X = subRadius
87.     Y = 0
88.  
89.     IF subRadius = 0 THEN PSET (CX, CY), K: EXIT SUB
90.  
91.     ' Draw the middle span here so we don't draw it twice in the main loop,
92.     ' which would be a problem with blending turned on.
93.     LINE (CX - X, CY)-(CX + X, CY), K, BF
94.  
95.     WHILE X > Y
96.         RadiusError = RadiusError + Y * 2 + 1
97.         IF RadiusError >= 0 THEN
98.             IF X <> Y + 1 THEN
99.                 LINE (CX - Y, CY - X)-(CX + Y, CY - X), K, BF
100.                 LINE (CX - Y, CY + X)-(CX + Y, CY + X), K, BF
101.             END IF
102.             X = X - 1
103.             RadiusError = RadiusError - X * 2
104.         END IF
105.         Y = Y + 1
106.         LINE (CX - X, CY - Y)-(CX + X, CY - Y), K, BF
107.         LINE (CX - X, CY + Y)-(CX + X, CY + Y), K, BF
108.     WEND
109. END SUB

 

But it's just a cheap trick not REAL Ball collisions like Fellippe's and this:

Code: QB64: [Select]

1. _TITLE "Delta angle by bplus 2018-10-05"
2.  
3. ' collisions on the topological rim of circle
4.  
5. 'given 2 ball angles A and B there are 2 distances the Major and the Minor
6. ' make a function that returns the Minor, so don't have to calculate the x, y coordinates of each ball
7.  
8.  
9. 'QB64 X 64 version 1.2 20180228/86  from git b301f92
10. CONST WW = 800
11. CONST WH = 600
12. SCREEN _NEWIMAGE(WW, WH, 32)
13. _SCREENMOVE (1280 - WW) / 2 + 30, (760 - WH) / 2
14. RANDOMIZE TIMER
15. CONST PI = 3.141492653589793
16. CONST PI2 = 1.570796326794897
17. CONST BIGR = 100
18. CONST x0 = 400
19. CONST y0 = 300
20. TYPE ballType
21.     a AS SINGLE '           all balls are located by angle a
22.     da AS SINGLE '          all balls move according to delta angle
23.     c AS _UNSIGNED LONG '   all balls are colored
24. END TYPE
25.  
26. nBalls = 7
27. DIM s(1 TO nBalls) AS ballType 'system array
28. nDiv = 24 'divide circle n times
29. angle = 2 * PI / nDiv ' 20 degrees, EDIT: no now it's 15 degrees! 360 / 24 = 15
30. smallr = BIGR * SIN(.5 * angle)
31.  
32. 'initialize system, setup like Fellippe's only make it a 2D system
33. FOR i = 1 TO nBalls
34.     s(i).a = PI / 2 - (nBalls * PI / 20) - (nBalls) * .5 * angle + i * angle + i * PI / 20 'the last + to separate balls
35.     IF i = nBalls THEN s(i).da = PI / 30 ELSE s(i).da = 0
36.     s(i).c = _RGBA32((nBalls - i + 2) * 20, (nBalls - i + 2) * 20, (nBalls - i + 2) * 20, 100) 'use transpaency to check overlap
37. NEXT
38. l = 30
39. WHILE _KEYDOWN(27) = 0
40.     CLS
41.     FOR i = 1 TO nBalls
42.         'collision check  just check the ball after ball i
43.         IF i = nBalls THEN j = 1 ELSE j = i + 1
44.         IF MinorDistance!(s(i).a, s(j).a) < angle THEN 'collision
45.             SWAP s(i).da, s(j).da
46.             SWAP s(i).c, s(j).c
47.         END IF
48.     NEXT
49.  
50.     'update
51.     FOR i = 1 TO nBalls
52.         s(i).a = s(i).a + s(i).da
53.         fcirc x0 + BIGR * COS(s(i).a), y0 + BIGR * SIN(s(i).a), smallr, s(i).c
54.     NEXT
55.     _DISPLAY
56.     l = l + .1
57.     IF l > 800 THEN l = 800
58.     _LIMIT l
59. WEND
60.  
61. ' all angles are in radian units
62. FUNCTION MinorDistance! (angleA, angleB)
63.     ab = angleA - angleB
64.     IF ab < 0 THEN ab = ab + 2 * PI
65.     ba = angleB - angleA
66.     IF ba < 0 THEN ba = ba + 2 * PI
67.     IF ab < ba THEN MinorDistance! = ab ELSE MinorDistance! = ba
68. END FUNCTION
69.  
70. 'Steve McNeil's  copied from his forum   note: Radius is too common a name
71. SUB fcirc (CX AS LONG, CY AS LONG, R AS LONG, K AS _UNSIGNED LONG)
72.     DIM subRadius AS LONG, RadiusError AS LONG
73.     DIM X AS LONG, Y AS LONG
74.  
75.     subRadius = ABS(R)
76.     RadiusError = -subRadius
77.     X = subRadius
78.     Y = 0
79.  
80.     IF subRadius = 0 THEN PSET (CX, CY), K: EXIT SUB
81.  
82.     ' Draw the middle span here so we don't draw it twice in the main loop,
83.     ' which would be a problem with blending turned on.
84.     LINE (CX - X, CY)-(CX + X, CY), K, BF
85.  
86.     WHILE X > Y
87.         RadiusError = RadiusError + Y * 2 + 1
88.         IF RadiusError >= 0 THEN
89.             IF X <> Y + 1 THEN
90.                 LINE (CX - Y, CY - X)-(CX + Y, CY - X), K, BF
91.                 LINE (CX - Y, CY + X)-(CX + Y, CY + X), K, BF
92.             END IF
93.             X = X - 1
94.             RadiusError = RadiusError - X * 2
95.         END IF
96.         Y = Y + 1
97.         LINE (CX - X, CY - Y)-(CX + X, CY - Y), K, BF
98.         LINE (CX - X, CY + Y)-(CX + X, CY + Y), K, BF
99.     WEND
100. END SUB
101.  

 

Append: Handling collisions when balls are next to or on top of one another is very tricky! Once again my brain is busted trying to handle this situation even on this 1D topology!

[FellippeHeitor]

Interesting mods, bplus.

[bplus]

Append: Handling collisions when balls are next to or on top of one another is very tricky! Once again my brain is busted trying to handle this situation even on this 1D topology!

Hi Fellippe,

It was interesting and once again I am baffled by collision handling, not detection, deciding what happens after detection, specially when balls are next to each other. I almost got to one version of Windows "wait" animation.

[TerryRitchie]

I'm curious about the CircleFill subroutine in your code. Is that a faster method than using PAINT?

Where on Earth did you come up with that? It's boggling my mind.

[FellippeHeitor]

The circleFill routine in my code is a courtesy of v: https://www.qb64.org/forum/index.php?action=profile;u=17

The fcirc routine in bplus's code was written by Steve.

It's not even a matter of being fast (didn't test it); it's a matter of being more efficient in that no paint ever leaks and you can also draw semitransparent filled circles with these.

[bplus]

I'm curious about the CircleFill subroutine in your code. Is that a faster method than using PAINT?

Where on Earth did you come up with that? It's boggling my mind.

Yes, mine came from Steve's example. I have tested it against anything else I found or invented. It was also discussed here:
https://www.qb64.org/forum/index.php?topic=298.45

Vince offered a similar routine that turns out to be basically the same when roundness fixed.

It is based on radical symmetry of circle, do calculations on 1/8th of circle and you have all you need to finish the circle AND no overlap or gap when using alpha colors.

Paint is slower and has to be used with cautions.

I wonder if this routine could be added to QB64.exe?

[SMcNeill]

I'm curious about the CircleFill subroutine in your code. Is that a faster method than using PAINT?

Where on Earth did you come up with that? It's boggling my mind.

Yes, mine came from Steve's example. I have tested it against anything else I found or invented. It was also discussed here:
https://www.qb64.org/forum/index.php?topic=298.45

I enjoyed that topic.  Pondering over Petr's code got me to rethink how I've calculated values for _MEM routines in the past, and has helped me find a better way to tweak a lot of my own code to optimize its speed/performance.  With just a little work, we took his routine, optimized it a bit, and made it run 10x faster!

..A trick I wish I could pull off for all my project So!  :D

[TerryRitchie]

I just read through that whole topic and it was fascinating to see this evolve. I'll need to remember the BF trick with making lines faster.