Pretty pattern

[+Vorticon]

So I came across this video by 8-bit Show and Tell on Youtube and I thought it would be interesting to recreate that on the TI, so I did  I could not get the pattern to work initially, until I figured out that the line length had to be Odd, not Even, for reasons that are unclear to me. That said, I'm not terribly happy with my code, and I was wondering if there is a way to optimize it somehow...

 

 

 

 

10 CALL CLEAR :: RANDOMIZE
20 CALL SCREEN(10)
30 CALL COLOR(1,7,10,2,9,10)
40 CALL CHAR(33,"FF7F3F1F0F07030180C0E0F0F8FCFEFF")
50 CALL CHAR(40,"FFFEFCF8F0E0C0800103070F1F3F7FFF")
60 B=B=0
70 FOR I=1 TO 27 :: B=B=0
80 Z=INT(RND+0.5)
90 IF Z=0 THEN CH=33 ELSE CH=40
100 IF B=-1 THEN IF Z=0 THEN CH=34 ELSE CH=41
110 IF I=27 THEN PRINT CHR$(CH)ELSE PRINT CHR$(CH);
120 NEXT I
130 GOTO 70

 

[JJB]

I was intrigued too and had a go with green instead of red 🙂:

 

100 CALL CLEAR
110 CALL SCREEN(3)
120 CALL COLOR(9,4,13)
130 CALL CHAR(96,"FFFEFCF8F0E0C080")
140 CALL CHAR(97,"0103070F1F3F7FFF")
160 CALL CHAR(98,"80C0E0F0F8FCFEFF")
165 CALL CHAR(99,"FF7F3F1F0F070301")
170 RV=INT(RND*4)
180 PRINT CHR$(96+RV);
190 GOTO 170

 

Definitely needs work

 

Edited February 21, 2023 by JJB
clatification

[+OLD CS1]

Give this a go.

 

10 CALL CLEAR :: RANDOMIZE :: CALL SCREEN(10):: CALL COLOR(1,7,10,2,9,10)
20 CALL CHAR(33,"FF7F3F1F0F07030180C0E0F0F8FCFEFF",40,"FFFEFCF8F0E0C0800103070F1F3F7FFF")
30 FOR I=1 TO 27 :: B=B=0 :: PRINT CHR$(34+INT(RND+.5)*7+B);:: NEXT I :: PRINT :: GOTO 30

 

[+OLD CS1]

Also, this is from my back-library of schtuff.

 

5 RANDOMIZE
10 PRINT CHR$(47+INT(RND*2)*45);:: GOTO 10

 

Surprisingly, a little slower than its TI BASIC equivalent.

[+OLD CS1]

15 minutes ago, OLD CS1 said:

Give this a go.

 

10 CALL CLEAR :: RANDOMIZE :: CALL SCREEN(10):: CALL COLOR(1,7,10,2,9,10)
20 CALL CHAR(33,"FF7F3F1F0F07030180C0E0F0F8FCFEFF",40,"FFFEFCF8F0E0C0800103070F1F3F7FFF")
30 FOR I=1 TO 27 :: B=B=0 :: PRINT CHR$(34+INT(RND+.5)*7+B);:: NEXT I :: PRINT :: GOTO 30

 

Oh, and to reverse the effect of B, change line 30:

 

30 FOR I=1 TO 27 :: B=B=0 :: PRINT CHR$(33+INT(RND+.5)*7-B);:: NEXT I :: PRINT :: GOTO 30

 

Lastly, you can move B=B=0 to between the PRINT and NEXT statements to have it start at 0 rather than -1.

[JJB]

16 minutes ago, OLD CS1 said:

Give this a go.

 

10 CALL CLEAR :: RANDOMIZE :: CALL SCREEN(10):: CALL COLOR(1,7,10,2,9,10)
20 CALL CHAR(33,"FF7F3F1F0F07030180C0E0F0F8FCFEFF",40,"FFFEFCF8F0E0C0800103070F1F3F7FFF")
30 FOR I=1 TO 27 :: B=B=0 :: PRINT CHR$(34+INT(RND+.5)*7+B);:: NEXT I :: PRINT :: GOTO 30

 

Yes that looks great. Shame the TI doesn't have several shades of gray

[JJB]

Cool that we both had the same idea of replacing the C64 inversion capability with a char definition.

[oddemann]

Will compiling it make it faster?

[Retrospect]

6 minutes ago, oddemann said:

Will compiling it make it faster?

Yes.  Many times faster.  And if there's no decimal points in the numbers, it will compile quite well.

[Retrospect]

40 minutes ago, JJB said:

PRINT CHR$(34+INT(RND+.5)*7+B)

@oddemann for example, this particular line won't go down too well with the compiler, it would have to be amended somehow.

[+OLD CS1]

1 hour ago, Retrospect said:

@oddemann for example, this particular line won't go down too well with the compiler, it would have to be amended somehow.

I had used RND*2 before, but I am not certain how RND returns.  @senior_falcon has some details on the best way to use RND when compiled.

[Retrospect]

Just now, OLD CS1 said:

I had used RND*2 before, but I am not certain how RND returns.  @senior_falcon has some details on the best way to use RND when compiled.

I must say those patterns are very pretty and would make a good "transition" effect between levels on an arcade game.  Not that I want to derail the thread or anything but still, they're nice.

[+TheBF]

2 hours ago, OLD CS1 said:

Also, this is from my back-library of schtuff.

 

5 RANDOMIZE
10 PRINT CHR$(47+INT(RND*2)*45);:: GOTO 10

 

Surprisingly, a little slower than its TI BASIC equivalent.

XB random number generator sucks from what Rich has seen. Could be that. 

[+Vorticon]

13 hours ago, JJB said:

Cool that we both had the same idea of replacing the C64 inversion capability with a char definition.

Yeah that was the fastest way to simulate inversion. One could write a routine to do that, but it would have been way too slow.

[+Vorticon]

14 hours ago, OLD CS1 said:

Give this a go.

 

10 CALL CLEAR :: RANDOMIZE :: CALL SCREEN(10):: CALL COLOR(1,7,10,2,9,10)
20 CALL CHAR(33,"FF7F3F1F0F07030180C0E0F0F8FCFEFF",40,"FFFEFCF8F0E0C0800103070F1F3F7FFF")
30 FOR I=1 TO 27 :: B=B=0 :: PRINT CHR$(34+INT(RND+.5)*7+B);:: NEXT I :: PRINT :: GOTO 30

 

That's what I'm talking about! 

[senior_falcon]

B=B=0   I don't understand the purpose of this? It sets B=0 and I see no place where B is ever changed

 

PRINT CHR$(34+INT(RND+.5)*7+B)

RND+.5 will be a value from .5 to 1.4999.....

so INT(RND+.5) will be 0 or 1 with an equal chance of either one.

so INT(RND+.5)*7 will be either 0 or 7

 

You could replace that with PRINT CHR$(34+INT(RND*2)*7); and since B=0 I left it out.

 

And with those changes I think this could be compiled.

Edited February 22, 2023 by senior_falcon

[+Lee Stewart]

27 minutes ago, senior_falcon said:

B=B=0   I don't understand the purpose of this? It sets B=0 and I see no place where B is ever changed

 

That is not what happens. B is set to -1 (TRUE) if B=0 going into the expression or 0 (FALSE), otherwise. The first '=' is an assignment operator and the second '=' is a relational operator. The expression is treated the same as though it were stated

B=(B=0)

...lee

[senior_falcon]

27 minutes ago, Lee Stewart said:

 

That is not what happens. B is set to -1 (TRUE) if B=0 going into the expression or 0 (FALSE), otherwise. The first '=' is an assignment operator and the second '=' is a relational operator. The expression is treated the same as though it were stated

B=(B=0)

...lee

You are right! I thought that is what would happen and tested this in the immediate mode:

B=B=0

PRINT B

and it prints 0

 

It must be that B was set to some value other than zero, because if you enter

B=0

B=B=0

PRINT B

then -1 is printed. 

 

And of course, the program 10 B=B=0::PRINT B 

will print -1 as you say. 

[+Lee Stewart]

7 minutes ago, senior_falcon said:

You are right! I thought that is what would happen and tested this in the immediate mode:

B=B=0

PRINT B

and it prints 0

 

It must be that B was set to some value other than zero, because if you enter

B=0

B=B=0

PRINT B

then -1 is printed. 

 

And of course, the program 10 B=B=0::PRINT B 

will print -1 as you say. 

 

Yeah, in Walid’s program, B will alternate between 0 and -1 as the loop containing it executes.

 

...lee

[senior_falcon]

You may need the parentheses for this to work compiled.

[RXB]

52 minutes ago, Lee Stewart said:

 

Yeah, in Walid’s program, B will alternate between 0 and -1 as the loop containing it executes.

 

...lee

I changed the program to see what was going on in B and it was just -1 0 over and over and over?

10 CALL CLEAR :: RANDOMIZE :: CALL SCREEN(10):: CALL COLOR(1,7,10,2,9,10)
20 CALL CHAR(33,"FF7F3F1F0F07030180C0E0F0F8FCFEFF",40,"FFFEFCF8F0E0C0800103070F1F3F7FFF")
30 FOR I=1 TO 27 :: B=B=0 :: S$=S$&STR$(B):: PRINT CHR$(34+INT(RND+.5)*7+B);:: NEXT I :: PRINT :: GOTO 30

Yea I know it crashes with string full but just FCTN 4 and type PRINT S$ to see result.

[senior_falcon]

20 hours ago, Vorticon said:

I could not get the pattern to work initially, until I figured out that the line length had to be Odd, not Even, for reasons that are unclear to me.

 

30 FOR I=1 TO 27 :: B=B=0 :: S$=S$&STR$(B):: PRINT CHR$(34+INT(RND+.5)*7+B);:: NEXT I :: PRINT :: GOTO 30

i'm guessing the B=(B=0) is getting out of sync when the line is even. I bet if you added B=(B=0) after NEXT I it would work properly with an even line length. (and improperly with an odd line length)

 

[+OLD CS1]

1 hour ago, RXB said:

I changed the program to see what was going on in B and it was just -1 0 over and over and over?

The value of B represents the orientation of the piece to the drawn.

[+OLD CS1]

55 minutes ago, senior_falcon said:

i'm guessing the B=(B=0) is getting out of sync when the line is even. I bet if you added B=(B=0) after NEXT I it would work properly with an even line length. (and improperly with an odd line length)

The line length is controlled by I, hence 27 characters and thus odd.

[senior_falcon]

Yes, and if you made it 28 and added the one extra B=(B=0) it would probably work.