11 replies to this topic

[Maverich40]

I have a routine that plots a Mandelbrot Set fractal (famous image of "overlapping circles" on complex plane...)

When drawn on fx 9750 gii, I get columns of blank pixels using Plot X,Y  or other Graphics commands...

What do I need to do to either set up proper Plotting mode, or what Plotting Command should I be using that does

not space out the pixels....

 

(I can not be absolutely certain there might not be a small problem with Program itself, but it's kind of lengthy to be

 typing now.)

 

Thanks for any help!

 

//RadSurfer//

 

[CalcLoverHK]

Hi Maverich40

 

Use other variables other than X and Y.

 

Reference: (Planete-Casio)

 

Cheers

CalcLoverHK

Edited by CalcLoverHK, 28 February 2020 - 07:50 AM.

[Maverich40]

Okay, so I replaced X, Y with G,H which were unused... I still get a mandelbrot plot that has blank columns spacing everything out! WHY?

 

Plot mode should be Connect or Plot? It takes over 20 minutes to draw this complete graphic on 9750gii (try it on a 7000g).

 

Why should simply plotting a Pixel be so non-intuitive and difficult!

 

//RadSurfer//

[CalcLoverHK]

@Maverich40

What is your View-Window setting?

 

Reference: (Planete-Casio)

[Maverich40]

GridOff

AxesOff

G-Connect

...

Viewwindow A,B,9,D,C,9

-2.1 to A, 1.1 to B, -1.1 to D, 9 being Scale

 

No one has addressed differences in Plot, PlotOn, etc, or just why simple addressing of Graphics Screen Pixels is apparently not straightforward...

 

//RadSurfer//

[Lephe]

Each Plot/PlotOn will draw a single pixel on screen (except if you have G-Connect in which case consecutive Plot will render lines). As a side effect, X and Y will be modified.

 

It is hard for us to guess what's going wrong without seeing the relevant looping and drawing. Here's a program that fills the screen at your View-Window setting:

ViewWindow -2,1,0,-1,1,0

For -2→A To 1 Step 3÷128
For -1→B To 1 Step 2÷128
PlotOn A,B
Next:Next

You might have one unlucky blank column if you approximate the coordinates in an irregular way, but that's a simple fix.
 

Drawing pixels is straightforward, but bugs can be hard to find. Hang on!

[Maverich40]

Here is a paste-bin which fully lists the Program, where it was originally discovered, and what changes were made

based on suggestion from this forum:

 

http://dpaste.com/2MHM0Z2

 

If you can get this to plot properly, PLEASE share. 

 

//RadSurfer//

 

[Lephe]

Have you noticed this line : (B-A) / 94 -> S ? This is clearly meant for a calculator with a screen of 94 pixels. The fx-9750G II has 128 pixels horizontally, and 127 are available to PRGM. You need to divide by 127!

[Maverich40]

How could I have missed that! Thank you... Running it right now... it does appear to be more connected, hopefully I can take a photo of the finished image, provided it does not disappear when completed :-|}

 

Because of this Screen Size difference, I should be looking for possible other refinements?  Like I hope this fills most of the screen...

 

StoPict 1 as last statement,

and

RclPict 1 to be able to recall it, should Calc power down... excellent!

 

That's progress I guess.

 

For those into FreeBASIC, I did create a FB32 program that does this in colour:

 

https://i.imgur.com/O4wTnZe.png | https://i.imgur.com/fnaq4NE.png | https://i.imgur.com/m8EsMGa.png

 

Which is based on a Python3 script of all things... but that was easily converted to FreeBASIC!

 

//RadSurfer//

Edited by Maverich40, 28 February 2020 - 08:12 PM.

[Lephe]

Glad you got it to work! You can improve on this program by using actual complex numbers, which might speed up the computation by a noticeable amount.

[Hlib2]

The size of display in fx-9750gii is 127×63 pxl, therefore X-pitch (B-A)÷126 should be a rational fraction (Edit: decimal fraction without rounding error in the calculator) here too for this plottings.
Make:
-2.15->A:1->B:(B-A)÷126->S
and you will get what you want.

Edited by Hlib2, 28 February 2020 - 09:32 PM.

[Maverich40]

https://i.imgur.com/...m/0tagf6T.jpg  It looks just a tiny bit "squashed" but otherwise just fine!  Using 63, 127 presently, and 40 iterations, might try playing with Iterations to see what I get...

 

Thanks everyone!

 

//RadSurfer//