Posted By
Luca
on 2009-11-24
03:33:21
 |  Circles in Multi Botticelli
I always hated the way Multi Botticelli has to draw circles: why non-symmetric, eh? 
A patched version is needed!
|
|
Posted By
MMS
on 2009-11-24
14:33:10
 | Re: Circles in Multi Botticelli
Hahaha, I spent hundreds of hours with that program, but I have never noticed that bug  It is really evident.
In fact I always drew my circles with hand. Do not ask why, I like pixel magic and hard work Anyhow, my style of grafix need no circles or boxes, just a lot of pixels and shades next to each other.
|
|
|
|
Posted By
Chicken
on 2009-11-24
18:37:58
 | Re: Circles in Multi Botticelli
I remembered this topic being mentioned before as well :D
The circles look like "drawn on hi-res screen looked at in multic color mode".
|
|
Posted By
SVS
on 2009-11-25
02:32:11
 | Re: Circles in Multi Botticelli
The problem is known. If I'm not wrong, in the standard source routine (from which it derives) there is a parameter named "aspect" (see GW-BASIC for example) that manages that feature.
It is clear that the planners of Botticelli did not fix it.
|
|
Posted By
Luca
on 2009-11-25
07:22:26
| Re: Circles in Multi Botticelli
No can't believe: 24 of November, the exact day. Autumn effects? 
|
|
Posted By
Chicken
on 2009-11-26
06:07:58
| Some details about circles in Multi Botticelli
SVS...
I think Luca is not refering to the aspect issue (which is caused by different pixel apect ratios).
Draw a circle in Multi Botticelli and you will see what he's talking about: The left or the right side (depending ONLY on the circles' "radius" not on the x/y position*) will look flattened and less curvy. I'm not sure but to me this looks like a rounding/truncation issue.
The BASIC circle command does NOT have this "flattening" issue.
*Since the radius in Multi Botticelli is set by moving the cursor upwards or downwards after setting the center, I think it's better to talk about pixel steps when explaining the exact behavior.
An even number of pixel steps away from the center will result in a flattened right side, an odd number in a flattened left side. BUT (too make it even more complicated) only if the radius was set by moving the cursor UPWARDS.
If you move the cursor downwards it's the other way around: Even number of pixel steps -> flattened left side. Odd number off pixel steps -> flattened right side.
There's another difference that should be considered when setting the radius:
If you move the cursor upwards, circle's height is 2*(pixel steps)+1.
If you move the cursor downwards, circle's height is 2*(pixel steps)-1
This means you'll get the very same circle (same side flattened) if you you move the cursor x pixel steps upwards or x+1 pixel steps downwards.
Furthermore, as you might have concluded already, ALL circles have an odd height, which makes it impossible to draw a circle vertically aligned exactly within 4x8 areas 
|
|
