antialias
Posted on Tue, 2010-08-31 10:18
Ive set antialias to both false and true and it does nothing. The vector lines are never antialiased.
application =
{
content =
{
width = 320,
height = 480,
scale = "none",
antialias = false
},
}
Replies
Posted on Sun, 2011-11-27 12:23
#2
Did this ever get fixed? antialias = false, seems to do nothing when I object.xScale, object.yScale
Posted on Sun, 2011-11-27 18:48
#3
Provide an example please.
Antialias is working in .591 - switch between true/false using this code and you'll notice a difference.
1 2 3 4 5 6 | local star = display.newLine( 0,-110, 27,-35 ) star:append( 105,-35, 43,16, 65,90, 0,45, -65,90, -43,15, -105,-35, -27,-35, 0,-110 ) star:setColor( 255, 102, 102, 255 ) star.width = 3 star.x = 160 star.y = 160 |
Peach :)
Posted on Sun, 2011-11-27 19:04
#4
ug. What am I doing wrong here then in my main.ui ?
Posted on Sun, 2011-11-27 20:15
#5
To get your code looked at/debugged you'd want to use this link; http://www.anscamobile.com/corona/support/ (Where you're posting now is the bug forum.)
Alternatively make a thread in Developer Support to ask the community to check it out.
Peach :)
Posted on Fri, 2011-12-02 01:35
#6
Hey Lava / Peach
Looking at the .zip and the discussion you do not need the initialising setting, that smooths the edges of vector lines and objects.
What you are after is a way of being able to set the smoothing of a texture. At the moment as the images is scaled opengl applies a routine to scale the texture which ends in it becoming blurred at scales over 100%.
What you need is nearest neighbour smoothing which then makes everything look pixelated as it is scaled.
Unfortunately Corona does not support this. I have been asking for this for about a year now along with other graphical techniques but so far nothing has been added.
Posted on Fri, 2011-12-02 01:43
#7
Yeah. I know. Don from Spriteloq filed me in on this. It's a real shame, and it seems like an easy thing for them to implement. I have had to create some work arounds. :(
I don't see it doing anything myself so I logged it as bug #920 (internal).
-Tom