On Jan 23, 2011, at 9:13 PM, Ricardo Moran wrote:
> Mmmm... maybe it is platform related. I really don't know. I'm on Windows XP.
>
> Pharo version: Pharo1.2rc1 of 18 January 2011 update 12311
> Squeak version: Squeak4.2alpha of 4 January 2011 update 10854
> Vm version: Squeak3.10.2 of ''5 June 2008'' [latest update: #7179]
>
> On Sun, Jan 23, 2011 at 5:00 PM, Udo Schneider <
udo.schneider@homeaddress.de> wrote:
> Hi Richo,
>
> could you tell me which Image/VM/OS you where using?
>
> On my System (Mac OS X 10.6.6) Pharo 1.1 and Squeak 4.1 on Cog VM (r2349) and regular VM (4.2.5beta1U) both show no antialiasing (image attached)
>
> Is this maybe somehow platform/plugin related?
>
> Best Regards,
>
> Udo
>
>
>
> On 23.01.11 20:28, Ricardo Moran wrote:
> Hi Udo,
>
> Out of curiosity, I executed your code in Squeak and the antialiasing
> works there. Debugging a little, I found a difference in
> BalloonCanvas>>#line:to:width:color:. If you update Pharo's
> implementation with Squeak's one it works in Pharo as well.
> There might be other differences so you might look into that.
>
> I hope this helps.
> Richo
>
> On Sun, Jan 23, 2011 at 4:09 PM, Udo Schneider
> <
udo.schneider@homeaddress.de
> <mailto:
udo.schneider@homeaddress.de>> wrote:
>
> � �All,
>
> � �I'm using the following code. IMHO the line is jagged ... but maybe
> � �I'm doing something wrong:
>
> � �form := Form extent: 640@480 depth: 32.
> � �"form fillWhite."
> � �canvas := BalloonCanvas on: form.
> � �canvas aaLevel: 4.
> � �canvas line: 0@0 to: 640@480 color: Color red.
> � �morph := ImageMorph new.
> � �morph image: form.
> � �morph openCenteredInWorld.
> � �"morph abandon."
>
> � �CU.
>
> � �Udo
>
>
>
> � �On 22.01.11 15:37, Udo Schneider wrote:
>
> � � � �All,
>
> � � � �does somebody has a pointer to a graphics lib/framework/engine
> � � � �for Pharo
> � � � �which works with Anti-Aliasing.
>
> � � � �My first tests with Balloon2D were disappointing (maybe I'm doing
> � � � �something wrong). But even setting aaLevel does not seem to
> � � � �bring any
> � � � �result. I.e. the line drawing code falls back to superclass code
> � � � �which
> � � � �simply draws the same jagged lines like always.
>
> � � � �Any pointers?
>
> � � � �CU,
>
> � � � �Udo
>
>
>
>
>
>
>
>
>