I thought there might be a system property for this but I couldn't find one. The "right" solution is probably to add a setFontRenderContext() method to the Platform class so applications can control this value.
On Jul 25, 2011, at 10:19 AM, Pierre Jansen wrote: > Quite possible, though I wouldn't have the faintest idea ;). Can you confirm > that the method suggested by Edvin below would be the best course of action > for a global override? > > On Mon, Jul 25, 2011 at 3:02 PM, Greg Brown <[email protected]> wrote: > OK. As Edvin mentioned, Pivot uses the AA hints provided by the platform by > default. Is it possible that Ubuntu/GNOME recently changed these defaults and > the JDK has not been updated yet? > > On Jul 25, 2011, at 8:55 AM, Pierre Jansen wrote: > >> >> On Mon, Jul 25, 2011 at 2:42 PM, Greg Brown <[email protected]> wrote: >> Sorry if this was mentioned earlier in the thread, but what platform are you >> using? >> >> I'm running Ubuntu 10.04 (GNOME) with the Sun JRE (1.6.0_22) / 32 bit. >> >> On Jul 25, 2011, at 8:22 AM, Pierre Jansen wrote: >> >>> >>> On Sun, Jul 24, 2011 at 11:38 PM, SYSE | Edvin <[email protected]> wrote: >>> I think your text is actually antialiased already, >>> >>> Thanks for looking into this. I'm not sure that this is the case though. As >>> a crude workaround, I sub-classed Label and overrode paint(Graphics2d), >>> allowing me to manually add the RenderingHints. I tried both >>> VALUE_TEXT_ANTIALIAS_LCD_ON and VALUE_TEXT_ANTIALIAS_LCD_HRGB. Both >>> returned different results to the default (VALUE_TEXT_ANTIALIAS_LCD_HRGB >>> seemed to be the best). >>> >>> but I found that you can override what kind of hint is used by adding this >>> code somewhere in your app before the rendering takes part, for example in >>> the Application#startup() method: >>> >>> Field aaHintValueField = >>> FontRenderContext.class.getDeclaredField("aaHintValue"); >>> aaHintValueField.setAccessible(true); >>> aaHintValueField.set(Platform.getFontRenderContext(), >>> RenderingHints.VALUE_TEXT_ANTIALIAS_GASP); >>> >>> Thanks, I'll give this a go - looks to be a far cleaner option than my >>> current workaround. >>> >> >> > >
