My vote would be to make it configurable between "direct" and
"buffered" rendering.  The default should probably be consistent with
the original behavior (ie, buffered) for backward compatibility. Our
applications don't use zooming, and fidelity is important to us, so we
would typically use "direct".

All of that said... I don't recall if fidelity was the only issue that
direct rendering solved.  I'll discuss with my colleagues to see if
any of them have a better memory.


On Mar 5, 3:36 pm, wrote:
> Comment #5 on issue 173 by Slow repaint of PSwing  
> components when zooming (from 1.2 to 
> 1.3)
> Before Issue 82, PSwing would render to a buffer and then render that  
> buffer to the Graphics2D. The result would be that the buffer would undergo  
> the applied transforms.
> Issue 82 changed this so that the Swing component rendered directly to the  
> (transformed) Graphics2D. It turns out the painting a JComponent with  
> transform applied is much more expensive than painting the buffer.
> Of course you lose fidelity when translating the buffered version (see  
> picture.)
> Should we restore the buffered approach? Should we make this configurable?
> Attachments:
>         2v3.jpg  30.6 KB

Piccolo2D Developers Group:

Reply via email to