[BUG][FIX][TEST] BitBlt alpha compositing ([sm][approved])

Brent Vukmer bvukmer at blackboard.com
Wed Apr 9 18:18:58 UTC 2003


Does this fix take care of the 32-bit display issues reported recently on the list? (See http://swiki.gsug.org:8080/sqbugs/760.html , for the most recent.)

> -----Original Message-----
> From: Doug Way [mailto:dway at riskmetrics.com]
> Sent: Wednesday, April 09, 2003 2:12 PM
> To: The general-purpose Squeak developers list
> Subject: [BUG][FIX][TEST] BitBlt alpha compositing ([sm][approved])
> 
> 
> 
> I also verified that the simulated tests pass, at least.
> 
> - Doug Way
> 
> 
> Andreas Raab wrote:
> > 
> > Hi Folks,
> > 
> > Attached is a test+fix for alpha compositing in BitBlt. It 
> turns out that
> > when you're using Form>>blend (aka: alpha compositing) the 
> resulting alpha
> > values are not at all what they should be. VERY problematic 
> for stuff like
> > anti-aliasing.
> > 
> > [Note to harvesters: Please consider integration ASAP. This 
> presence of this
> > fix in the appropriate VMs will be critical for Croquet].
> > 
> > Cheers,
> >   - Andreas
> > 
> >   
> --------------------------------------------------------------
> ----------------
> >                          Name: BBCompositing.2.cs
> >    BBCompositing.2.cs    Type: ChangeSet 
> (application/x-unknown-content-type-cs_auto_file)
> >                      Encoding: quoted-printable
> > 
> >                      Name: BitBltTests.st
> >    BitBltTests.st    Type: unspecified type 
> (application/octet-stream)
> >                  Encoding: quoted-printable
> > 
> >   
> --------------------------------------------------------------
> ----------------
> 
> 



More information about the Squeak-dev mailing list