I'm secretly sorta surprised that using two framebuffers (one for the background and one for the foreground/text) would make a significant difference to just blitting the background and rewriting the text (as we currently do). If it does make a significant difference doesn't it just mean our text blit code is in need of optimising?
  ----- Original Message -----
  From: Paul Louden
  To: Rockbox development
  Sent: Sunday, June 25, 2006 12:45 PM
  Subject: Re: Another list optimization

  Isn't the text and the mask the same thing? You just have two framebuffers. The "foreground" which is only drawn in the foreground color and is primarily text. And the actual framebuffer. The backdrop is stored wherever it is currently (which I suppose you could call a framebuffer but it never changes really.) Since the foreground image is always 1-bit, 1 is drawn in the foreground color, 0 uses the pixel at that location from the backdrop buffer instead, right?

  I'll admit that I don't know the code at all, but it seems kinda simple since text is always drawn 1-bit. I'm probably missing something.

  On 6/25/06, postmaster@diffenbach.org <postmaster@diffenbach.org> wrote:
    postmaster@diffenbach.org wrote:

> I've now implemented another optimization, used when items "move off"
> the top or bottom of the list. The implementation just copies the
> framebuffer over itself, to move the unchanged items up or down in the
> viewport.

    If you have a 60 GB ipod only, you can try it here:
    http://diffenbach.org/rockbox/listopt2-noscroll/rockbox.zip
Received on Sun Jun 25 15:32:35 2006