[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Gnash-dev] Re[2]: Text quality (was: Questions about trapezoids, triang

From: Udo Giacomozzi
Subject: [Gnash-dev] Re[2]: Text quality (was: Questions about trapezoids, triangle stripper etc.)
Date: Mon, 2 Oct 2006 19:30:08 +0200

Hello strk,

Monday, October 2, 2006, 5:48:58 PM, you wrote:
s> Outlines are available, and rendered. But it seems that iff
s> generate_font_bitmaps is called, every glyph would *also* have
s> a bitmap associated, and the latter is used, which is just nicer.

We should let the renderer decide what's better.

>> As for the AGG backend, the rendering of small glyphs is not that bad
>> I think. See the "glyph-test" at

s> Really good!
s> Is it done using the shape or the bitmap ?

Currently no bitmaps are supported by the AGG renderer. I still have
to study that. The flexibility of AGG makes it difficult to learn. For
example, there are at least ten (!) different methods of
rotating/rescaling a image available. The good thing is that gradients
work very similar to bitmaps. So, when one of them works, the other
should be easy to implement.

s> This is what I'm also trying to understand.
s> There's also an attempt to cache identical bitmaps for different glyphs...

Identical bitmaps? For glyphs that look identical?

s> I'm tempted to trash everything and leave only the render-from-shape
s> part...

Do we still need the Cairo backend (provided that AGG works)? Is there
any really working backend except for the OpenGL one?

Anyway, we could use the AGG routines to draw the bitmap cache
in-memory even for the OpenGL backend. That would make everything much

Can someone verify that the AGG licence allows us to integrate AGG
into the Gnash source tree?


reply via email to

[Prev in Thread] Current Thread [Next in Thread]