Re: [Nautilus-list] current thinking on canvas vs. anti-aliased canvas

[Maciej Stachowiak <mjs eazel com>]

Seth Nickell <seth eazel com> writes:

> I realize that there are various factors governing the deployment of the
> anti-aliased canvas vs. the regular canvas. However it would help me
> decide how much / how good / how specific gradient code I plan to write
> depending on current thinking and factors (and guesses) going into the
> decision. While the canvases are somewhat compatible (eg it shouldn't be
> too hard to move code between the two) I think it would be preferable to
> at least point the code I'm writing in a particular direction. This code
> could be deferred...but it looks like people are going to be wanting
> things like good 3-point gradients fairly soon (and it'd be better to
> use a generic framework ripped off, say, the gimp rather than dealing
> specifically with every case). Flame me!
> 

OK, you asked for it. :-) I think we should do the minimum work
required to do the current required features (decently rendering
simple 2-point linear gradients), or, if we decide we need more
complex gradient rendering, add appropriate tasks to the `required'
milestone.

As far as the canvas goes, it is possible we may ship with the ability
to switch between the two modes still present, since there is a major
tradeoff of performance vs. graphics quality, and the desired point in
the tradeoff may not be the same for all users.

 - Maciej