<div dir="ltr">I'm sure that there are many others beside me who would like to have inkscape working on ARM tables without having to resort to floating point emulation libraries for cairo, as that will probably slow down inkscape quite a lot. (But I'm just guessing.) Thus, an integer solution, might be preferable for performance reasons. (At least optionally).<br>
<br>Dov<br><br><div class="gmail_quote">On Wed, Aug 11, 2010 at 11:50, Jean-Claude Repetto <span dir="ltr"><<a href="mailto:jcrml@mxm.eu">jcrml@mxm.eu</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Le 11/08/2010 08:00, M Joonas Pihlaja a écrit :<div class="im"><br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>
On Wed, 11 Aug 2010, Soeren Sandmann wrote:<br>
<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
For the matrix in pixman_image_t, I think we need to just move to<br>
single precision floating point unconditionally. There are several<br>
cases where 16.16 fixed point simply doesn't have enough either range<br>
or precision.<br>
</blockquote>
[snip]<br>
<br>
Before deciding on the kind of floating point, could we think very<br>
hard about the precision and accuracy guarantees pixman would like to<br>
make for each of its operations? Single precision floating point is<br>
not very far off from 16.16 after all, so I worry that we'll be in the<br>
same boat very quickly again if we're not careful.<br>
<br>
Joonas<br>
</blockquote>
<br></div>
Please don't forget that Cairo is used on some processors that don't have FPU (such as ARM processors). So if you introduce floating point, it MUST be optional.<br><font color="#888888">
<br>
Jean-Claude</font><div><div></div><div class="h5"><br>
<br>
--<br>
cairo mailing list<br>
<a href="mailto:cairo@cairographics.org" target="_blank">cairo@cairographics.org</a><br>
<a href="http://lists.cairographics.org/mailman/listinfo/cairo" target="_blank">http://lists.cairographics.org/mailman/listinfo/cairo</a><br>
</div></div></blockquote></div><br></div>