<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div style="font-family: 'DejaVu Sans'; font-size: 9pt; font-weight: 400; font-style: normal;">
<div class="im">
> > > I'm waiting whether somebody approve my commit which uses QImage as<br>
> > > buffer in canvas although we have many buffers in the render chain...<br>
> ><br>
> > That is still a concern.<br>
><br>
> Any chance to test this patch through beta stages and revert if it shows<br>
> performance problems?<br></div>
I leave that decision to Boudewijn :)</div></blockquote><div><br>Of course the last word is for Boudewijn, but i vote for joining prescaled projection's buffer and canvas' one. This will lead to some changes in canvas architecture, but it'll create really good, clear, easy to understand system.<br>
</div><div><br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div style="font-family: 'DejaVu Sans'; font-size: 9pt; font-weight: 400; font-style: normal;">
<div class="im">
> > What's wrong with the black&white lines solution ?<br>
><br>
> I really think that some composite operation is better because it can help<br>
> you identify e.g. the borders of some area you paint. You can see on the<br>
> outline where is different colour. Black and white just covers part of the<br>
> image.<br>
><br>
> Second argument is pure subjective -- it is ugly :)<br></div>
since xor is ugly as well...</div></blockquote><div><br>For me Xor is not so ugly as two pixel heavy lines. More than that, this is the way other editors work. That's why i'm for xor, but with clear system.<br></div>
</div><br clear="all"><br>-- <br>Dmitry Kazakov<br>