Just checked in a possible fix. Again, a bit tough for me to test this. It's kind of a hack, but it might work. It should automatically detect whether antialiased (multisampled) drawing is supported to an FBO.<br><br>Michael<br>
<br><div class="gmail_quote">On Sat, Aug 7, 2010 at 8:08 PM, Michael Lawrence <span dir="ltr"><<a href="mailto:michafla@gene.com">michafla@gene.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
As I said, it's only used for antialiasing cached drawing. So at worst, any cached layers will not employ antialiasing. I need to decide between figuring out a way to "catch" this error or simply leave it to the user. I think auto-detection is the best, but it will take some work.<br>
<font color="#888888">
<br>Michael</font><div><div></div><div class="h5"><br><br><div class="gmail_quote">On Sat, Aug 7, 2010 at 4:19 PM, Tengfei Yin <span dir="ltr"><<a href="mailto:yintengfei@gmail.com" target="_blank">yintengfei@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div>Yes,your idea can solve the problem of detecting different hardware and toggle cache automaticlly, could be placed on the top of our package.</div><div><br></div><div>But I still have a concern that sometime turning off cache make interactive graphics slow, not quite responsive(there is a case in my demo here), which means although users with integrated graphic could run the demo, but it could be slow sometimes. I am just wondering, do we have to leave it that way or is it still possible for integrated graphic users to turn on cache in some other way?</div>
<div><br></div><div>Regards</div><div><br></div><div><font color="#888888">Tengfei<br><div><br></div></font><div><div><div></div><div> <br><br><div class="gmail_quote">On Sat, Aug 7, 2010 at 4:03 PM, Yihui Xie <span dir="ltr"><<a href="mailto:xie@yihui.name" target="_blank">xie@yihui.name</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">I'm thinking about a function like par() in our development (qpar()?)<br>
to specify some global options...<br>
<div><br>
Regards,<br>
Yihui<br>
--<br>
Yihui Xie <<a href="mailto:xieyihui@gmail.com" target="_blank">xieyihui@gmail.com</a>><br>
Phone: 515-294-6609 Web: <a href="http://yihui.name" target="_blank">http://yihui.name</a><br>
Department of Statistics, Iowa State University<br>
3211 Snedecor Hall, Ames, IA<br>
<br>
<br>
<br>
</div><div><div></div><div>On Sat, Aug 7, 2010 at 6:00 PM, Tengfei Yin <<a href="mailto:yintengfei@gmail.com" target="_blank">yintengfei@gmail.com</a>> wrote:<br>
> ...This concerned me, in my project here, I sometimes use cache=TRUE in the<br>
> package just for efficiency issues, which means my demo may not be possible<br>
> to run in your laptop or other computers with integrated graphic card.<br>
> It looks like if I want to satisfy all the users, I always have to set<br>
> cache=FALSE, is it possible to use cache and satisfy all the users with or<br>
> without discrete graphics?<br>
><br>
</div></div></blockquote></div><br><br clear="all"><br></div></div><div>-- <br>Tengfei Yin<br>MCDB PhD student <br>1620 Howe Hall, 2274,<br>Iowa State University<br>Ames, IA,50011-2274<br>Homepage: <a href="http://www.tengfei.name" target="_blank">www.tengfei.name</a><br>
<br>
</div></div></div>
</blockquote></div><br>
</div></div></blockquote></div><br>