On Sat, Mar 29, 2008 at 3:58 PM, Steven Watanabe <<a href="mailto:watanabesj@gmail.com">watanabesj@gmail.com</a>> wrote:<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;"> AMDG<br> <div class="Ih2E3d"><br> Robert Dailey wrote:<br> > Well I'm not 100% sure of the performance impact RTTI will have on the<br> > application. What I've read is that RTTI can get very expensive in<br> > terms of executable size if you have a lot of classes in your application.<br> <br> </div>You should be able to switch between using RTTI and using ID's in the<br> dispatcher with<br> a few lines of code. *Measure the difference* before making decisions<br> for performance<br> considerations.<br> <div class="Ih2E3d"><br> > Unfortunately I don't think I could get away with using strings since<br> > the network is assigning integral values to the packets in order to<br> > uniquely identify them.<br> </div>Of course, if you need the ID's around anyway for other reasons, they are<br> not a problem. I wouldn't expend Herculean efforts to eliminate them.<br> <div class="Ih2E3d"><br> > I like the string idea because it doesn't have the repetitive code in<br> > each derived class (the overridden GetID method that will be the same<br> > in each derived class).<br> <br> </div>struct WalkPacket : CRTPPacket<WalkPacket, PID_WALKPACKET> {}; ?</blockquote></div><br>What exactly is this structure supposed to present?<br>