<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><br><div><div>On Jan 4, 2010, at 7:01 PM, Chris Ribe wrote:</div><blockquote type="cite"><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left-width: 1px; border-left-style: solid; border-left-color: rgb(204, 204, 204); margin-top: 0pt; margin-right: 0pt; margin-bottom: 0pt; margin-left: 0.8ex; padding-left: 1ex; position: static; z-index: auto; "><font class="Apple-style-span" color="#000000"><br></font> I feel we have established the latency concerns are real and of no significance to this mode of operation.<div><div></div><div class="h5"><br></div></div></blockquote><div><br>Latency is absolutely a significant issue. When it is on the order of 50-500ms, it will cause a horribly annoying echo at any frontends that are within earshot of each other.</div></div></blockquote><div><br></div><div>If we take 50 ms as the threshold, ethernet will be more than OK, but wireless will be a problem. Pinging wired hosts on my home network, I get round trip times of less than 1 ms, even with large packets. However, my wireless hosts go as high as 100 ms -- and it's inconsistent, so even an adaptive approach wouldn't work. Keeping consistently tight sync over wireless may not be a solvable problem without adding a lot of complexity.</div><div><br></div></div></body></html>