<pre><span lang="EN-CA" style="font-family:Arial;
mso-ansi-language:EN-CA">&gt;</span><span lang="EN-CA" style="color:black"> </span><span style="color:black">Do you have any reason to believe that if you implement your own protocol (or re-implement existing protocols) you will actually reduce latency?</span></pre>


<p class="MsoNormal"><span style="font-size:10.0pt;font-family:Arial"></span>It bypasses
OS and transport and application layer by going directly to wire is the reason
why I think it will reduce latency. </p>

<p class="MsoNormal"><span style="font-size:10.0pt;font-family:Arial">These two
articles on <a href="http://wincap.org">wincap.org</a> also imply the same</span></p>

<p class="MsoNormal"><a href="http://www.winpcap.org/docs/WinPcap-SBAC03.pdf">http://www.winpcap.org/docs/WinPcap-SBAC03.pdf</a></p>

<p class="MsoNormal"><a href="http://www.winpcap.org/docs/iscc01-wpcap.pdf">http://www.winpcap.org/docs/iscc01-wpcap.pdf</a><span style="font-size:10.0pt;font-family:Arial"></span></p>