Page 1 of 1
New "fast TCP" being developed at Caltech will allow download speeds of up to 8.6 Gbs... pretty nifty.
PostPosted:Sun Jun 08, 2003 8:50 pm
by ManaMan
<div style='font: 12pt Arial; text-align: left; '><b>Link:</b> <a href="
http://www.zap2it.com/movies/news/story ... ">Internet 2</a>
New "fast TCP" being developed at Caltech will allow download speeds of up to 8.6 Gbs... pretty nifty.</div>
PostPosted:Sun Jun 08, 2003 9:15 pm
by SineSwiper
<div style='font: 10pt "EngraversGothic BT", "Copperplate Gothic Light"; text-align: left; '>For what I hear on the threads from Slashdot, it's just fluffed up bullshit.</div>
PostPosted:Sun Jun 08, 2003 9:46 pm
by Kupek
<div style='font: 10pt verdana; text-align: left; padding: 0% 10% 0% 10%; '>What I understood was that the popular press coverage is - surprise - not representing the research accurately. The research is legit, but it is portrayed inaccurately.</div>
PostPosted:Sun Jun 08, 2003 10:20 pm
by SineSwiper
<div style='font: 10pt "EngraversGothic BT", "Copperplate Gothic Light"; text-align: left; '>I've read that taking away the critical error checking/correction out of TCP theorically increases speeds, but it falls apart in the real world.</div>
PostPosted:Mon Jun 09, 2003 7:30 pm
by Kupek
<div style='font: 10pt verdana; text-align: left; padding: 0% 10% 0% 10%; '><b>Link:</b> <a href="
http://netlab.caltech.edu/FAST/overview ... ew.html</a>
I don't think it has anything to do with error checking. They're changing the underlying algorithms. But this stuff is over my head - or, at least, beyond my understanding unless I commit to understanding it.</div>