This isn't a rigorous benchmarked lab test yet, but I found something bizarre this morning. After fixing a little router problem, I tested the download speeds on my DSL line from AT&T and my cable Internet line from Time Warner Cable. I have both for product testing needs.
My DSL is pretty low speed, officially about 1000 kilo bits per second on download and 256 kilo bits per second upload. Testing that with my new (and still aggravating) Vista machine on the Bandwidth.com test meter, I got 1223 kpbs download and 321 kbps upload. That's about normal.
Then I tested my Time Warner cable connection on my primary work machine running Ubuntu 8.04. My cable modem connection is always much faster than my DSL by a factor of 10 on the download speeds. I forget what Time Warner's latest promises are for download speeds, but I think it's 10 mega bits per second to 12 mbps. Upload speeds are throttled down to 1mbps.
My Ubuntu machine returned a rating from the Bandwidth.com test of 22-25mbps over several tests. That's darn fast today, faster than normal. Then I did the same test from a Windows XP PC and got results from 12-14mbps. Still fast, but not nearly as fast as the Ubuntu machine.
The two computers are almost identical: both off-lease Compaq small form factor D515s, part of the very popular corporate desktop D500 family. Both have Pentium 4 processors running at 2GHz. The Ubuntu machine has 768MB of RAM, while the XP box has only 512MB of RAM. Both run Firefox 3 as their browser.
Can a little extra RAM make that much difference in Internet download speeds? I don't think so. Sure, more RAM helps, but can it add 7-10mbps to your download speed? No. That makes me think that Ubuntu handles networking faster than Windows XP.
If you have both operating system in your company, running on comparable PCs, try the test yourself. It could be I got some random weirdness, but I tested the two systems back and forth for seven iterations for good luck. There's a big difference in my two systems. Let me know how much difference you find.