Date: Wed, 15 Oct 1997 10:03:58 -0600 (MDT) From: Nate Williams <nate@mt.sri.com> To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> Cc: Donald Burr <dburr@poboxes.com>, ports@freebsd.org Subject: Re: 8 days until 2.2.5... Administrative notices. Message-ID: <199710151603.KAA11877@rocky.mt.sri.com> In-Reply-To: <199710151416.KAA17656@khavrinen.lcs.mit.edu> References: <9710150749.AA14643@bragg> <XFMail.971015020727.dburr@POBoxes.com> <199710151416.KAA17656@khavrinen.lcs.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
[ Mega-reduction of Cc line ] > > Getting foo.tar.gz (380k/1522k), 25% done, 3.6 K/sec, 0:18 remaining... > > > The k/sec figure would be useful to make sure you're getting the most ouf > > or your modem (or to brag about your high speed internet connection), and > > the time remaining is would be very useful for deciiding whether you want > > to sit there and wait for the port to download, or if you want to abort > > it). > > Except that the way TCP works, they are both totally bogus, No, they *can* be totally bogus, but most of the time they are a *great* approximation of what is happening. > particularly in the presence of even a small packet loss rate. With small and/or big packet loss, you can still get a pretty good 'average' rate. Think like a user who needs feedback and not like a scientist. Scientists need hard numbers, users will settle for approximations. Nate - The user who likes approximate download times and K/sec fields....
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199710151603.KAA11877>