On Wednesday 27 February 2002 02:37 pm, you wrote: > On Wed, 27 Feb 2002, Troy.A Johnson wrote: > > 8am is an odd time for Quake play > > (I am assuming it isn't 8pm), but to > > each their own, I suppose. > > Two seperate thoughts... users log on around 8 AM, Quake playing just > slows it down in general. > > -Brian > Logging on does -- and I've monitored it -- not consume a whole lot of bandwidth, not when we're talking about a 100-meg network or network segment that isn't a constant problem when people are doing more bandwidth-intensive stuff than that. But the time coincidence is . . . interesting. But even if everybody is, say, refreshing their IE caches of whatever websites they monitor upon boot (easy to set up in IE), that would have to be a lot of people monitoring a lot of websites AND refreshing caches at the same time AND some majorly fast Internet connection, given that you can't even flood a 10meg network with a T1 line. The tool I'm most familiar with for this sort of thing is a Windows-only one -- Observer (see http://www.netinst.com) -- and, when I was writing their sales collateral stuff a year or more ago, I couldn't find a competitive Linux package that did all the stuff it did as easily. What I'd want to see is a time-scaled chart of protocols used and stations using them, and packet statistics by station, and (possibly at the same time) see what the network error level is like. I've heard of network cards misbehaving until they're properly warmed up, but I've not seen it. ------------------------------------- There's a widow in sleepy Chester Who weeps for her only son; There's a grave on the Pabeng River, A grave that the Burmans shun, And there's Subadar Prag Tewarri Who tells how the work was done. -------------------------------------