LL>       Watch the ones that are built onto the motherboard.  I have two
 LL> that aren't capable of running at any real speed.  A 
 LL> third board I didn't even bother to check, I just 
 LL> dropped a COM board in and moved on.
The motherboard-based ones are the good ones.  The ones that are on the Dual 
Async/A cards are the ones that don't quite perform as well. 
 > One of my IP (actually a local college) providers uses old
 > DECservers which lock the attached USR Sportster 33.6k modems at
 > 19.2, but in reality, because LAT is so inneficient, they only
 > get 14.4k speeds.  (the term servers aren't smart enough to run
 > their own PPP, so the PPP has to be run on a VAX system).
 LL>       LAT shouldn't have any trouble running 19.2, we use it on
 LL> ethernet.  I suspect the problem is in tunneling TCP/IP 
 LL> or the VAX itself.  We've upgraded our system to use 
 LL> Shivas, but with a little work we could have done it 
Yeah, it's done through UCX.  It is an older model of VAX (a MicroVAX 
3100-80) machine, so maybe that has something to do with it.
 LL> with Multinet or UCX through the VAX.  The drawback 
 LL> there is that it adds to the work of an already 
 LL> overloaded VAX cluster. Just tell them to get a new hub 
 LL> with some PPP capable servers for their modems. 
They're too cheap around here.  First of all, Sportsters are *not* 24/7 
modems designed to be used in modem pools.  With that out of the way, they 
actually have the equipment sitting there, and have for the past two years, 
but they are too lazy to deploy it, saying "if we set PPP up properly, then 
everyone would want to use it, and the 10 lines would easily be saturated"  
(only 2 lines, right now, at any given time, are in use).  The excuse used to 
be "our 56k net connection just can't handle it".  They went and upgraded the 
net connection to a 10megabit/second connection (fractional T3), and they 
just had to change the excuse..
--- Maximus/2 3.01
---------------