Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.6.2.17 $; site uiucuxc.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!ihnp4!inuxc!pur-ee!uiucdcs!uiucuxc!kline
From: kline@uiucuxc.UUCP
Newsgroups: net.micro
Subject: Re: Orphaned Response
Message-ID: <34500020@uiucuxc.UUCP>
Date: Tue, 19-Feb-85 17:52:00 EST
Article-I.D.: uiucuxc.34500020
Posted: Tue Feb 19 17:52:00 1985
Date-Received: Fri, 22-Feb-85 08:29:38 EST
References: <1312@eosp1.UUCP>
Lines: 17
Nf-ID: #R:eosp1:-131200:uiucuxc:34500020:37777777600:763
Nf-From: uiucuxc!kline    Feb 19 16:52:00 1985


Yes the can't-type-while-redrawing has certainly been fixed in the 3278
series terminals. I wouldn't give up my 3178 for the world.

While we're discussing 80-users-on-a-4341-running-VM versus 80-users-on-
a-780-running-4.2, let's also remember that a LOT of effort went into the VM
scheduler to give extremely quick service to interactive requests. The less
of your timeslice you use and the longer the interval between becoming
eligible, the better scheduling breaks you get. 

Of course I'm not denying that generating one interrupt for a whole
screenful of typed data on a 3270 isn't a big factor also. And having
intelligent users who don't press ENTER or PF2 after every line typed helps
too.

Charley Kline
University of Illinois Computing Services Office