Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 exptools; site cbuxc.UUCP
Path: utzoo!watmath!clyde!cbosgd!cbuxc!gmv
From: gmv@cbuxc.UUCP (Mike Vrbanac)
Newsgroups: net.micro,net.micro.pc
Subject: Re: Multitaskers (really DoubleDos)
Message-ID: <353@cbuxc.UUCP>
Date: Mon, 14-Oct-85 12:29:35 EDT
Article-I.D.: cbuxc.353
Posted: Mon Oct 14 12:29:35 1985
Date-Received: Tue, 15-Oct-85 20:26:12 EDT
Distribution: net
Organization: AT&T Bell Laboratories - Columbus Comp. Center
Lines: 24
Keywords: DoubleDos
Xref: watmath net.micro:12354 net.micro.pc:5601

As a user of DoubleDos, I have one thing to add to the discussion.
DoubleDos keeps an image of the video buffer in memory. Thus, as the
background task writes to the screen via DOS calls, DD intercepts it
and updates its own buffer instead. So far so good.  The drawback 
to this approach is that each and every software package you run
that writes to the video buffer *directly* must be modified in order
to work under DoubleDos.  Softlogic provides a bunch of debug zap
batch files to fix 1-2-3, dBase-n, Framework, etc, etc to issue a
DoubleDos INT instead of blasting away to the video buffer. I run
Symphony 1.1, but Softlogic has not issued the zaps for it yet.

One further point: you need not modify a package if you never run it
as the invisible task. This obviously limits the effectiveness of
the product.

The above comments aside, I have found DoubleDos to be a solid product.

-- 

 +--------------+------------------------+-----------------+
 | Mike Vrbanac | AT&T Bell Laboratories | Columbus, Ahiya |
 +--------------+------------------+-----------------------+
 | Unix: {cbosgd|ihnp4}!cbuxc!gmv  |  CompuServ: 76054,113 |
 +---------------------------------+-----------------------+