HP3000-L Archives

February 1997, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Wirt Atmar <[log in to unmask]>
Reply To:
Date:
Sat, 22 Feb 1997 00:40:57 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (93 lines)
Before Christmas, I mentioned that we were going to put together a new HP
terminal emulator for windows-based PCs that would be distributed for free. I
also mentioned at that time that we would probably have it ready for release
in either February or March. I thought that I would write and give a brief
update.

As with all software projects, we're going to be a bit late :-). We're
clearly not going to make February as a first release date -- and if we make
March, it will be towards the very end of March. Indeed, I would say that
middle-April is now a more reliable date for a first pre-alpha,
take-what-you-get release.

However, saying that, the terminal is actually working out quite nicely.
Although it is still some ways from being releaseable, I use it almost all
day, every day now. About 30-50 man-hours a week is being put into its
development, almost all of that time being supplied by Vickie Kurtz, an
engineer in our Albuquerque office. And she's doing a really nice job.

Vickie worked for AICS Research in Las Cruces for five years, fifteen years
ago, and obtained her Masters in Electrical Engineering while she worked
here. After her masters, she left to begin work for Sandia National
Laboratories in Albuquerque, where she worked on nuclear weapons effects for
13 years until she grew disgusted by the work. She then taught science and
mathematics at a Catholic middle school in Santa Fe for two years before
rejoining AICS last summer.

The terminal is being put together wholly in Visual Basic, and I'm pleased
with the speed and performance that we're getting out of it. Although neither
QCTerm nor Reflection are as fast nor as well optimized as a standard HP
700/92 terminal, QCTerm is proving to be 20% faster and noticeably smoother
in its screen handlings than Version 5.1 of Reflection on a 16MB, 133MHz
Pentium-based processor running Windows 95. On the other hand, for reasons we
do not yet fully understand, QCTerm is about 20% slower than Reflection's
Version 4.10 on a 4MB, 486 66MHz/DX2/Windows 3.1 machine.

To date, we've tested QCTerm on several Windows 3.1 486 machines, a Windows
3.1 Cyrix 586 machine, and one 133MHz Pentium running Windows 95. We've had
no trouble on any platform. >Theoretically<, it should also run without
trouble on a Windows NT machine, but we've yet to test it there.

Some time ago, the question was asked on HP3000-L: "What's Reflection?" The
answer that was provided was that Reflection was a program that turned your
colorful, GUI-based, icon-populated PC into a "dull and boring terminal that
you used to see 20 years ago." I took the "dull and boring" comment rather
seriously, so we're trying to put QCTerm together so that it will be a little
less dull-and-boring.

Clearly, QCTerm still has to be a terminal that is fully compatible with all
of the software that has been written for terminal displays, but we're
putting it together more in the idiom of a browser than a strict terminal.
Indeed, the order in which the various idioms that have been adopted for the
construction of QCTerm probably ranks in this manner:

     o Standard HP 700/92 terminal
     o Netscape/Internet Explorer browsers
     o Reflection
     o Standard Macintosh command sequences

The last may seem strange, but Microsoft has adopted these command key
sequences for most of its products (such as VB and IE), as has Netscape, AOL,
and many others. We intend on using them, also. (With Bill Gates at the helm,
I'm not sure that the Macintosh is ever going to die. It's simply going to
change platforms -- but Apple unfortunately isn't going to be part of its
revenue stream any longer).

Putting QCTerm together has reinforced to us what a well designed product
Reflection is. We probably won't be fully competitive with (meaning equal or
better than) Reflection until the end of the year.

We're designing QCTerm especially for the new generation of PCs that will
appear later this year. We expect these new PCs to have these minimum
capabilities: >100MHz Pentium processor, 16MB RAM or greater, >1GB drive,
65,000 colors, running Windows 95 or later, and priced at or less than
US$1,000. These new machines will be basically in the same ball park as the
price that HP would like to charge for a new terminal. A free terminal
emulator should make such PCs an attractive alternative.

And we're designing QCTerm to be a good citizen. As a test, I keep QCTerm
running all of the time that I'm on my Pentium and 486 processors simply to
see if I can detect any noticeable load on the machine. On the 4MB 486
processor, there is some noticeable slowdown, most especially associated with
virtual memory swaps. But on the Pentium, it has proven to be essentially
invisible.

When the first release does occur, please do not widely distribute it. This
is going to be a pre-alpha test copy. You're going to get while the bits in
its program are still hot from the forge. A good, hard distributable release
should be available perhaps by mid-autumn.

As things progress, I will continue to update you on our schedule.

Wirt Atmar

ATOM RSS1 RSS2