Consumer Report
Microsoft Windows 3.0
Part II: A Closer Look
Open Forum
2.88MB Extra-High Density
3^-lnch Disk Drive
This & That
PC TAP Consumer Report Index
New ERN Reps
Report #9
October 1990
PC Technology Assessment Program
EPA National Data Processing Division
Information Centers Branch - RIC II, MD-35
Research Triangle Park, NC 27711
Telephone: (919) 541-0568 (FTS) 629-0568
Printed on Recycled Paper
-------
PC TAP CONSUMER REPORTS
From the Editor's Desk
Shortly after the last PC TAP Consumer Report was published, we received an anonymous piece of mail
addressed to "Editor, PC TAP." The only return address was a zip code in an EPA Regional Office city.
Here in its entirety is our unnamed correspondent's message:
While your pub. is printed on recycled paper (at a higher cost to
taxpayers!); because it was dved blue (not inked), it is no longer
recycleable (sic) as white ledger (which has a high value). It's now
considered mixed waste with a low value & EPA does not recycle
this type of paper. What a waste!
We usually ignore the opinions of those who don't have enough confidence in what they're saying to
identify themselves. And we didnt think anyone would want to recycle our reports anyway, but would
keep them indefinitely for reference! But because of the seriousness of the allegation that we're
environmentally irresponsible, we decided to investigate the validity of this charge. We called the EPA
print shop, where our Consumer Reports are printed, and we talked to the folks at the EPA Headquarters
Recycling Office. The lessons we learned in paper recycling put us at ease about our publication.
For recycling purposes, paper falls into one of three categories. White paper is "high grade," and it usually
brings the highest price from recycling companies. An intermediate grade includes colored bond paper.
Then there's a "low grade" category into which everything else falls; newsprint is a familiar example of low
grade recyclable paper. Some organizations, and some recycling companies, eliminate the intermediate
grade; they just classify paper for recycling as high grade or tow grade. Although recyclers pay more for
high grade, the cost difference isnt significant enough to warrant discontinuing use of other papers;
besides, the initial cost of low grade papers often is lower. Generally, EPA recycles all grades of paper,
although at some sites there may be local exceptions to this practice.
PC TAP Consumer Reports are printed on recycled paper, and that paper (even though it's blue) is
recyclable again. We have been assured that our use of this paper is economically and environmentally
responsible, and that there's no reason for us to change. The EPA Headquarters Recycling Office invites
anyone who has questions about recycling to call them at FTS 382-6980.
On a brighter note, this Consumer Report focuses again on Microsoft Windows 3.0. It includes more in-
depth testing than was discussed in our preliminary report on Windows in July, and it reflects input from
many more people. We're grateful for the product assessment feedback we received from our PC TAP
External Resource Network representatives in the regional offices and some labs. These folks are what
PC TAP is all about: users telling their peers about their experiences using microcomputer technology.
Open Forum begins on page 19. It contains a report about extra-high density disk drives for PCs. In This
& That, beginning on page 16, you will find an index of reports published during the first year of PC TAP
Consumer Reports. We're pleased to have passed the one-year milestone, and look forward to continuing
to research and write about microcomputer technology.
David A. Taylor
PC TAP CoonHmtor
-------
Microsoft Windows 3.0: A Closer Look
Introduction
Along with our First Impressions of Microsoft Windows, published in July, we promised a more detailed
evaluation would follow. This Closer Look includes results of additional testing by PC TAP and others
within the National Data Processing DMsion, as well as input from a diverse group of end users
around the Agency through the PC TAP External Resource Network.
This report is aimed at readers who have at least an acquaintance with Microsoft Windows 3.0. No
fundamentals will be presented, and it’s assumed that you’re familiar with the terminology of graphic
user interfaces in general and Windows 3.0 in particular. If you don’t have that background, we
recommend you read our earlier report, Microsoft Windows 3.0 Part I: First Impressions (PC TAP
Consumer Report #8, July 1990). Copies are available from Anne Harrison at the Washington
Information Center, FTS 382-7412 or Email A.HARRISON.
In the previous report on Windows we discussed the tremendous amount of printed material we had
gathered about the product. Well, the avalanche hasn’t let up yet, and our stack of Windows articles
keeps growing. Before examining the results of our own testing and that of users around the Agency
who have been participating in our Windows evaluation project, we’ll present a synopsis of information
we’ve gathered from the media since our first report was prepared. With that background in mind, we
then will discuss our own recent experiences, and those of others within NDPD, as we’ve probed
Windows further than we had when we prepared our first report. Finally, we’ll see how the feedback
from others correlates with our own assessment.
The Ongoing Saga of Windows 3.0
Articles about Windows continue to occupy prominent space in many of the trade publications. We
thought the hoopla that began in May with the announcement of Microsoft Windows 3.0 would subside
after the initial wave of new-product information passed. However, as we’ve already mentioned, the
media blitz continues. The opinions expressed In the trade press are overwhelmingly favorable, but
there have been some grumbles as well. First, lets look at excerpts from some negative comments.
Problems With Windows
In our first report on Windows we noted that incompatibilities had been discovered between Windows
3.0 and Ontrack’s DMDRVR disk manager, when running on a 386 machine. This problem, along with
some others, received some media attention during July and August.
Windows 3.0 users concerned about hang-ups or data loss because of disk
utility conflicts can run one of several tests to determine their systems’ status,
Microsoft said. The problems occur only in several very specific configurations
—“Disk Utility Conflicts With Windows Acknowledged,” InfoWorld, July 9, 1990,
p.3.
Three months after Microsoft Corp.’s flashy Windows 3.0 release, some users
putting the new environment through its paces are reporting a slew of minor
problems in getting the software up and running.. . . Though few of the reported
problems are bugs in Windows, the complexity of the environment and the
2
-------
demands it makes on system resources are exposing conflict that previously went
unnoticed, users said. . . . —“Early Users Face Ups and Downs Of Windows 3.0,”
PC Week, August 13, 1990, pp. 1, 6.
Just about every PC manufacturer claims its computers are ‘100 percent IBM
compatible.’ But as the programmers and users of Windows 3.0 have recently
discovered, 100 percent compatibility is not a foregone conclusion. . . . Because
not all PC hardware is completely compatible, software vendors must detect
differences and accommodate them. . . Whether Windows 3.0 overcomes these
incompatibilities, however, is a point of contention between Microsoft, which has
spent thousands of hours trying to weed out bugs, and frustrated users, who are
having difficufty running Windows.—”Windows 3.0 Exposes Hardware
Incompatibilities,” PC Week, August 20, 1990, p. 19.
Although there seems to be no question that some users are having difficulty running Windows 3.0,
it appears the problems aren’t within Windows; they’re specific to particular configurations of hardware
and/or software. We explained the DMDRVR problem in our previous Consumer Report, and we
haven’t seen a more definitive treatment of that issue since then. Remember, though, that problem
is specific to running Windows 3.0 in 386 enhanced mode on a 386 processor that is also running
Ontrack’s DMDRVR software. Since relatively few EPA users currently are running on 386 machines,
this should not be a significant issue around the Agency. Those of you who are fortunate enough to
have 386 PCs, or who may have upgraded 286 PC5 with 386sx or true 386 chips (especially the
Epson Equity 111+, which was shipped with DMDRVR installed), should carefully investigate this issue
before installing Windows on such machines.
More Kudos
No single product will ever satisfy all the people all the time, and Windows is no exception. However,
the majority opinion of Windows 3.0 seems to be quite favorable. Here are quotes from some of the
articles we’ve read that express that opinion.
Here at last is an elegant, smoothly working interface that multitasks, takes full
advantage of the extended memory in a 286 or 386, and supports networks, yet
requires as little as 1MB of RAM for efficient operation... . Even if you dislike
icons and windows, Windows 3.0’s solution to DOS’s 640K memory limit is too
compelling to ignore.—”DOS Updated for the ‘90s,” PCResource, July, 1990, pp.
20-21.
Yes, we’re stuck with Windows, but I’m smiling as I say this.. . My favorite
DOS tools. . . fly like eagles in the 386 Enhanced mode of Windows 3.0, with its
disk-based virtual memory and EMS emulation. Under Windows, my 386 acts as
if it has twice its actual RAM.—”Users Stuck in the GUI Gumdrops May Not Want
To Escape,” PC Week, July 23, 1990, p. 33.
it’s true that more and more semi-power users are moving to Windows 3.0
every day. . . - Windows is here to stay, and that’s good news too.—”Career
Opportunity: Windows 3.0 Configurations Engineer,” InfoWorld, August 13, 1990,
p. 25.
At $149 (list price—Ed.], Microsoft Windows 3.0 is a real bargain, when you
consider all the functionality it offers. . . . Windows is, of course, a graphical
environment, and while it can multitask DOS applications gracefully, that is hardly
its primary focus. The graphical user interface. . . conveys an image of refined
quality.—”386 Multitasking Environments,” PC Magazine, October 16, 1990, p. 202.
3
-------
To summarize what we’ve read, there certainly are mixed reviews on Windows. The predominant
opinion seems to be that it Is a strong product that’s destined to be a trend setter for some time.
Forecasts of just how long that influence will prevail vary widely; we’ve seen figures ranging from one
to ten years. Ultimately, time and user demand probably will dictate the answer.
PC TAP Test Results
PC TAP’s primary test machine is an Epson Equity 111+ with 640K of memory and a 40mb hard drive.
We use that configuration because it is typical of most machines in use throughout EPA. Much of the
data for our earlier Windows report resulted from our experience running Windows in real mode on
this machine. You may remember that Windows operates in real mode on a 286 processor with no
more than 640K of memory.
Running Windows In Standard Mode
Standard mode is the next step up from real mode as you climb the Windows performance ladder.
A 286 processor with at least a megabyte of memory (640K conventional memory and at least 256K
of extended memory) is required for standard mode. The primary advantages of standard mode over
real mode are increased performance and the capability to run more applications concurrently. The
significance of the actual increase in performance depends upon several factors, as does the limit on
the number of concurrent applications you can run. Applications written specifically for the Windows
3.0 environment can be muititasked in all Windows’ modes.
Adding Memory
To enable us to run Windows in standard mode on our test machine, we installed an Intel Above
Board with two megabytes of memory, all of which was configured as extended memory. Although
we didn’t detect any noticeable improvement in the speed of Windows’ operation, we did gain
increased capacity. Within Windows, Microsoft defines capacity as how many applications you can
run simultaneously, and how much data applications can store in memory at a time. Before we
installed the Above Board, the initial amount of available memory reported by the Windows Program
Manager was around 390K. With the 2K of added memory, this figure increased to something in the
neighborhood of 2500Kl
The reason why memory is enclosed within quotation marks in the previous paragraph, and why the
amounts of available memory in real and standard modes are given as approximations, requires some
explanation. The Navailable memory value reported by the Windows Program Manager is based on
two factors: (1) the actual amount of conventional and extended memory that’s available for
applications; and (2) the amount of hard disk space that’s available to Windows. This becomes
apparent if you have a volatile hard disk. For example, we’ve added several more applications since
we installed the Above Board, and the available memory TM has decreased accordingly. The additional
applications didn’t eat up memory, but they do occupy disk space that Windows could otherwise use
for temporary files, and for application swapping.
To give you an idea of how this process works, we started Windows and checked available memory
and disk space. Then we started loading applications and data files, while keeping tabs on memory
and disk space as we went along. Here are the steps we went through:
4
-------
1. Start Windows, check values: Disk spate available, 2,498,560
bytes; memory available, 2236K.
2. Launch WordPerfect 5.1; bring in a 179-page document.
3. Switch from WordPerfect to Windows Program Manager.
4. Launch Lotus 1-2-3 from Windows desktop; load a 31 .275-byte
spreadsheet.
5. Switch from Lotus 1-2-3 to Windows Program Manager.
At this point, Windows Program Manager reported 1.077,248 bytes of free disk space and 2076K of
memory still available. The memory-resident portions of the applications and data files we had
activated occupied only 160K of memory. But when we switched out of the applications, Windows
swapped them to disk where they consumed a megabyte and a half of space. This shows how
Windows keeps memory available for the application in which you happen to be active at a given time,
but it also illustrates why you need lots of free disk space if you want to get optimum performance from
Windows in standard mode. You can keep starting applications and switching out of them. But
eventually you’ll try to switch out when there’s no longer enough hard disk space to hold any more
applications, and Windows will issue a message instructing you to terminate the active application
because there’s not enough memory available to accommodate the switching operation. In fact, disk
space is the constraint, not memory.
One final note on the memory issue. Even though you might have plenty of memory and disk space
available, it’s still possible to get a ‘not enough memory’ message if you try to load an application that
doesn’t recognize any memory above 640K, and there’s not enough available memory below 640K
to hold that application. We experienced this with Freelance Plus (version 3.0), whIch needs 520K of
memory available to run at all. Although we ran Freelance under Windows earlier In our testing, after
installing an additional device driver, which is always memory-resident, we no longer had enough
memory to load Freelance. Even though Windows said there was over two megabytes of memory
available. Freelance only recognizes memory addresses up to 640K, and 520K was no longer
available below 640K.
Speed Tests
We’ve mentioned already that we didn’t detect any noteworthy improvement in the execution speed
of Windows in standard mode over that of Windows in real mode on the same PC. However, we had
heard it said that DOS programs run slower under Windows than they do outside the Windows
environment. We set out to check into this allegation with respect to WordPerfect, and found some
slight basis for it. The table below illustrates our findings. The test document was 179 pages long.
Functfon,
Wmd ws Wor iPerfect
Standalone WP
Reposition from beginning
to end of document
10 Seconds
10 Seconds
Global replace. 27 occurrences,
last occurrence on laSt ge.
22 Seconds
20 Seconds
5
-------
As the table shows, it took two seconds longer for the global replace function when WordPerfect was
running under Windows. We repeated the test several times, and the results were consistent—the 2-
second gap was preserved. While this test confirms that some functions take longer when a program
is running under Windows, it also indicates the magnitude of the differences we’re talking about.
In another test of operating speed under Windows, we started up three applications—WordPerfect,
Lotus 1-2-3, and DOS—and switched between them. Here are the times for the switching operations.
p_o
p....
WordPerfect to Lotus 1-2-3
11-13 Seconds
Lotus 1-2-3 to Windows File Manager
8 Seconds
Lotus 1-2-3 to DOS
11-13 Seconds
In our first Windows report we listed context-switching times of from 8 to 10 seconds, and these times
are In the same ballpark. Again, whether you consider these times OK or not depends on your
perspective. If you’ve never had a context switching capability before, 12 seconds to switch from
Lotus to Windows Program Manager, or from Lotus to dBase or WordPerfect, is probably quite
satisfactory. But if you’ve been running on a 386 PC where you can switch between windows
applications about as fast as you can hit the keys, then 12 seconds can seem like an eternity.
To summarize our findings relative to moving from real mode to standard mode in Windows 3.0, the
primary gain is an increased processing capacity. In other other words, you can launch more
applications before you exhaust your machine’s resources, and you can accommodate larger files.
The magnitude of the capacity increase depends upon how much free disk space and memory is
available in your machine. Performance improves as the amount of available memory and disk space
Increase. But the really significant performance boost occurs when you move up to a 386 processor.
Running Windows In 386 Enhanced Mode
Windows’ 386 enhanced mode provides access to the 386 processor’s virtual memory capabilities.
A 386 processor with a minimum of 2 megabytes of memory is the minimum required hardware
configuration. (A 386 machine with less than 2MB of memory operates in standard mode.) In
enhanced mode, true Windows applications have access more memory than is physically present in
the machine. For example, Softmart Inc., a software retailer, tested Windows 3.0 on a 386 machine
with eight megabytes of RAM. In that configuration, Windows provided access to 16 megabytes of
expanded memory to Lotus 1-2-3 Release 2.2 (Microsoft Windows Version 3.0, Softmarf Product
Review, September 1990). Another advantage of 386 enhanced mode is the capability to multi-task
Windows 3.0 versions of non-Windows applications.
We ran Windows in 386 enhanced mode on an IBM PSI2 Model 70 with 6 megabytes of memory. As
with the move from real mode to standard mode, this is another step up in the Windows performance
ladder. That’s where reasonable comparisons end, however. The performance improvement from real
or standard mode to enhanced mode could be likened to moving from a shetland pony to a race horse.
6
-------
We’ve established that Windows will run on a basic 286 machine with 640K, and that it runs better on
a 286 with additional extended memory and several megabytes of available hard disk space. With the
move to a 386 machine, Windows doesn’t get better; it moves into another performance dimension.
While context switching definitely beats having to exit from one application in order to start up another,
having several applications (including Windows 3.0-compatible non-WIndows applications) running
simultaneously with all visible in windows on your screen at the same time is something elsel
Let’s consider an example. From Windows in 386 enhanced mode, we started up the Windows Write
program, a basic word processor. Then we activated the Windows Clipboard and copied the text from
Write to the Clipboard. If you have the windows positioned such that you can see them both on your
screen, you can actually watch the text as it’s written onto the Clipboard. Next, we launched
WordPerfect in a third window, and pasted the text from the Clipboard into a WordPerfect document.
Finally, we cleared the clipboard, entered new text in WordPerfect, and reversed the process.
The cut-and-paste capabilities vary, depending on the types of applications you’re working with. That
is, Windows application to Windows application, Windows application to non-Windows application (with
other variations depending on whether the non-Windows application is running within a window or in
full-screen mode), and so on. You can cut and paste text, graphics, or a combination of text and
graphics; but these operations are constrained by the applications involved. In other words, you can’t
paste graphics into an application that doesn’t recognize the graphic image. In the exercise described
in the preceding paragraph, we tried to paste a graphic image into Write, and what we got was a solid
black square where the image should have been. Write didn’t recognize the graphic image. The
user’s guide is quite clear on how each alternative is accomplished, and with a little practice one could
easily master the technique for often-needed operations.
With respect to speed, there’s no point in preparing comparison charts. For all practical purposes, just
assume that tasks like context switching, moving between active windows on the screen, and
performing cut-and-paste operations will happen about as fast as you can strike the necessary keys
or click the mouse. After spending some time on the PSI2, we gained some insight into why people
who are used to such machines say Windows on a 286 is too slow. It’s easy to get spoiled by the
power and speed of the 386.
In summary, 386 enhanced mode provides more computing power and a lot more speed. It also offers
capabilities, like multiprocessing of non-Windows applications, for example, that aren’t possible in real
or standard modes. Enhanced mode presents the Windows environment at its best, and it’s easy to
understand why Microsoft (and a lot of others who’ve evaluated Windows 3.0) recommend the 386
processor for optimum Windows performance. Keep in mind, however, that the basic system on the
American Coastal Industries (ACI) contract is a 386sx with only 1 megabyte of RAM. So if you buy
that system and intend to run Windows in 386 enhanced mode, you must order additional memory.
The standard and high-performance systems come with 6 and 8 megabytes of RAM, respectively.
Windows’ Appetite for Resources
In some circles, Windows has been given the rather indelicate label of uresource hog.N It’s true that
it requires significant amounts of memory and disk space; the User’s Guide recommends 6-8
megabytes of free disk space. On the other hand, you get a lot back in performance: memory
management, context switching or multitasking, a file manager, a print manager, and lots of other
goodies, some of which we’ve already discussed. By contrast, OS/2 Extended Edition version 1.2 with
all the its options requires around 20 megabytes of hard disk space.
7
-------
On our Epson Equity 111+ we were hurting for disk space, so we devised our own version of “Windows
optimization ” we deleted some nonessentials. Windows comes with a lot of “nice but not necessary”
features. First, there are two games, Solitaire and Reversi; they’re neat, but hardly a requirement to
get the job done. There are also quite a few bitmaps that you can use to put a fancy background on
your desktop. Within Windows, they’re called “wallpaper.” Pretty, but unessential. You might also
consider deleting some of the accessones. (Do you really need another calendar? Calculator? How
about the Clock?) There are files that are specific to running in 386 enhanced mode that you don’t
need if your machine is a 286. We deleted a number of such files, along with Windows’ Help, and
reduced our Windows directory down to a little over 3MB. By contrast, the Windows directory on our
PS/2 test machine, which includes all the files that come with the package, occupies 5.3 megabytes
on the hard disk. In Chapter 13, “Optimizing Windows,” the User’s Guide provides guidance for
selecting files that can be eliminated to free up disk space.
More Test Data
Some interesting test data on Windows 3.0 performance was just published in a Software Digest
Ratings Reporr (Volume 7, Number 10, October 1990). The Software Digest publisher, NSTL, Inc.,
of Plymouth Meeting, PA, has generously granted permission for us to pass along the following
information from that report.
Data for the report was gathered while testing four applications under Windows 3.0 (Microsoft Excel,
Ami Pro, PageMaker, and Word for Windows) on a variety of computer configurations (IBM PS/2 286,
386, and 486; Compaq 286 and 386). The performance of Excel and PageMaker under Windows also
was compared with Macversions of those packages running on Macintosh llfx, Ilci, and SE computers.
Windows 3.0’s performance was considered totally acceptable, and significant improvements in
performance over previous versions of Windows and other environments were noted. For multitasking,
OS/2 was found to offer advantages over Windows, but in terms of cost and single-application
performance Windows was considered superior. On a 286 or 386 processor, Windows outperformed
comparable Macintosh systems. This was attributed to Windows’ superior memory management
capabilities. However, a Mac llfx beta system was found to be slightly faster than Windows on a 486
machine. The following quotation from the report’s conclusions concisely summarizes the findings.
Users do not have to add memory because Windows 3.0 runs in 640KB, but
adding memory lets Windows take advantage of extended memory.... Windows
provides an environment that is truly easy to learn and use, and one that comes
bundled with enough extras to justify the cost. The costs of upgrading applications
to run under version 3.0 are minimal, especially when weighed against
performance benefits (p. 14).
Running WIndows on a LAN
One obvious solution to the problem of limited resources is to run Windows as a local area network
application. From a resource point of view, the advantage of Windows on the LAN is that you don’t
have to install it on your own hard disk. The program files go into the LAN system directories, and
certain user-specific files can be placed in your LAN user directory. On our LAN at the RTP
information centers, the Windows subdirectory for an individual user occupies about 1.1 megabytes.
It should be noted, however, that if you want to, you can install the user-specific files on the PC’s hard
disk.
8
-------
Unlike previous releases of Windows, version 3.0 is very LAN aware.” For example, if your LAN
drivers are in memory when you start up Windows, the Windows File Manager displays your LAN
drives along with those that are physically on your PC. You can list directories and manipulate files
on the LAN drives in the same way you do the files on your own machine. With this approach, you
get all the advantages of running under Windows without having to pay the penalty in terms of impact
on your own machine’s resources.
Windows’ performance on a LAN is determined in part by the capabilities of your workstation. In other
words, Windows will operate in the mode appropriate for the amount of memory and the
microprocessor in your PC. And since networked Windows uses the file server in the ways the
standalone version uses your hard disk, you can run LAN Windows from a diskless workstation.
Based on our experience running LAN Windows In RIP, and on industry reports, Windows is well
suited for use as a networked application.
The End Users Speak
To support our evaluation of Windows 3.0, Microsoft Corporation supplied copies of Windows to all
the PC TAP External Resource Network (ERN) representatives. There are ERN representatives in
each regional office and in a number of laboratories (see This & That on page 16). Along with
Windows, Microsoft also shipped copies of their software suite for Windows 3.0: Word, Excel,
PowerPoint, and Project. Although the main thrust of the PC TAP assessment project was to evaluate
Windows as a platform for EPA’s supported PC application software (Lotus, dBASE, WordPerfect, and
CrossTalk), we asked our evaluators to optionally consider testing Microsoft’s application software.
Feedback from those who were willing and able to do so is included in the following discussion.
PC TAP provided a questionnaire to help evaluators record their assessments of Windows 3.0. The
questionnaire divides the evaluation of Windows into these broad areas:
• Installation of Windows
• The Windows File Manager
• Running non-Windows Applications (DOS applications)
• Running Windows Applications
• Rating Prominent Windows Features
Finally, we asked each participant to assign an overall rating based on their own testing of Windows.
Interpreting the Evaluation Data
Twenty-two questionnaires were returned to PC TAP from project participants in Regions ll-IV and VII-
X; the labs in Ada, Ann Arbor, Athens, Cincinnati, and Corvallis; Washington, DC; and RTP. However,
all the evaluators did not respond to every questionnaire item. Within the graphs that accompany the
discussions that follow, the percentages of evaluators responding to each item are given. On the bar
graphs, the bar indicates the mean score for a product, on a 5-point scale with 5 being the most
favorable rating. The number of respondents who actually rated that particular package is indicated
by a percentage atop each bar.
9
-------
In the following paragraphs, we’ll see how the participants In our study rated Windows In each of the
five broad evaluation areas we defined eailier. In some cases, when our own experiences Illuminate
the evaluation data or when our conclusions differ from those of our evaluators, we may elaborate on
those aspects of the study.
Installing WIndows 3.0
We said In our preliminary report that the Windows Installation process was easy and straightforward.
Data from our evaluators supports our original position on this Issue. All 22 participants responded
to the questions about using Windows’ setup program to Install the product, or to modify the original
configuration sometime later. On a 1-to-5 scale, with 5 beIng the most favorable rating, here’s how
the Installation process fared:
Installation
1 % 100%
• II .
LrI I IL
A frequent comment about the Installation process was that there is a lot of documentation and general
Information on the disks (in read-me files) that Is not included in the manual, particularly Information
about Windows’ disk space requirements. Afthough a number of people commented on Windows’
capability during the installation process to recognize executable files on the hard disk as programs
the user might want to Install under Windows, several commented that programs had been mis-
identified during the process. In one case PC Wnte was Identified as Multimate; In another, a program
named QuIc*en was labeled Quattro. As the graph shows, however, our evaluators were generally
well satisfied with the installation process and with Windows’ Setup program.
The Windows File Manager
We asked evaluators to compare the Windows File Manager with AUTOMAXX for file management
functions. Not surprisingly, responses to this item on the questionnaire varied widely. Some people
thought File Manager was the greatest; others preferred AUTOMAXX. The majority fall into a third
10
-------
management from the DOS prompt, who are indifferent about
group, many of whom prefer to do file
File Manager versus AUTOMAXX. The
accompanying chart shows how the
responses were distributed. One person
said File Manager Is “very nice, easy to ______
understand, easy to navigate,” but pointed
out that It shouldn’t read the directory
structure each time it’s started. The latter
comment refers to the fact that the first time
you call up File Manager, It builds a
directory tree for your primary drive
(normally C). If you exit from File Manager,
then Immediately call it up again, the
program rebuilds the tree again as If you’d
never been there before. It would be more
efficient If the tree were retained In memory
for the duration of the Windows session.
RunnIng non-Windows Applications
One of the major objectives of this evaluation project was to dig further into the question, “Is Windows
3.0 a viable platform on which to run EPA’s current standard menu of products’?” To help answer this
question, we asked all our evaluators to exetclse the software on the AUTOMAXX menu under
Windows. While not everyone reported on every product, all our respondents rated one or more of
them.
In addition to the standard products, we also inquired about other non-Windows applications that
respondents may have tested. Among those “others” respondents reported testing were the following
(in alphabetic order): Arbiter, DrawPerfect, First Publisher, FOCUS, GoScript, Harvard Graphics,
Kermit, Knowledge Pro, Novell Utilities, PageMaker, PCWrite, Procomm, O&A, Turbo Pascal, and
WordPerfect Office. Some of these were noted by more than one respondent. Here are the ratings.
File Manager Preferences
Overaii Windows File Mgr Rating - 3.5
V FkM
•AUTOMA)O(
L NoP an s
Non-Windows Applications
77 % 4S%
4 4M1
1 1 1 1
d ss sds XT Oisi.
11
-------
Running Windows Applications
A ‘Windows” application is one that’s designed to run under Windows 3.0. It takes full advantage of
the Windows graphical environment, and uses Windows’ conventions for pull-down menus and dialog
boxes. Althcsugh all the ERN representatives received the four Microsoft products for Windows we
mentioned earlier, we were pleased that several respondents reported testing other Windows
applications as well. Packages mentioned in this category included CoreIDRAW, Kappa, Knowledge
Pro, Level 5 Object, and Ventura Publisher. Sixty percent of all respondents rated Windows
applications. Within the following chart, the percentages atop the bars indicate how many of that sixty
percent rated each package.
Windows Applications
5 60% of Respondents Reporting 1 -
Excel PowerPoint Word Project Others
Prominent Windows Ffatures
One part of the evaluation asked for respondents opinions about the major features and capabilities
that set Windows apart from other environments. The primary items are reflected in the graph on the
following page. The “Accs ” category included the Windows accessories (Write, Paintbrush, Notepad,
Recorder, Cardfile, Calendar, PIF Editor, Terminal, Calculator, and Toolbook). While some of the
accessories were rated by a majority of respondents, others were not. For example, 86% tried the
Write accessory, while only 23% said they looked at the Recorder.
12
-------
. : Prominent Windows Features
5
4
3
2
I
0
Other significant items on the questionnaire dealt with whether using function keys and arrow keys,
as opposed to operating with a mouse, is a viable way to function within Windows; with the merits of
the graphical Interface as opposed to a menU-driven character-based system; and with Windows’
operating speed. Eight respondents had done their evaluations of Windows without benefit of a
mouse. Seven of those eight people felt that a mouse should be considered a necessity for Windows’
GUI environment. One person felt that using Windows without a mouse is an OK situation. On the
five-point scale, with 1 being I like commands” and 5 representing “GUI beats entering commands,”
the average score with all participants responding was 3.8. Opinions about use of a GUI vs. a menu
system and about Windows’ operating speed are reflected in the graphs below. It should be noted
that the speed rating for 486 machines reflects input from only one person.
•is
___ Windows 3.0
a Execution Speed
User Interlace Preference
:NoA.spons thsr
t3
-------
An Overall RatIng
Last but not least, we asked all respondents to Indicate their overall assessment of Microsoft Windows.
Here is a graphical representation of the overall mean rating.
In conjunction with the overall product rating, we asked for general
quotations from those comments:
comments.
Have you thought about the cost of converting to Windows? Mouse, extra
memory, cost of Windows software itself? Windows is nice, but it’s not for
everyone.
I’m very disappointed that Windows doesn’t run under my DOS emulator (from
within Unix).
I feel Windows is an outstanding product which is destined to become a de facto
industry standard.
I currently prefer PC Tools over Windows because of memory and disk space
limitations.
I like Windows 3.0 for most tasks, but for some it really does not add much of
value...
Windows 3.0 is a vast improvement over earlier versions, but as an everyday
operating environment it does not represent a step forward on the type of
machines we currently use.
On a 640K 286 PC, I prefer to use a traditional menu system.
After our users learned the mouse and mastered hand-eye coordination, Windows
received a favorable response. In fact, it is now the platform of choice for graphics
and desktop publishing.
Here are some
Windows 3.0
Overall Rating
I-Pow
14
-------
On our 8MHz PC/AT, the response was too slow to fully realize Windows’
potential. Furthermore, most of the EPA-supported software would not be
immediately enhanced by installing Windows.
Windows’ greatest benefit, outside its beauty, may be to the less experienced user.
Windows is really easy and very intuitive, but the hardware requirements are hard
to satisfy. -
Windows is a good GUI on a platform with lots of disk space 4 lots of memory,
12MHz or better speed, a mouse, VGA monitor, and ALL Windows-compliant
software.
I do not use Windows on a 286 PC. On a 386 with at least 4 MB of memory, I
would definitely use Windows.
Much improved over previous versions of Windows. Should use on a 386 PC.
Highly recommended, since protected mode negates memory headaches.. . no
more RAM crami
Summary
We haven’t encountered anyone who was ambivalent about Windows 3.0; most users express clear
opinions about it. Many of those who are critical of Windows coma from the “power user” camp.
They’ve tailored their systems using a variety of sophisticated software tools, many of which may not
be compatible with the Windows environment. They say, “Why should a give up this, this, and this,
in order to be able to operate under Windows? I don’t like GUIs and mice anyway.”
The obvious answer for this group is, “Windows probably isn’t a good choice for you at this time.” On
the other hand, some of those people may find that the functions they’re getting from several other
products can be performed by a single product, Windows. Memory management, multitasking or
context switching, cut-and-paste between applications, and a graphical user interface are examples
of such functions.
The longer Windows 3.0 has been around, the less vocal its detractors seem to have become. The
product clearly is a success in the marketplace. Reports of alleged bugs have quieted. The
advantages and viability of the product, both in its own right and as a transitional environment on the
way to OS/2, are widely touted in the trade press. Examples of successful implementations of
Windows as an organizational standard are common. Our opinion is that the Agency should give
serious consideration to supporting Windows 3.0 as an available alternative environment for MS-DOS
microcomputers. The feedback from participants in our PC TAP evaluation of Windows indicates that
in general they agree.
Certainly there’s no urgency to replace AUTOMAXX with Windows in the installed base of machines
that use the 286 microprocessor, although we think its a viable option in that environment for those
who wish to do so. However, all the workstations on the current PC contract have at least a 386sx
processor, and Windows certainly is worth considering on that platform. Furthermore, a general trend
toward graphical user interfaces appears to be developing. Regardless of whether Windows is or
should be a “standard” in the Agency or industrywide, it would be prudent to consider It carefully as
we develop computing strategies for the next several years.
15
-------
This & That
PC Technology Assessment Program
New External Resource Network Representatives
The PC TAP External Resource Network (ERN) Is made up of people in the Regions and Labs who
are Interested In technology assessment, and who are willing to serve as local contacts for PC TAP
at their respective sites. In our last Consumer Report (#8, published in July) we printed a list of all the
ERN representatives. Since then, Erica Duvai, our Washington representative, has left her position
at the Washington Information Center, so we have a new ERN rep in DC. Also, the labs in Ann Arbor,
Michigan, and Edison, New Jersey, have Joined the ERN. Here are the representatives at these three
locations:
Representative
Phone/EMAIL
Mitch Cumberworth
Motor Vehicles Lab
2565 Plymouth Road
Ann Arbor, Ml 48105
Rich Koustas
EPA RREL, Mail Stop 104
3890 Woodbridge Ave.
Edison, NJ 08837-3679
8-374-8342 M.CUMBERWORTH
8-340-6898 RCB/STDD/RREL
Leslie Yambor
EPA WIC
401 M Street SW
Washington, DC 20460
8-475-7414 LYAMBOR
We also note with sorrow the recent sudden passing of Diana Smith, our Region IV representative.
Her replacement will be announced later.
Site
Ann Arbor
Edison
Washington
16
-------
PC TAP Consumer Reports Index
August ‘89 thru July ‘90
Eight PC TAP Consumer Reports have been published since the first one appeared in August 1989.
Heres a listing of the reports that have been published to date.
By Issue
Report Date Features
#1 Aug. 89 Personal Information Management Systems
Open Forum: Sun Workstation/DOS Windows
#2 Sep. 89 Graphics Software for Scientific Applications
Desktop Pnnters
Open Forum: MS Windows DOS Extension
Macintosh 35mm Slides
#3 Oct. 89 Color Hardcopy Output
Technology Assessment Around EPA
Open Forum: PC 386 Upgrades
#4 Nov. 89 PC Graphics File Transfers
#5 Jan. 90 Desktop Scanners
#6 Feb. 90 WordPerfect 5.1 Evaluation
Open Forum: Scientific Graphics Followup
#7 May 90 Text Retrieval Software
Open Forum: PacificPage PostScript Cartridge
This & That: HP LaserJet Ill
#8 July 90 Microsoft Windows 3.0—
Part I: First Impressions
Open Forum: ChemDraft II Chemical Structures
This & That: The PC TAP ERN
#9 Oct 90 Microsoft Windows 3.0—
Part II: A Closer Look
Open Forum: Low-Cost Macintosh Printers
This & That: Consumer Report Cumulative Index
17
-------
Articles
By Thie
Title ReDort Nr. Date
ChemDraft II Chemical
Structures Software 8 Jut. 90
Color Hardcopy Output 3 Oct. 89
Desktop Printers 2 Sep. 89
Desktop Scanners 5 Jan. 90
Graphics Software
for Sdentific Applications 2 Sep. 89
Graphics Software for Scientific
Applications: A Followup 6 Feb. 89
HP LaserJet Ill Printer 7 May 90
Macintosh Printer Options 9 Oct. 90
Macintosh 35mm Slides 2 Sep. 89
Microsoft Windows 3.0—
Part I: First Impressions 8 Jul. 90
MS Windows DOS Extension 2 Sep. 89
PacificPage PostScnpt Cartndge 7 May 90
PC Graphics File Transfers 4 Nov. 89
PC TAP External Resource Network 8 Jul. 90
PC 386 Upgrades 3 Oct. 89
Personal Information Management Systems 1 Aug. 89
Sun Workstation/DOS Windows 1 Aug. 89
Technology Assessment Around EPA 3 Oct. 89
Text Retrieval Software 7 May 90
WordPerfect 5.1 Evaluation 6 Feb. 89
18
-------
Open Forum
Open Forum provides an opportunity for users to share with others their own
innovations, or the results of their own technology assessments. The PC Technology
Assessment Program neither verifies nor endorses the contents of Open Forum Items,
but we are pleased to offer them as a service to users.
2.88MB Extra-High Density
3½-Inch Disk Drive for PCs
This nwassul miued by Gienr Piper atormeGmStedent ntemlü
the RTP Ubtary. Glenn has since left the Obraty to aecept a position wfth
American Coastal b AjStz1OSb 0* wlnt* o thø Interim P0 Contract. P TAP
appreciates GIenn s conttibution.
The TEAC disk drive kit is intended for installation in IBM PCs and compatibles. The kit includes the drive,
a controller board, mounting hardware, installation instructions, and a 5¼-inch diskette containing the
necessary software. According to the documentation, the drive will read, write, and format 720K (double
density), 1.44MB (high density), and 2.88MB (extra-high density) 3½-inch diskettes. The drive will read
720K and 1.44MB diskettes formatted on other drives, but in order to format diskettes the special formatting
program that comes with the kit must be used.
We installed the TEAC drive in an Epson Equity 111+. The only problem we encountered was that the
mounting bracket that came with the kit would not fit in the Epson. Except for this problem, which we
overcame by obtaining the proper bracket, the installation was trouble-free. The instructions were clear
and obviously thought-out.
Although the physical installation of the controller card was simple, getting it to work was more difficult.
The factory settings for the controller card were incompatible with the Epson’s configuration (several other
boards had previously been installed in the machine). After two calls to Computer Technologies’ technical
support line, the correct jumper settings on the controller board were determined, and the installation was
completed successfully.
With the hardware installation behind us, we began testing to see if the drive would perform as advertised.
The Practidisk software was very easy to install, and its automatic configuration option worked well. In
many ways, we found the Practidisk formatting utility easier to use than the DOS formatting commands.
We had no problems reading from and writing to 3½-inch diskettes of all three densities. Diskettes of all
three densities that were formatted and written upon on other drives were handled with no problems by the
TEAC drive. Furthermore, on other computers we had no problem reading or writing upon diskettes that
had been formatted and written to on the TEAC drive.
All things considered, the TEAC extra-high density drive lived up to its advertised capabilities, and it passed
all our tests with flying colors. For the most part, the documentation was clear, understandable, and easy
to follow, and Computer Technologies’ fast and courteous technical support staff more than made up for
the gaps we found in the literature. However, those who plan to install this drive in an Epson Equity 111+
should be sure they have the proper mounting hardware on hand before beginning the installation.
19
-------
How to Submit Items for Open Forum
In keeping with the PC Technology Assessment Program's objective to have the user community actively
involved in TAP projects, users are encouraged to submit items for inclusion in future PC TAP Consumer
Reports. If you have independently investigated the capabilities of a software product or a hardware
component, we would like to hear from you. We'd also like you to share with others your solutions to any
problems you may have encountered with a particular application or device, and about tricks, shortcuts,
or unique applications you have devised. Although we can't promise to publish every contribution, we will
evaluate them all in terms of their potential interest to our readers and their conformance to the spirit and
intent of PC TAP.
There are no additional rules for Open Forum contributions, but here are some guidelines:
1. Contributions must be typed. Our first preference is that they be
submitted on a floppy disk in WordPerfect format. If that isn't
possible, the next best method is to EMAIL the text to PCTAP,
EPA30647. The least preferable method, but still acceptable, is to
mail a typewritten article to TAP at the address on the cover of this
publication.
2. The length of your contribution will be determined somewhat by its
complexity. However, keep in mind that we're primarily interested in
the purpose of your study project and how pleased you were with the
results, not in the nitty-gritty details of how you did it. We will publish
your name, address, and phone number for those who want more
details. Two to three pages is probably a reasonable maximum
length. On the other hand, a paragraph containing a nugget that may
be useful to others would be equally welcome.
3. All material submitted by users is subject to our editing, and you will
not be given an opportunity to review the final manuscript before
publication. Sorry, you'll just have to trust us. If we have questions
or don't understand any part of your text, we'll contact you for
clarification.
We hope you enjoy PC TAP Consumer Reports, and we look forward to hearing from individuals who
have insights or discoveries to share with others. Thanks for your interest and your participation in the
PC Technology Assessment Program.
------- |