home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Fritz: All Fritz
/
All Fritz.zip
/
All Fritz
/
FILES
/
INFOTEXT
/
REVIEWS.LZH
/
SIIG386.TXT
< prev
next >
Wrap
Text File
|
1988-11-07
|
8KB
|
127 lines
Review of the SIIG S386, Model 2100A
Copyright 1988 David W. Batterson
While the average end user is not going to rush out and buy a 386 system,
my guess is that within three years the 386 may well replace the AT as the
entry level PC for most businesses.
By that time, 80486 machines should be starting to grab a nice piece of
the pie too.
For many companies, it's definitely worth an extra $1500 or so to get a
386 instead of a 286 system.
SIIG (Suntek Information International Group) is pushing its S386 series
of high-performance PCs toward large companies, MIS professionals, VARs (Value
Added Resellers) and consultants. You won't see these 386s at ComputerLand or
Sears.
Their S386 2000/2100 series is designed around the VLSI (Very Large Scale
Integration) chip set. The BIOS is from AMI (American Megatrends Inc.), which
has captured a major part of the 386 BIOS market.
The 2100A model I reviewed features 1MB RAM, dual speeds of 20MHz/8MHz,
and 0 wait state. You switch between 8MHz and 20MHz with the keyboard, using
<Ctrl> <Alt> and the - or + key. You can switch from 0 to 1 wait state by
pressing the <Ctrl> left <Shift> <Alt> - keys.
There were 2 serial ports, one parallel and a game port. The power supply
was a hefty 200 watts. The supplied floppy drives were the usual 1.2MB
high-density 5 1/4" and 1.44MB 3 1/2", both from Toshiba.
The S386 comes with SIIG's own Suntouch 101 click-tactile keyboard. It
has a nice touch, though not as good as Northgate's OmniKey/102. While it
features oversized <Shift> keys, the <Backspace> key is too small, and hard to
reach because the \| key is placed between it and the =+ key.
The video board supplied was SIIG's own VideoStar ECH2. The monitor I
reviewed was a 14" QC-1448 EGA monitor from Quadrant Components. It
automatically switches to your graphics adapter, whether it's MDA, CGA or EGA
(or a compatible one).
The dot pitch is a respectable .31mm. The actual usable picture area is
about 13 1/4" (I don't know why monitor manufacturers count the area that is
covered up). The QC-1448 features a non-glare screen with dark tint.
Although manuals are not crucial with a monitor, I always feel a little
better about a product when it comes with a professional looking manual. This
one consisted of poorly photocopied pages, with some words hard to decipher.
The monitor's controls are easy to reach, with power switch, brightness
and contrast on the right front. There are pre-set sub-brightness and
precontrast controls in back. Two vertical size controls (for CGA and EGA
modes) in back adjust picture height.
The swivel and tilt base is adequate, with the tilt varying 25 degrees.
Colors were rich and accurate, and text was sharp.
Now what about the PC's setup?
The BIOS features a built-in setup and diagnostics program, so you don't
have to boot a floppy to get started on the installation. You only have to set
the time and date, and indicate the hard disk type. My unit came with a
Seagate 40 MB, model ST251, with a fast access time of 28ms.
The setup works either in automatic or manual mode (for power users who
know EXACTLY what they are doing). In most cases, the automatic setup will
work fine.
It automatically detects the type of display card, size of real memory,
memory beyond 1 MB, presence of a 360K floppy drive, and presence of an 80287
math coprocessor.
In the diagnostics section, which you run by choice, there are numerous
tests for the hard drive, floppies, keyboard, video, and misc. tests, including
parallel and serial ports.
The setup seemed to proceed OK. The unit, however, did not come with
MS-DOS and GW-BASIC disks. I already had these, so it was nothing to be
concerned about. A new user would probably get a bit agitated.
The setup lets you choose how you want the hard disk partitioned, such as
20MB in C: and D:, or 32MB in one, and 8 MB in the other. I opted for the
latter, saving the smaller storage area of the D: drive as a "spare tire."
That way, if you fill up the C:, you just remember that you still have 8MB
left!
As formatted, the hard disk actually came up with 33.4 MB for the C: drive
and 9.2MB for the D:, making a total of 42.6MB.
The SIIG comes with a disk cache utility, although I wasn't able to
determine its source. It wasn't commercial software, and had a typed label on
the disk. It's menu-driven, with help screens, and it has a disable/re-enable
feature. Depending upon your system's memory, you can have a cache size up to
3.2MB. It supports conventional memory, Extended memory and LIM EMS 3.2 or
greater.
However, there is what I consider to be a misnamed file on the disk. Any
time you see README.1ST, README.TXT, READ.ME, and any similar file, you know
it's a text file you should read before doing anything. On this disk, there's
a batch file named README.BAT.
Therefore, you might assume you would just type README at the DOS prompt.
But that only brings up the batch file which is used in conjunction with the
cache program. So it's not a good idea naming it README, as it confuses the
issue.
There is also the DISK MANAGER/DISK MANAGER DIAGNOSTICS program from
Ontrack Computer Systems Inc., which is pretty straightforward.
The S386 itself performed well during the evaluation period. However,
there was a BIG problem with the 3 1/2" drive that I discovered right away.
And what happened was a new one for me.
I did a DIRectory on a 3 1/2" disk, took it out and popped in another
disk. After typing DIR again, I could hardly believe what my eyes were telling
me. I saw the SAME directory displayed of the PREVIOUS disk!!! Doing a DIR
repeatedly still gave me the directory of the disk I had REMOVED!
The only solution came when I would go to A: and then back to B: again.
Then I could get a proper directory. Obviously something was wrong.
A call to a computer consultant friend brought the answer. He chuckled,
and asked me if it was a Toshiba drive. I told him it was. He said he had had
the same experience with Toshiba 1.44MB drives in the PCs that he sells.
He told me the drives have logic board problems which cause the weirdness
with the drive's buffer, making it "remember" the previous directory! This
model has been discontinued by Toshiba he added, and a later model is
supposedly OK. However, he's switched to Mitsubishi 3 1/2" drives, and he said
they perform very well.
Therefore, before purchasing the SIIG S386, I would advise buyers to ask
for a Mitsubishi or some other drive that they KNOW is reliable. The
consultant told me I'd probably have an identical problem with the Toshiba
1.2MB 5 1/4" drive. However, it seemed to work fine.
I ran a check on the 3 1/2" drive, using the 386's built-in diagnostics
program. BINGO! Although the proper speed for that drive is 300, a slight
variance is all right, with 294 to 306 as the acceptable limits.
The test showed the drive varying wildly, from 328 to 355, so it was way out of
tolerance.
The system I evaluated had a price tag of $3423. When someone spends that
much money on a system, the drives should be totally reliable. In my view,
every system should be bench-tested before sending it out to a customer. If I
had purchased this 386, I'd have to ship it back for replacement of the 3 1/2"
drive, and I wouldn't be happy about it.
While you cannot judge the overall performance and value of a computer by
one of its components, I'm always a bit nervous when something like this
happens. It makes me wonder what else might be wrong inside the case.
If you're a power user and need that 386, you can still consider the S386
2000/2100 series. But call the company first to see if they have cleared up
that 3 1/2" drive problem.
Contact Suntek Information International Group, 5369 Randall Place,
Fremont, CA 94538; (415) 657-0567; FAX (415) 657-5962.
#
David W. Batterson is a computer journalist/reviewer, business writer and
PR consultant. Contact him via MCI MAIL: DBATTERSON.