Idiotic Article in Radio Electronics...
Daniel J Rubin
or Radio Electronics. The article is a crock and I was hoping that others
out there would agree or disagree and add some input. The article is in the
column Computer Connections, written by Jeff Holtzman titled "Microsoft and
Intel take on the world." and it is on page 96. The article basically states
that "Mega-PCs" are going to dominate the computer market in the near future.
I am not sure what "Mega-PCs" refers to because the article is written so
poorly. The article states also that UNIX is, "like the horror-flick creature
that just won't die...Despite its tenacity, UNIX never comes close to
achieving the dominance to which it aspires. ". Further on in the article he
states that UNIX is only good for a few circumstances, "The UNIX market has
consistently overrated its importance in the overall scheme of things. Its
strongholds include CAD, manufacturing, financial analysis, and, thanks to
Next, multimedia development.". The main problem I have with the idea this
person is trying to convey is that he does not mention how powerful the UNIX
environment is for developing and testing software and documentation. I would
always choose an UNIX environment for this type of work over anything from
Microsoft and I believe other major companies are also...
PS. I was going to email a letter to Jeff Holtzman, but of course the man who
knows so much about what is going on with computers does not have an email
address or at least he does not have it listing in the TCP/IP Internet user
name directory service.
- Dan
--
Daniel Joseph Rubin ru...@cis.ohio-state.edu
GO BENGALS! GO BUCKS!
Dave Mc Mahan
>I just finished reading an article in the April 1993 issue of Electronics Now
>or Radio Electronics. The article is a crock and I was hoping that others
>out there would agree or disagree and add some input. The article is in the
>column Computer Connections, written by Jeff Holtzman titled "Microsoft and
>Intel take on the world." and it is on page 96. The article basically states
>that "Mega-PCs" are going to dominate the computer market in the near future.
Take another look at the date on the front of the magazine. Notice that it
says 'April'. Do you think the article could be written as a silly attempt
at an april fool's pun? Many magazines seem to think that is worth wasting
print on.
>The main problem I have with the idea this
>person is trying to convey is that he does not mention how powerful the UNIX
>environment is for developing and testing software and documentation. I would
>always choose an UNIX environment for this type of work over anything from
>Microsoft and I believe other major companies are also...
It seems that there will forever be the makings of a religious war over
this topic. You have your opinion (I tend to agree with it in most cases
but not all). Keep in mind that there are lots more folks who are computer
phobic than computer-compatible. Unix tends to be no fun for those who
don't want to learn. Likewise, the MS-DOS computer has seemingly become much
easier to use with the advent of Windows 3.0+. If you were writing
magazine articles for the masses and wanted to make money, where would you
side? If MS-DOS computer are relatively cheap, easy to get, easy to use,
and powerful, what would you buy if you just wanted to get the job done?
If you price equivalent software for things like word processing,
spreadsheets, or databases, where would you park your dollars?
These and other points can be argued forever with no clear winner in sight.
It is only in hind-sight that we realize how blind we used to be.
>PS. I was going to email a letter to Jeff Holtzman, but of course the man who
>knows so much about what is going on with computers does not have an email
>address or at least he does not have it listing in the TCP/IP Internet user
>name directory service.
Perhaps he has a Compu$erve, MCI-mail, BIX or Prodigy account? These also
count as e-mail addresses.
> Daniel Joseph Rubin ru...@cis.ohio-state.edu
-dave
--
Dave McMahan mcm...@netcom.com
Jim Campbell
>I just finished reading an article in the April 1993 issue of Electronics Now
>or Radio Electronics. The article is a crock and I was hoping that others
>out there would agree or disagree and add some input. The article is in the
>column Computer Connections, written by Jeff Holtzman titled "Microsoft and
>Intel take on the world." and it is on page 96. The article basically states
>that "Mega-PCs" are going to dominate the computer market in the near future.
[deleted stuff]...
I haven't read RE in a while because it arrives in our supermarkets
in a plastic bag, and since I wouldn't buy a car without a test drive, I find
myself not getting the magazine (I'm not allowed to examine it before I buy).
However, could this maybe be an "April fools" article? RE has done that
before. The last one I remember was something about a "Poor Man's Laser
Printer", or some such, and involved strapping a monitor to a xerox machine!
I know I may be on thin ice here, suggesting that it might be a spoof
when I haven't even read the article. But I thought I'd throw that in anyhow.
regards,
--
Jim Campbell | "Remember to tweet!"
ji...@megatek.com | Is it my imagination, or am
WB6ZPB NSS#36691 ASA TNS | I just imagining things?
Daniel J Rubin
>says 'April'. Do you think the article could be written as a silly attempt
>at an april fool's pun? Many magazines seem to think that is worth wasting
>print on.
>
Well actually I did not think of this at the time of the posting, but after
about 20 replies saying that it must be an April fools joke I thought about
it. I went back and read the article again and decided that it really was
not "off the wall" enough to be an April fools joke. Take a look at it and
I think you will agree. Also I remember some of the outlandish April fools
article in Radio Electronics in the past and I do not think this one fits...
On thing though, every time I read the stupid article I get more and more set
on my opinion that the author's knowledge of computers is lacking diversity
and has limited scope...
Bob Myers
> when I haven't even read the article. But I thought I'd throw that in anyhow.
Sure, it might be a spoof. That, of course, has nothing to do with the
fact the Unix IS a thing of the devil.
Bob Myers | "One man's theology is another man's belly laugh."
my...@fc.hp.com | - Lazarus Long/Robert A. Heinlein
|
Tad Cook
>or Radio Electronics. The article is a crock and I was hoping that others
>
>PS. I was going to email a letter to Jeff Holtzman, but of course the man who
>knows so much about what is going on with computers does not have an email
>address or at least he does not have it listing in the TCP/IP Internet user
>name directory service.
>
Jeff's email address is 360...@mcimail.com.
--
Tad Cook | Phone: 206-527-4089 (home) | MCI Mail: 3288544
Seattle, WA | Packet: KT7H @ N7DUO.WA.USA.NA | 328...@mcimail.com
| Internet: t...@ssc.com or...sumax!ole!ssc!tad
Doug Wellington USGS
While UNIX may have some good places to be, it still is
not a very good _production_ environment. I am convinced
that the only reason UNIX is so big is that it was FREE.
Schools that teach computer science love it, because they
are worried about budgets, and want anything that comes
with the source code, so a whole generation of computer
scientists has only been taught UNIX. Since we are all
subject to "The Baby Duck Syndrome", we think that the
first half-way capable thing that we run into is the best.
Then, with the current trend in holding down prices, all
of these hardware manufacturers have to cut everything to
the bone, and since UNIX doesn't cost much, and is relatively
portable, they don't have to invest much in software development
and therefore can remain competitive in a tough market. Face
it, nobody wants to pay for operating systems now, they just
want the latest, greatest, fastest hardware for the cheapest
price. And cheap means the only thing hardware manufacturers
can afford is UNIX.
Now, truthfully, how many of you UNIX bigots have ever REALLY
learned any other operating system? (Besides Macs! -which
aren't real operating systems anyway...)
Let the flames begin!
-Doug Wellington
Who IS a UNIX systems administrator and has also worked
in production shops with other operating systems.
Frank - Hardware Hacker - Borger
>Hey folks, I think your religion is showing! ;-)
>
>While UNIX may have some good places to be, it still is
>not a very good _production_ environment. I am convinced
>that the only reason UNIX is so big is that it was FREE.
>
The system and source code were free, but so was the support. You got
it for free so you had to support it. Not my idea of fun. (And yes,
I do and have done a lot of system level coding, I've written handlers
etc, so I know somewhat of what I speak.)
It is a religion. Steeped in its own arcane language known only to
the chosen elite. Containing mystic symbols and animals, (CATs and
DOGs if you remember BIFF was a dog.)
It's also much like the Bible. Although many consider the Bible to be
one book, it is actually a collection of many short books, written
by many different people, having many different views, (not all of
them agreeing.)
DGMW. All systems have their strengths and weaknesses, but UNIX is the
only one I know of whose practicioners have such a holy grail quality.
Frank R. Borger - Physicist ___ George Halas had two heirs; one carried
Michael Reese - U of Chicago |___ his seed and one carried his soul. ...
Center for Radiation Therapy | |_) _ One can imagine the whisper of George
net: Fr...@rover.uchicago.edu | \|_) Halas, rooting for his soul. -
ph: 312-791-8075 fa: 567-7455 |_) Jeannie Morris
Henry Spencer
>Then, with the current trend in holding down prices, all
>of these hardware manufacturers have to cut everything to
You haven't priced commercial Unix source licenses lately, I see. Unix
was cheap to universities; it's not cheap to commercial users except in
binary-only form.
The main reason why so many companies -- including ones with major
investments in their own operating systems -- are supporting Unix is
that their customers have made it clear that they want Unix.
--
All work is one man's work. | Henry Spencer @ U of Toronto Zoology
- Kipling | he...@zoo.toronto.edu utzoo!henry
Scott Dorsey
>Hey folks, I think your religion is showing! ;-)
>
>While UNIX may have some good places to be, it still is
>not a very good _production_ environment. I am convinced
>that the only reason UNIX is so big is that it was FREE.
That is an excellent reason for it to be big. What's more, now that there
is a free BSD system completely devoid of AT&T code available for free to
anyone through uunet.uu.net, it's an excellent reason for it to become even
bigger.
It's not the best possible OS design that I can think of, but it's relatively
simple and it's free. Plenty of software is available for it. It's already
been ported to plenty of boxes. I think I'll remain a member of this religion.
--scott
Ken Hornstein
>In article <24MAR199...@penny.telcom.arizona.edu>, do...@penny.telcom.arizona.edu (Doug Wellington USGS) writes...
>>Hey folks, I think your religion is showing! ;-)
>>
>>While UNIX may have some good places to be, it still is
>>not a very good _production_ environment. I am convinced
>>that the only reason UNIX is so big is that it was FREE.
>>
> TANSTAAFL
>
> The system and source code were free, but so was the support. You got
> it for free so you had to support it. Not my idea of fun. (And yes,
> I do and have done a lot of system level coding, I've written handlers
> etc, so I know somewhat of what I speak.)
I dunno, I think it _is_ fun ... but that's just me :-)
I don't think sci.electronics is the appropriate place to get into a "my
OS can beat up your OS" flame war, but let me add my $0.02:
My first experience with a multiuser operating system was VM/CMS. I used
that for about 3 years, and I feel that I know it pretty well. In the last
two years, I've converted over to Unix. I've also used VMS on and off during
that time, and I now manage a VMS and Unix workstation cluster.
IMHO, I feel that Unix is far superior to VMS and VM/CMS. Of course, the case
could be made that this wasn't much competition :-) But what operating systems
are out there? MS-DOS? A glorified bootstrap loader. Windows? A graphical
front-end to a bootstrap loader. I can't think of any other operating systems
that could give Unix a run for it's money. Sure, now we have Plan 9, Windows
NT, but Unix has a strong installed base.
--Ken
Larry D. Pyeatt
|> Hey folks, I think your religion is showing! ;-)
|>
|> While UNIX may have some good places to be, it still is
|> not a very good _production_ environment. I am convinced
|> that the only reason UNIX is so big is that it was FREE.
HUH?! UNIX is not free. In the last couple of years there
have been a few UNIX-like Operating systems becoming availabe,
but if you want the real thing, get ready to pay a minimum of
$1000. The reason UNIX is so big is because it is so good.
It costs 10 times as much as DOS and gives 10 times as much
power, with 1/10 the headaches.
|> Schools that teach computer science love it, because they
|> are worried about budgets,
Completely wrong. The like it because it is very good for
experimentation and rapid software development. Also, it
is modular, elegant, and powerful.
|> and want anything that comes
|> with the source code,
just try getting the source code for SunOS, Solaris, AIX, or
any other popular version of UNIX. True, BSD386 and Linux
provide free source code, but they represent a small minority
of UNIX machines.
|> so a whole generation of computer
|> scientists has only been taught UNIX.
During my undergrad days, I learned to use a clunky old IBM 3091,
a VAX running VMS, a CDC with a keypunch, IBM-PathetiComputer,
Macintoy, CP/M, and an Apple II with AppleSloth. I never
experienced real POWER until I began using UNIX as a GRADUATE
student. You'll have to pry it out of my dead fingers. ( okay,
so I'm a zealot )
|> Since we are all
|> subject to "The Baby Duck Syndrome", we think that the
|> first half-way capable thing that we run into is the best.
No, I am the first to say that UNIX has some faults. However,
I think that I have had enough experience to make an informed
choice. Until something new and better comes along, I will
stick with the best OS I have ever used -- UNIX. I must admit
that in many ways, VAX VMS is superior, but the VAXen are
so slow that it is not worth the effort of maintaining one.
I also like NextStep.
|> Then, with the current trend in holding down prices, all
|> of these hardware manufacturers have to cut everything to
|> the bone, and since UNIX doesn't cost much,
$1000 is a lot more than DOS, but a lot less than IBM's sorry
mainframe setup, and quite a bit less than DEC's VMS.
|> and is relatively
|> portable, they don't have to invest much in software development
|> and therefore can remain competitive in a tough market. Face
|> it, nobody wants to pay for operating systems now, they just
|> want the latest, greatest, fastest hardware for the cheapest
|> price. And cheap means the only thing hardware manufacturers
|> can afford is UNIX.
What else are they going to use? There really is no useable
alternative. OpenSystems is very important in the marketplace
right now, and the most open OS is UNIX.
|> Now, truthfully, how many of you UNIX bigots have ever REALLY
|> learned any other operating system? (Besides Macs! -which
|> aren't real operating systems anyway...)
The same can be said of MS-DOS,
See above.
--
Larry D. Pyeatt The views expressed here are not
Internet : pye...@texaco.com those of my employer or of anyone
Voice : (713) 975-4056 that I know of with the possible
exception of myself.
Neal Howard
>In article <25MAR199...@rover.uchicago.edu> fr...@rover.uchicago.edu writes:
>>>Hey folks, I think your religion is showing! ;-)
>>>
>>>While UNIX may have some good places to be, it still is
>>>not a very good _production_ environment. I am convinced
>>>that the only reason UNIX is so big is that it was FREE.
>
>OS can beat up your OS" flame war, but let me add my $0.02:
That's right but since we're mostly all 'puter geeks here, then it's as good
of a place as any.
>
>IMHO, I feel that Unix is far superior to VMS and VM/CMS. Of course, the case
>could be made that this wasn't much competition :-) But what operating systems
>are out there? MS-DOS? A glorified bootstrap loader. Windows? A graphical
MPE/iX on the HP3000/9xx series minicomputers is quite an elegant O/S geared
towards "transaction processing based" applications. It took me less than a
month from never having ever heard of this system before, to becoming totally
proficient (to a demi-guru level) in this environment. It's a fantastic system
for such mundane things as accounting, inventory control, customer billing,
etc. sorts of applications and has an extremely fast hashed DBMS (Turbo Image)
package built in. It has a really neat method of file access called "mapped
files" in which opening a file gives you a pointer into memory so you can
simply treat your file as if it were a simple array already loaded into memory
and don't have to worry about bothering with regular file reads & writes. All
files are memory cached by the hardware anyway so this saves you one layer of
O/S to go through to get to your files. The native directory structure is only
a flat, one layer directory structure which is ultra easy to navigate thru (
although with MPE/iX 4.5, an upcoming release, we're getting a full tree
directory AND the old flat directory concurrently in the same O/S with a POSIX
C.I. shell too). Don't compare MPE/iX too closely to the old MPE/V system for it
is light years ahead in capabilities for the applications developer and end user
(yes, I use lotsa foul language when working with MPE/V too (-: ). The only
other gripe I have about MPE is the lack of true full duplex serial I/O
when using a terminal port as a "device file" (from an applications program's
perspective) although this is supposed be
one of the enhancements in upcoming versions of MPE/iX.
Disclaimer: I don't work for HP or represent them in any way, I just use the
heck out of HP3000's and HP9000's almost every day and get a lot of good work
done on these machines ;-)
--
=============================================================================
Neal Howard '91 XLH-1200 DoD #686 CompuTrac, Inc (Richardson, TX)
doh #0000001200 |355o33| ne...@cmptrc.lonestar.org
Std disclaimer: My opinions are mine, not CompuTrac's.
"Let us learn to dream, gentlemen, and then perhaps
we shall learn the truth." -- August Kekule' (1890)
=============================================================================
Joseph Chiu
I suggest you look at OS/2 2.0 as a possible platform on the "PC" architecture.
It has at least two of the great attributes of UNIX: true multi-tasking, and
virtual memory. This OS will allow you to compile programs that are not
limited by DOS's 64K-at-a-time segmented architecture.
No, it's not free like UNIX (or, rather, 386bsd and Linux for the PC's), but
for the price ($90), it includes DOS and Windows support as well as the
OS/2 operating system and it's GUI: WPS. It also has a fair amount of on-line
documentation to get you started.
--
Joseph Chiu | jos...@cco.caltech.edu "OS/2: You gotta get this thing!"
MSC 380 - Caltech |
Pasadena, CA 91126 | Turn your ordinary PC into a personal workstation...
+1 818 449 5457 | OS/2 2.0 : Giving you the power to move ahead...
Ken Hornstein
>I suggest you look at OS/2 2.0 as a possible platform on the "PC" architecture.
>It has at least two of the great attributes of UNIX: true multi-tasking, and
>virtual memory. This OS will allow you to compile programs that are not
>limited by DOS's 64K-at-a-time segmented architecture.
I've heard mixed things about OS/2. The one copy of it I saw running on a 486
ran like a dog (made a Sun 3 running Xkernel look really spiffy).
I don't think I want to be tied down to the PC artitecture, tho. For a while
I've been toying with the idea of desinging my own computer based on the 68030
or 68040, and porting Unix to it. Sure, it's not going to be cheap or easy,
but I'd think it be a lot of fun to do. All I need is a real job, $20K-$30K
of test equipment (minimum), and I'd be ready to go :-)
--Ken
John Whitmore
>In article <24MAR199...@penny.telcom.arizona.edu>, do...@penny.telcom.arizona.edu (Doug Wellington USGS) writes:
>HUH?! UNIX is not free. ...get ready to pay a minimum of
>$1000.
>$1000 is a lot more than DOS, but a lot less than IBM's sorry
>mainframe setup, and quite a bit less than DEC's VMS.
It's possible to buy VMS in many forms, but unless you
have a really LARGE computer, you'd probably be getting a
two-user license for VMS, and the current cost of that is $600.
Four-user license is $1200. This is somewhat complicated
by the 'base license' that isn't priced separately but
comes with the computer, and by the extra cost for the
full documentation set ($3600). Depending on how many
users a typical UNIX machine serves, the prices overlap.
As I understand it, the license limits simultaneous
logged-in users (though there is no limit to batch and
background processes) and is enforced in software.
Parenthetically, Mr. Pyeatt is correct in noting that VMS
has not run on competitive hardware, but with the Alpha AXP
machines coming out now, that is changing. DEC is improving
their UNIX as well, so they plan to support EVERY popular
system (including NT and DOS, alas).
John Whitmore
Ken Hornstein
> It's possible to buy VMS in many forms, but unless you
>have a really LARGE computer, you'd probably be getting a
>two-user license for VMS, and the current cost of that is $600.
>Four-user license is $1200. This is somewhat complicated
>by the 'base license' that isn't priced separately but
>comes with the computer, and by the extra cost for the
>full documentation set ($3600). Depending on how many
>users a typical UNIX machine serves, the prices overlap.
Of course, when you buy VMS, you get just that ... VMS, nothing else. No
compilers, no networking, no nothing. All you can do is log on and send
VMS MAIL to your buddies on the same system. But since you can only use
directly connected terminals, you probably won't need more than a two-user
license. But don't you have to renew those licenses each year?
In contrast, with BSD 4.3, you get:
Built-in networking (TCP/IP, not DECnet)
Full set of software development tools (C & fortran compilers and debuggers)
The source code to the OS, so you can add other neat features.
An equivalant VMS system would cost much more than $600.
I'm not going to get into the "Which is better?" argument, but here's something
interesting: VMS running a VAX 11/750 was completely unbearable to use with
more than one person on it. The same 11/750 with BSD 4.3 now can support 7-8
people all reading news simultaneously quite nicely.
--Ken
Daniel J Rubin
>have been a few UNIX-like Operating systems becoming availabe,
>but if you want the real thing, get ready to pay a minimum of
>$1000. The reason UNIX is so big is because it is so good.
>It costs 10 times as much as DOS and gives 10 times as much
>power, with 1/10 the headaches.
I agree totally... well almost totally, you can buy UNIX for PC compatibles
for a couple hundred or ftp memory and diskspace efficient semi-scaled-down
versions from almost any ftp site that has PC stuff.
As for the power, it is overwhelming. I got started in computing on an IBM
8088 running MSDOS many years ago and I though it was the greatest thing ever,
but even DOS 5.0 and the whole philosophy around MSDOS cannot compare to UNIX
because it lacks that ability to easily manipulate the system AND NETWORK
like UNIX.
You can whip up a shell script to do just about everything on a UNIX machine
or right a program using all the UNIX tools available to accomplish tasks on a
single machine or hundreds in a network. The idea of solving a large problem
in chunks on many machines is easily implemented at a UNIX site. Many Sun
workstations working together on parts of a complex problem can solve it
faster than a single super computer can in some cases. Need more power add
more workstations. The bigger the network the more powerful the site is
which makes expansion easy unlike mainframes etc... I think this is one of
the main reasons UNIX is so popular right now and the reason many companies
are going that route. When I was working as a intern I witnessed the company
switch over to a Sun network for development and roof test their IBM mainframes
so to speak.
>During my undergrad days, I learned to use a clunky old IBM 3091,
>a VAX running VMS, a CDC with a keypunch, IBM-PathetiComputer,
>Macintoy, CP/M, and an Apple II with AppleSloth. I never
>experienced real POWER until I began using UNIX as a GRADUATE
>student. You'll have to pry it out of my dead fingers. ( okay,
>so I'm a zealot )
I think many people out there are reluctant to dig into UNIX and discover
the power of it. Some people just don't like change which I think is a
shame because if you think about it there will always be something better
to replace the things you are using now...
oep
: environment is for developing and testing software and documentation. I would
: always choose an UNIX environment for this type of work over anything from
: Microsoft and I believe other major companies are also...
The problem is ofcourse that the general public don't want a userunfriendly
machine like unix. They want a box to put on their desk to do exactly the tasks
it's suppose to do, to a reasonable price. Thats why unix never won the computer
war. ( UNIX does a lot of things you don't want to do, and you need a
computer engineer to get it to run properly, things that you unix-gurus love
to waste your life to do)
I do develop software, and i do like UNIX, and all the source-code/freeware
that comes with it, but i don't have one on my desk because I can't afford a
personal Sparc or next. Another problem os ofcourse that users of my software
don't like to setup a unix, just to use my software. (Which I would have had
to cross-compile i n versions to cover all platforms)
Lets just hope that someone makes a bug-free unix with a mac-like interface,
that run all unix-software, burn it into ROM and sell a complete system for
the price of a PC-clone :->
-oep
Bill Mayhew
IIcx. It really is a pretty nice implementation of the UNIX
environment. It is big and not terribly fast. None the less, it
is pretty easy to install, easy to administer and comes with very
good start-up manuals for neophytes. I was quite impressed with
A/UX's flexibility I was able to have someone log in at the console
and run the Macintosh (non A/UX) based game, Solarian, that
basically takes over the entire machine, yet still run an
interactive ksh shell session over a CSL/IP link to Chameleon
running on my DOS machine. I'm not often an Apple fan, but I have
to give them credit for doing a very nice job here.
I like A/UX better than either OS/2 or Solaris 2.0. My own
personal machine is an IPC 4/40 with SunOS 4.1.2. I'm curious to
see what comes of NeXT's work for the '486 hardware.
I've used most of the big iron operating systems over the last 20
or so years. I have to admit that as a technical user that I much
prefer UNIX to VM or VMS, etc. For neophyte users, it doesn't make
too much difference what's running on the other side of the CRT as
long as it is easy to use. Unfortunately, none of the above are
very user friendly.
The basic Macintosh user interface is about the most user friendly
interface around; I don't like it, however, because it is too
successful at hiding the complexity of the machine from me. For an
occasional user, however, it is great stuff. I prefer Windows for
low end desktop use at the moment becuase it satisfies my wishes
for configurability and availabilty of tools. I've used OS/2 and
have been disappointed by OS/2's voracious appetite for resources.
OS/2 does have an awful lot of neat features and is very flexible.
I wish that IBM would change a number of things on the workplace
shell. The property sheets and the like are neat for people like
me, but rather confusing for novice users. The workplace shell
suffers from mixed metaphor problems in some areas, but not as much
as Windows. Windows NT sounds like it is going to have many of the
problem attributes of OS/2 when it finally makes it market.
Windows NT is going to have a lot of neat capabilities for
technical users, but really isn't going to be very appealing for
novices. I keep hearing conflicting information about what
features will be in NT's final release. The last I heard, TCP/IP
may be out and mail services will be limited. I'm curious to see
what's going to be in the final of the upcoming OS/2. The OS/2
beta CDROM has lots of neat stuff.
In summary, there really is no best operating system or control
environment. Much depends on the task one desires to accomplish.
UNIX fans that like the appearance of the Macintosh environment
should definitely take a look at A/UX 3.0.
--
Bill Mayhew NEOUCOM Computer Services Department
Rootstown, OH 44272-9995 USA phone: 216-325-2511
w...@uhura.neoucom.edu (140.220.1.1) 146.580: N8WED
Dave Martindale
>should definitely take a look at A/UX 3.0.
Now if Apple only had some decent hardware to run it on. I work in Unix
all the time and like it. I also like Macs for the things they do well.
But Apple *hardware* is pretty unimpressive compared to current cheap
Unix workstations from anyone else. I hope the new Apple RISC machines will
be out soon.
Larry D. Pyeatt
I agree, but I really don't expect Apple to go all out on fast hardware.
For years they have been building machines with the cheapest processors
they can get. The 68040 is a pretty speedy processor, but Apple uses a
cheap version with a low clock rate. The problem is compounded by the
fact that the CPU is used to do things which could easily be handled by
I/O processors. Are they just cheapskates?
Bruce Cheney
>In article <1993Apr1.0...@imax.imax.com>, da...@imax.imax.com (Dave Martindale) writes:
>|> Now if Apple only had some decent hardware to run it on. I work in Unix
>|> all the time and like it. I also like Macs for the things they do well.
>|> But Apple *hardware* is pretty unimpressive compared to current cheap
This doesn't match my experience at all. I use UNIX, DOS, MAC, and Windows,
and in fact have one machine of each on my desk. Uniformly MAC's outperform
DOS machines when you compare the nearest equivalent microprocessors at the
same speed. Since they are priced about the same, the MACs are a much better
deal.
On top of that, our MACs have outlasted 3 generations of PC's. The guy in
the cubicle next to me has purchased 3 PC's and 4 operating systems (DOS upgrades and Windows) to get the capability and performance I purchased 5 years ago.
On the hardware side, since the MACs have a closed architecture and the interfaces are truly standards, adding peripherals and enhancements is very easy.
On the DOS side, for example, I tried to upgrade the display to an NEC 4FG and
and ATI Mach 32 card. Two weeks later after 3 pc experts have worked on it, it
still doesn't work.
Bruce Cheney
Dave Martindale
>
>This doesn't match my experience at all. I use UNIX, DOS, MAC, and Windows,
>and in fact have one machine of each on my desk. Uniformly MAC's outperform
>DOS machines when you compare the nearest equivalent microprocessors at the
>same speed. Since they are priced about the same, the MACs are a much better
>deal.
Who said anything about PCs? The article of mine that you quoted
(which you cut off in mid-sentence) was comparing Mac Hardware to
other manufacturers' low-end Unix workstations *as a platform for running
Unix*. If you compare the fastest Mac with an R4000 Indigo, or a cheap
HP Snake, or a Sparcstation 10, it's pretty sadly lacking in performance.
Maybe a PC with a fast 486 is faster than the fastest Mac, maybe not.
I don't know. But it's not competitive with the SGI/HP/Sun machines
listed above.
>On top of that, our MACs have outlasted 3 generations of PC's. The guy in
>the cubicle next to me has purchased 3 PC's and 4 operating systems
>(DOS upgrades and Windows) to get the capability and performance I
>purchased 5 years ago.
You're talking about Macs running MacOS and PC's running DOS/windows/whatever.
We're talking about platforms for Unix. Why are you joining this discussion?
>On the hardware side, since the MACs have a closed architecture and the interfaces are truly standards, adding peripherals and enhancements is very easy.
>On the DOS side, for example, I tried to upgrade the display to an NEC 4FG and
>and ATI Mach 32 card. Two weeks later after 3 pc experts have worked on it, it
>still doesn't work.
Didn't you even read the articles you quoted? Someone said that the Mac
made a good Unix box. I was criticizing the Mac hardware's speed relative
to other Unix boxes. Then you come to the Mac's defence by pointing out how
its hardware is better than the PC? Nobody even mentioned the PC until
you!
Dave
John Whitmore
>|> But Apple *hardware* is pretty unimpressive compared to current cheap
>|> Unix workstations from anyone else. I hope the new Apple RISC machines will
>|> be out soon.
>I agree, but I really don't expect Apple to go all out on fast hardware.
>For years they have been building machines with the cheapest processors
>they can get.
They've used 68000, 68HC000, 68020, 68030, 68040; they've
used ALL the processors that they could get that run their
core software (the 68010 and Signetics 68070 and 68300 that
they HAVEN'T used aren't particularly pricey).
> The 68040 is a pretty speedy processor, but Apple uses a
>cheap version with a low clock rate.
Motorola selects the fastest chips from the assembly line
to get the highest clock rates; they probably WON'T guarantee
delivery of those chips in the volume Apple wants. Apple simply
can't buy the fastest CPU chips in volume with guaranteed delivery
dates (and this is presumably more important in the purchase
decision than the price).
>The problem is compounded by the
>fact that the CPU is used to do things which could easily be handled by
>I/O processors. Are they just cheapskates?
Only for a multiuser machine is I/O bandwidth an important
bottleneck; single-user use of a PC-like box is the main application
that Apple is concerned with, and the I/O is good enough for that.
Heck, it's a lot BETTER than good enough...
Various third-party suppliers sell smart boards for the
Nubus machines: I haven't noticed any particular popularity for
any of these.
John Whitmore