[comp.sys.amiga] NeXT vs. Amiga ad infinitum.

WHE46@CCVAX.IASTATE.EDU (Marc Barrett) (12/08/90)

   All this incessant quibling in comp.sys.amiga and alt.religion.computers
reminds me of two people fighting in a burning house.  The point about 
which system is better between the NeXTStation and the Amiga 3000UX is
moot because neither system is likely to succeed in the workstation market.

   For Iowa State's Project Vincent research grant, ISU is getting a large
number of DECStation 2100 systems from DEC for approximately $3000 each.
These are RISC systems with approximately the same overall performance
as a 25Mhz 68030 system, and include a 104M hard drive, 12M of RAM, 19"
1280x1024 monochrome monitor, ethernet connectivity, and (of course) ULTRIX
and DECWindows (derivitives of UNIX, X-Windows, and Display PostScript).

   There is no way that either Commodore or NeXT are going to succeed 
against systems like these from a company like DEC.  For one thing, the
strength of the company behind a product is VERY important to most
edicational institutions and corporations, and the fact that DEC is
stronger (by far) than both NeXT and Commodore combined is enough to
tip most of these types of buyers in favor of DEC.  
   
   Even if NeXT and Commodore priced their systems less than the systems
from DEC, most buyers would still opt for DEC because of the very reasons
I stated above.  NeXT Inc. is a very unstable company, and may not be
around a year from now.  And although Commodore is a stronger company 
than NeXT, Commodore has a very unstable management that could completely
pull out of the workstation market at any given moment.  Combine these
with the higher cost of the systems from both companies, and you have a
formula for disaster.  This can be verified by DEC's proven track record
in the workstation market.

   
                                     -MB-

   

greg@ccwf.cc.utexas.edu (Greg Harp) (12/08/90)

In article <F2CC11F177FF004282@ISUVAX.BITNET> WHE46@CCVAX.IASTATE.EDU 
  (Marc Barrett) writes:
>   For Iowa State's Project Vincent research grant, ISU is getting a large
>number of DECStation 2100 systems from DEC for approximately $3000 each.
>These are RISC systems with approximately the same overall performance
                                               ^^^^ ^^^^^^^ ^^^^^^^^^^^
>as a 25Mhz 68030 system, and include a 104M hard drive, 12M of RAM, 19"
 ^^ ^ ^^^^^ ^^^^^ ^^^^^^
>1280x1024 monochrome monitor, ethernet connectivity, and (of course) ULTRIX
>and DECWindows (derivitives of UNIX, X-Windows, and Display PostScript).

>                                     -MB-

Aigh!  The end is near!  With these new "powerful" DEC's out, NO Amigas or
NeXT's will EVER sell!  Why, I might as well throw my Amiga out the window,
for all the support I'll be getting.  Oh...Woe-is-me....Woe-is-me...

Just when I was going to buy CBM stock, too...  :-)

Marc, I hope you don't think that machines like this new DEC are going to 
revolutionize the workstation market.  A RISC box with the "same overall 
performance of a 25Mhz 68030 system" is disgracefully slow by today's
standards...

-- 
       Greg Harp       |"How I wish, how I wish you were here.  We're just two
greg@ccwf.cc.utexas.edu|lost souls swimming in a fishbowl, year after year,
Amiga: (n.) multimedia |running over the same ground.  What have we found?
 "Yes, but NeXT what?" |The same old fears.  Wish you were here." - Pink Floyd

dsherif@csserv2.ic.sunysb.edu (Darin D Sheriff) (12/08/90)

In article <F2CC11F177FF004282@ISUVAX.BITNET> WHE46@CCVAX.IASTATE.EDU (Marc Barrett) writes:
>
>   All this incessant quibling in comp.sys.amiga and alt.religion.computers
>reminds me of two people fighting in a burning house.  The point about 
>which system is better between the NeXTStation and the Amiga 3000UX is
>moot because neither system is likely to succeed in the workstation market.
>
>   For Iowa State's Project Vincent research grant, ISU is getting a large
>number of DECStation 2100 systems from DEC for approximately $3000 each.
>These are RISC systems with approximately the same overall performance
>as a 25Mhz 68030 system, and include a 104M hard drive, 12M of RAM, 19"
>1280x1024 monochrome monitor, ethernet connectivity, and (of course) ULTRIX
>and DECWindows (derivitives of UNIX, X-Windows, and Display PostScript).
>
>   There is no way that either Commodore or NeXT are going to succeed 
>against systems like these from a company like DEC.  For one thing, the
>strength of the company behind a product is VERY important to most
>edicational institutions and corporations, and the fact that DEC is
>stronger (by far) than both NeXT and Commodore combined is enough to
>tip most of these types of buyers in favor of DEC.  
>   
>   Even if NeXT and Commodore priced their systems less than the systems
>from DEC, most buyers would still opt for DEC because of the very reasons
>I stated above.  NeXT Inc. is a very unstable company, and may not be
>around a year from now.  And although Commodore is a stronger company 
>than NeXT, Commodore has a very unstable management that could completely
>pull out of the workstation market at any given moment.  Combine these
>with the higher cost of the systems from both companies, and you have a
>formula for disaster.  This can be verified by DEC's proven track record
>in the workstation market.
>
>   
>                                     -MB-
                                       ^^^^^
                                       Oh lord!  Here we go again.
-- 
           Darin Sheriff.  Just a College student with an Amiga.
"According to the classical laws of Aerodynamics, it is impossible for a
 bumblebee to fly."           --- DR WHO ---
 Disclaimer:  Wasn't me.

torrie@Neon.Stanford.EDU (Evan James Torrie) (12/08/90)

greg@ccwf.cc.utexas.edu (Greg Harp) writes:

>In article <F2CC11F177FF004282@ISUVAX.BITNET> WHE46@CCVAX.IASTATE.EDU 
>  (Marc Barrett) writes:
>>   For Iowa State's Project Vincent research grant, ISU is getting a large
>>number of DECStation 2100 systems from DEC for approximately $3000 each.
>>These are RISC systems with approximately the same overall performance
>                                               ^^^^ ^^^^^^^ ^^^^^^^^^^^
>>as a 25Mhz 68030 system, and include a 104M hard drive, 12M of RAM, 19"
> ^^ ^ ^^^^^ ^^^^^ ^^^^^^
>Marc, I hope you don't think that machines like this new DEC are going to 
>revolutionize the workstation market.  A RISC box with the "same overall 
>performance of a 25Mhz 68030 system" is disgracefully slow by today's
>standards...

  Actually, the DECStation 2100 has been out for well over a year.
It's just that DEC has dropped the bottom out of the price recently.
The chip in the DECStation 2100 is a 12.5MHz R2000, which is rated at
about 9-10 MIPS ( take MIPS to mean what you will ).  I programmed and
used 2100s for about a year, and in my experience, they are easily two
times as fast as a 25MHz 68030, more so on floating point.


-- 
------------------------------------------------------------------------------
Evan Torrie.  Stanford University, Class of 199?       torrie@cs.stanford.edu   
"I didn't get where I am today without knowing a good deal when I see one,
 Reggie."  "Yes, C.J."

yoo@well.sf.ca.us (Young-Kyu Yoo) (12/09/90)

>NeXT, Inc. is a very unstable company, and may not be around a year from
>now.

It think it is Standard & Poors that gives NeXT its highest rating for
financial stability.  This is because of NeXT's very large capital base.
Canon paid $100 million for 16% of NeXT.  This makes NeXT a $600 million
company.  Couple this with the fact that NeXT only has several hundred
employees and one of the most cost efficient manufacturing plants in the
world (Fortune or Forbes had an article devoted to the NeXT factory alone),
you have a company that will be around for quite a while.  

True may schools are stuck in the Sun/Dec mindset, but NeXT will make inroads
with its new machines.  Iowa State purchased 68030 Decs for $3000 each.
It could have bought 68040 NeXTs for $3000 or less (the educational price of
$3000-$3500 include institutional markups).  Yes, procurement at many 
universities is often a political issue with little pretense to objectivity.
But, the prices, power, and features of the NeXTs are gonna make some 
procurers take notice.

barrett@jhunix.HCF.JHU.EDU (Dan Barrett) (12/11/90)

In article <F2CC11F177FF004282@ISUVAX.BITNET> WHE46@CCVAX.IASTATE.EDU (Marc Barrett) writes:
>ISU is getting ... DECStation 2100 systems from DEC for approximately $3000
>each.  These...include a 104M hard drive, 12M of RAM, 19" 1280x1024
>monochrome monitor, ethernet connectivity, and (of course) ULTRIX and
>DECWindows (derivitives of UNIX, X-Windows, and Display PostScript).

	So go buy one yourself, Marc.  What?  You mean that price is not
available to individuals?  Just institutions?  Oh well.

	But suppose you manage to get one anyway, somehow.  Try to run
DECwindows.  What?  Your workstation is going SWAP SWAP SWAP SWAP so much
that applications are loading SLOW SLOW SLOW?  Didn't DEC tell you that 12
MB is not enough to run DECwindows comfortably?  After all, the ordinary
"dxterm" program is almost THREE MEGABYTES in size.  Open 4 of these
windows, and you have used ALL your available RAM.  DEC charges about $2000
for 4 MB RAM.  Better shop around....

	Suppose you get your RAM and things seems smoother.  Start
developing a program.  What?  There are OS files missing?  Of course:  you
didn't have enough room on your disk!  The OS take up 70-150 MB.  And they
recommend 48 MB swap space (3x available memory).  And you really don't get
104 MB; more like 96 MB once the filesystem is made.

	So, you hook up your DS2100 to a fileserver machine, or buy some more
disk space.  Add more $$$$.  And now you may have to deal with the SLOWNESS
of NFS (network file system) for remote disk access.

	OK... so suppose you get everything the way you like it.  One
day, your monitor blows, or the CPU board goes bad.  Ever calculate the
price of a repair for DEC equipment??  You will go RUNNING back to your
Amiga, full of apology and shame.

	Marc, you just don't know what you are talking about sometimes.
You can't just praise one company and bash another, unless you point
out the good and bad points of BOTH companies.  It just isn't fair.

>...DEC's [stability and] proven track record in the workstation market.

	DEC is not the world's most stable company, either.  First of all,
they have to deal with a large, internal split between the Ultrix and the
VMS people.  You would not believe how much trouble this can cause the end
user.  Don't believe me?  Ask a DEC salesperson why the supplied "troff"
doesn't work under RISC Ultrix 4.0.

	"Proven track record?"  If you want a list of idiotic workstation
design decisions made by DEC, just ask any Ultrix administrator what "DMS"
is. :-)

LACK OF DISCLAIMER:	I am the UNIX systems administrator of a network of
			DECstation 3100's.

DISCLAIMER:		DEC is OK.  They are not the worst company, nor the
			best company.  Like all companies, and all equipment,
			they have their problems.

			But it is stupid to praise one company over another
			without pointing out both company's faults.

                                                 Dan Barrett
						 (No relation!)

 //////////////////////////////////////\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
| Dan Barrett, Department of Computer Science      Johns Hopkins University |
| INTERNET:   barrett@cs.jhu.edu           |                                |
| COMPUSERVE: >internet:barrett@cs.jhu.edu | UUCP:   barrett@jhunix.UUCP    |
 \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\/////////////////////////////////////

mwm@raven.relay.pa.dec.com (Mike (My Watch Has Windows) Meyer) (12/11/90)

In article <22056@well.sf.ca.us> yoo@well.sf.ca.us (Young-Kyu Yoo) writes:
   True may schools are stuck in the Sun/Dec mindset, but NeXT will make
   inroads with its new machines.  Iowa State purchased 68030 Decs for $3000
   each.

Sorry, but you misread the post, Yoo. ISU didn't by "68030 DECs"; it
bought R2000 DECs. The poster claimed they were in the same
performance range as a 25MHz 68030.

While this may be true for some applications, it's not true in
general. The R2000 at 12.5 MHz is somewhat below the low end of the
quoted specs for the 68040 (I have as yet to see hard numbers from
people actually using an '040; the quotes from Moto & others vary from
sub-14MIPS to 24+MIPS), but faster than any '030 implementation you're
ever liable to run into.

Note that, unlike the Next/Amiga comparisons, you're comparing the
_low_ end of the DEC line with the NeXT. The 2100 is the slowest
DECStation sold. Options from there include the 3100 (16MHz R2000),
and the 5000/200 (25 MHz R3000) with various graphics hacks. The R3000
@25MHz is at the _high_ end of the quoted specs for the '040 - and
systems with that chip (second sourced, even) in them have been
shipping for at least six months. The DECStations 5000/200 is almost
fast enough to do animations as well as the Amiga.

As a final note, I know of three projects to produce MIPS (Rx000 chip)
architectures with 1-5 ns clock times. Any of those is going to be an
order of magnitude faster than a 25MHz '040. This plus open systems
pretty much spells the death of any CISC architechture in the
workstation market.

	<mike
--

bsyme@cs.strath.ac.uk (Brian J Syme IE88) (12/15/90)

I'll settle for one of each, thanks.

-- 
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
<> Brian Syme            <> Why make things difficult, when with just a     <>
<> bsyme@cs.strath.ac.uk <> little more effort you could make them          <>
<>                       <> impossible.                                     <>
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

p554mve@mpirbn.mpifr-bonn.mpg.de (Michael van Elst) (01/03/91)

In article <40996@ut-emx.uucp> greg@ccwf.cc.utexas.edu (Greg Harp) writes:
>Marc, I hope you don't think that machines like this new DEC are going to 
>revolutionize the workstation market.  A RISC box with the "same overall 
>performance of a 25Mhz 68030 system" is disgracefully slow by today's
>standards...

Except for the DECstation 2100 is about two times as fast as a
25 MHz 68030 and about 4 times as fast as a 25 MHz 68882 for
floating point operations.

Regards,
-- 
Michael van Elst
UUCP:     universe!local-cluster!milky-way!sol!earth!uunet!unido!mpirbn!p554mve
Internet: p554mve@mpirbn.mpifr-bonn.mpg.de
                                "A potential Snark may lurk in every tree."