Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

no more hardware

82 views
Skip to first unread message

John J Kuszewski

unread,
Feb 9, 1993, 11:49:08 AM2/9/93
to
I heard on WTOP radio (Washington, DC) this morning that NeXT is getting
out of the hardware business and trying to sell its factory to Canon.
Their story was based on a San Francisco Chronicle article. Does anyone
have the original?

Observations:

1. So half the observations I made before--that NeXT would dump its own
low-end hardware in favor of NextStep/Intel powered peecee clones--are
right. :-/

2. The stories of non-peecee clone Intel machines are almost certainly
wrong. Think about how much effort it would take to write new device
drivers for some wacko machine that would be as non-standard as the
slabs and cubes.

3. This obviously makes the HP deal that much more important for high-
performance NextStep. Remember that we have some very compute-intensive
software (Renderman) as standard equipment. For it to be used as intended,
which would be a real holy-cow sort of thing, fast CPUs and fast memory
would be needed.

4. How does this affect the struggles between SJ and Canon that were
talked about in InfoWorld and this group a few months ago? Will the
factory be enough to buy off Canon so that NeXT can continue unimpeded?

Predictions? NeXT'll have their own OEMed line of peecee clones, complete
with localbus video and (perhaps) DSP cards, in the near future. If
we're all lucky, they'll also have an HP port done soon.


--
John Kuszewski
jo...@jhunix.hcf.jhu.edu
I'm not an idiot, but I play one on USENET

Michael Brill

unread,
Feb 9, 1993, 1:51:13 PM2/9/93
to
>I heard on WTOP radio (Washington, DC) this morning that NeXT is getting

>out of the hardware business and trying to sell its factory to Canon.
>Their story was based on a San Francisco Chronicle article. Does anyone
>have the original?

The gist of the article title "Next Inc. to Drop Hardware - 300 losing
jobs in strategy shift":

* laying off 300 of 540 employees
* negotiating with Canon to sell manufacturing plant
* 69,300 workstations shipped last year (they meant _as of_ last year)
* analysts state public offering 18+ months away
* Official announcements will be made 2/16/93

Nothing else new, just garden variety "NeXTSTEP is good, hardware is not
so competitive," "NeXT is 2-3 years ahead of their competitors" and let
us not forget "NeXT will have a hard time competing because they're small."

Current equity breakdown:
* Jobs 46%
* Perot 11%
* Canon 17.9%
* Other (100 - 46 - 11 - 17.9)%

May I make a suggestion? Nobody say "I told you so!" Thanks! :)

...Michael Brill (yi...@netcom.com)

--
----

Art Isbell

unread,
Feb 9, 1993, 4:02:46 PM2/9/93
to

Although NeXT's official word on its hardware future is not scheduled for
another week, I'd like to ENCOURAGE NeXT to make the announcement ASAP due to
the widespread leak.

While I agree with the move, I find the timing to be very bad. My client's
first "go live" is scheduled for May on 5 NeXTstations. Our customer isn't
happy about buying discontinued hardware (who would be ?-( But we'll have no
alternative for him in May because NS-Intel will still be in beta. So my
client could lose his first sale because of this which will make his survival
very difficult. NeXT had better have a very convincing announcement VERY SOON.
--

Art Isbell Cubic Solutions
NeXT Registered Developer #745 NeXT software development and consulting
NeXTmail: isb...@cats.UCSC.EDU Voice: (408)335-1154
USmail: 95018-9442 Fax: (408)335-2515

Art Isbell

unread,
Feb 9, 1993, 4:11:04 PM2/9/93
to

The Sun salespeople are already out trying to scare NeXT customers away from
NeXT. We've received 2 calls today, one of them mentioning Bud Tribble (who's
now at Sun). They're trying to convince us to abandon NeXTSTEP and go with
Solaris (boy, do they have a lot of convincing to do :-) But for customers who
have taken a chance to go with NeXT and are nervous about having done so, Sun
may be able to scare them away from NeXT. It's a nasty business out there.

Vin Locke

unread,
Feb 9, 1993, 11:14:15 PM2/9/93
to
>> The Sun salespeople are already out trying to scare NeXT customers
>> away from NeXT. We've received 2 calls today, one of them mentioning
>> Bud Tribble (who's now at Sun). They're trying to convince us to
>> abandon NeXTSTEP and go with Solaris (boy, do they have a lot of
>> convincing to do :-) But for customers who have taken a chance to
>> go with NeXT and are nervous about having done so, Sun may be able
>> to scare them away from NeXT. It's a nasty business out there.
>
>If NeXT succeeds with NeXTSTEP on 486 & Pentium machines, I'd expect people
>to be a lot less worried about going with NeXT then they were before. If
>they also have NeXTSTEP running on HP chips, it will be Sun who should be
>the nervous one. With NeXT running on 486's as the low end and HP-PA at the
>high end, it'll be an awfully attractive system.

Do people realize that as of this "no more hardware" leak,
NeXT no longer has a product? NS 486, oh yeah. er, when did they say it
would be out?

Also, NeXT, to make it as a software company, will have to have software to
sell. I could see some pretty intersting discussion coming up on the
net . . .


--vin
v...@netcom.com
------------------------------------------------------------------------
If productivity can really be measured as the rate at which lines of
code are produced, why not just use a tight loop to spew code as fast
as possible, and send the programmers home?
--Brad Cox


Garance A. Drosehn

unread,
Feb 9, 1993, 6:30:34 PM2/9/93
to
isb...@cats.ucsc.edu (Art Isbell) writes:
> The Sun salespeople are already out trying to scare NeXT customers
> away from NeXT. We've received 2 calls today, one of them mentioning
> Bud Tribble (who's now at Sun). They're trying to convince us to
> abandon NeXTSTEP and go with Solaris (boy, do they have a lot of
> convincing to do :-) But for customers who have taken a chance to
> go with NeXT and are nervous about having done so, Sun may be able
> to scare them away from NeXT. It's a nasty business out there.

If NeXT succeeds with NeXTSTEP on 486 & Pentium machines, I'd expect people

to be a lot less worried about going with NeXT then they were before. If
they also have NeXTSTEP running on HP chips, it will be Sun who should be
the nervous one. With NeXT running on 486's as the low end and HP-PA at the
high end, it'll be an awfully attractive system.

Of course, it is certainly true that it's a nasty business out there...

--
Garance Alistair Drosehn = g...@eclipse.its.rpi.edu
ITS Systems Programmer (handles NeXT-type mail)
Rensselaer Polytechnic Institute; Troy NY USA

Neil Greene

unread,
Feb 10, 1993, 9:58:20 AM2/10/93
to
In article <1l965m...@darkstar.UCSC.EDU> isb...@cats.ucsc.edu (Art
Isbell) writes:
>
>
> Although NeXT's official word on its hardware future is not scheduled
for
> another week, I'd like to ENCOURAGE NeXT to make the announcement ASAP
due to
> the widespread leak.
>
> While I agree with the move, I find the timing to be very bad. My
client's
> first "go live" is scheduled for May on 5 NeXTstations. Our customer
isn't
> happy about buying discontinued hardware (who would be ?-( But we'll
have no
> alternative for him in May because NS-Intel will still be in beta. So
my
> client could lose his first sale because of this which will make his
survival
> very difficult. NeXT had better have a very convincing announcement
VERY SOON.


Well said Art.

It is hard to beat out some of the competing platforms and it will be real
hard to beat them with discontinued hardware and only intel machines to
sell. Which turns us immediately into Gateway/Dell salesperson. Now, of
course I have been talking to Gateway for some time now in anticipation of
any move. As a reseller, it places you in a holding pattern.

--
Neil Greene
President, Kentucky NeXT User Group, Inc.
Email: ne...@kynug.org [NeXTMail]

Robert Stephen Rodgers

unread,
Feb 9, 1993, 7:30:05 PM2/9/93
to
In article <1l96l8...@darkstar.UCSC.EDU> isb...@cats.ucsc.edu (Art Isbell) writes:
>
>
>The Sun salespeople are already out trying to scare NeXT customers away from
>NeXT. We've received 2 calls today, one of them mentioning Bud Tribble (who's
>now at Sun). They're trying to convince us to abandon NeXTSTEP and go with
>Solaris (boy, do they have a lot of convincing to do :-) But for customers who


Out of curiousity, I keep seeing posts regarding the alienation
of users that Solaris is causing. Anybody willing to send me
email detailing why? (Solaris hasn't gotten a lot of press--a
bit of a dark horse--so I have no idea..)


MATIS stephane

unread,
Feb 10, 1993, 8:26:31 AM2/10/93
to
In article <justin.729319693@hp750> jus...@hp750.me.engr.uky.edu (Justin Sullivan) writes:
>
>RUN - don't walk - to your nearest HP representative. Don't let your NeXT
>do it: It crawls like a crippled sea slug. You'll be old and gray by the
>time it gets anything done.

I usually stay out of these "mini" flames, but Justin, most of
us don't do "Computational Fluid Dynamics" or our NeXT.... and trully, a
HP 9000/750 is $$$ more than my lowlly NeXTstataion color.

I'm sure my DSP can crank some horsepower ... but not for what
you have in mind.

To each his own ... and honsetly ... the NeXT ain't a "slug".
>--
>* Justin P. Sullivan * jus...@engr.uky.edu (Internet) *
>* System Administrator * jus...@mik.uky.edu (NeXTMail) *
>* Computational Fluid Dynamics * JUSTIN@UKCC (BITNET) *
>* (606) 257-2268 * ukma!engr.uky.edu!justin (UUCP) *


+---------------------------------+ ___ ___ ___
| Stephane I. Matis | / \_BATLLETECH /\__\ Viva NeXT!
| E-Mail : kgn...@cs.concordia.ca | \___/ \___/ \/__/ NeXTSTEP 3.0!
| "It Just Works..." - Steve Jobs | \___/
+---------------------------------+ Wolfnet Operative & NeXThead

Justin Sullivan

unread,
Feb 9, 1993, 11:48:13 PM2/9/93
to
jp...@macbeth.umd.edu (Robert Stephen Rodgers) writes:

> Out of curiousity, I keep seeing posts regarding the alienation
> of users that Solaris is causing. Anybody willing to send me
> email detailing why? (Solaris hasn't gotten a lot of press--a
> bit of a dark horse--so I have no idea..)

Over in the group of the One True Computing Platform (comp.sys.hp :), a fellow
just commented: "Sun really dropped the ball on that one.." referring to their
abandoning of their most precious creation: SunOS. He mentioned that he
really missed BSD4.3-based stuff, and let's face it.. Jumping from that
arena into SVR4 in less than a year is quite a step.

You're now going to have lots of people (misguided purists, hopeless academics
and lazy scientists) scratching their heads when they get a new SPARCstation
(10?) with Solaris 2.x preinstalled.. The local NeXTadmin spent over an hour
trying to figure out how to make a printing modification on this HP 9000/750.
It wouldn't have taken me ten seconds, but the point is that he was completely
unaccustomed to SVR4. He must have pulled his hair out when he didn't find
'lpd' running. Then he called it 'nonstandard.' <shaking head> Am I to blame
because he's been corrupted by a mangy little workstation? Has it come as a
surprise to everyone that their little black slabs are being orphaned by the
faltering company that created them?

RUN - don't walk - to your nearest HP representative. Don't let your NeXT
do it: It crawls like a crippled sea slug. You'll be old and gray by the
time it gets anything done.

Michael Brill

unread,
Feb 10, 1993, 12:38:53 PM2/10/93
to
[miscellaneous doom and gloom happily deleted]

Apart from the occasional "NeXT should only make X terminals" type comments,
it seems that the overwhelming view of csna has been that NeXT should drop
hardware and concentrate on making NeXTSTEP ubiquitous. Hmmmm... lessee...
NeXT drops hardware and is concentrating on NeXTSTEP. I expected to hear
the sound of champagne corks popping, not whines about the market value of
their slabs! :/

I, for one, am more optimistic than ever (and I've nothing vested in the
whole game).

BTW: from today's SF Chronicle (I don't subscribe, really!):
-----------------------------------------------------------------------
Next Confirms Exit From Hardware
Next Computer Inc., of Redwood City, the computer-workstation company
founded by Apple Computer co-founder Steve Jobs, confirmed a report
in yesterday's Chronicle that it will sell its hardware business to Canon,
Inc., the Japanese electronics company [*not* the Indiana-based poultry
packer!].

Canon owns 17.9 percent of Next after investing $165 million over the past
five years. Under the agreement, Canon will forgive Next's debt from a
$55 million credit line [thank you very much] that it negotiated last year.

Next, which employs 535 people, will lay off 280 workers outright. Fifty-
five works will leave Next but may have an opportunity to join Canon, a
Next spokeswoman said. Next, which will keep 200 workers, expects to
complete the transaction by May.
--------------------------------------------------------------------------

Garance A. Drosehn

unread,
Feb 10, 1993, 3:47:43 PM2/10/93
to
yi...@netcom.com (Michael Brill) writes:
> [miscellaneous doom and gloom happily deleted]
>
> Apart from the occasional "NeXT should only make X terminals" type
> comments, it seems that the overwhelming view of csna has been that
> NeXT should drop hardware and concentrate on making NeXTSTEP
> ubiquitous. Hmmmm... lessee... NeXT drops hardware and is
> concentrating on NeXTSTEP. I expected to hear the sound of
> champagne corks popping, not whines about the market value of
> their slabs! :/
>
> I, for one, am more optimistic than ever (and I've nothing vested
> in the whole game).

While this move is pretty much what many people have pestered NeXT to do,
the timing of it is odd (to say the least). If this happened after NS-Intel
was a real, shipping product (and people were confident that it works as
well or better than NS 3.0 on NeXT hardware), then this move would seem
almost inevitable.

Doing it now when NS-Intel isn't a buyable product yet, and doing it after
this "awful" hardware has just helped to give NeXT a profitable quarter,
seems rather bizzare timing. What, they couldn't stand to have two
profitable quarters in a row or something?

The biggest problem with this is probably not the move itself, but
opportunity it gives to a number of other companies to rush around and
misrepresent what the story is. Reading some of these accounts, one gets
the impression that NeXT was never and could never produce competitive
hardware. That alone is not true. When the 040 NeXTstations were
introduced they were extremely good boxes for the $$$.

Most the articles are long on details on how the hardware is not competitive
(which, at the moment, is true), and then toss in some little blurb saying
"Oh yeah, the operating system is pretty nice". This doesn't bother me
personally, but I can see how the ratio of good news to bad news might make
a number of NeXT corporate buyers a little uneasy.

--
Garance Alistair Drosehn = g...@eclipse.its.rpi.edu
ITS Systems Programmer (handles NeXT-type mail)
Rensselaer Polytechnic Institute; Troy NY USA

(I'm speaking as someone who *almost* bought a NeXTstation color turbo two
weeks ago, but am now resigned to waiting for some Intel or (preferably)
HP-based box to run NeXTstep on)

Ron Pomeroy x(Coop)

unread,
Feb 10, 1993, 11:57:24 AM2/10/93
to
In article <C28Io...@newsflash.concordia.ca> kgn...@cs.concordia.ca (MATIS
stephane) writes:
>>In article <justin.729319693@hp750> jus...@hp750.me.engr.uky.edu (Justin
Sullivan) writes:
>>>
>>>RUN - don't walk - to your nearest HP representative. Don't let your NeXT
>>>do it: It crawls like a crippled sea slug. You'll be old and gray by the
>>>time it gets anything done.
>>
>> I usually stay out of these "mini" flames, but Justin, most of
>>us don't do "Computational Fluid Dynamics" or our NeXT.... and trully, a
>>HP 9000/750 is $$$ more than my lowlly NeXTstataion color.
>>
>> I'm sure my DSP can crank some horsepower ... but not for what
>>you have in mind.
>>
>> To each his own ... and honsetly ... the NeXT ain't a "slug".
>>>--

Justin,

RUN - don't walk - to your nearest CRAY representative. Don't let your HP


do it: It crawls like a crippled sea slug. You'll be old and gray by the
time it gets anything done.

As Einstein said...It's all relative.

--
Ronald Pomeroy "VisualWorks - the *only* real competition for IB"
Advanced Micro Devices
CAM Applications Group
rpom...@aunext1.amd.com
(NeXTmail preferred)

Don Yacktman

unread,
Feb 10, 1993, 2:31:50 PM2/10/93
to

In article <1993Feb10.1...@netcom.com>, yi...@netcom.com (Michael Brill) writes:
>[miscellaneous doom and gloom happily deleted]
>
>Apart from the occasional "NeXT should only make X terminals" type comments,
>it seems that the overwhelming view of csna has been that NeXT should drop
>hardware and concentrate on making NeXTSTEP ubiquitous. Hmmmm... lessee...
>NeXT drops hardware and is concentrating on NeXTSTEP. I expected to hear
>the sound of champagne corks popping, not whines about the market value of
>their slabs! :/

I think the timing is perhaps the biggest problem. At this point, it
would be silly to go and buy a NeXT, since it's to be discontinued. But
there's no other way to get NeXTSTEP! That means that until the 486 version
ships, the NeXTSTEP market _will_not_grow_! If you're a small start up
software company hoping to make money off your products, this will not
make you happy. It would have been much better (for the NeXT developer)
if the announcement were to come _after_ NS486 is on the shelves in the
local software store... then there would be an alternative to move to!
I'll bet NeXT sales drop to near zero starting now and will stay that way
until NS486 has shipped. And how much longer will it be delayed? Perhaps
only way NeXT will be able to get any sales at all is to drop the
prices on the hardware. (Can you say fire sale?) If that happens, I'll
snap up a few slabs, though. It's good hardware, and will be good
for several years. At a reduced price... well, everyone is this group
has been telling NeXT to drop their prices anyway. Wish granted? :-)

>I, for one, am more optimistic than ever (and I've nothing vested in the
>whole game).

I haven't lost a positive outlook, and I simply hope to high heaven that
somebody at NeXT knows what they're doing...perhaps there's a few hidden
factors that we aren't aware of that make this decision rosier than it
appears on the surface. At any rate, I'm trying to speed up the projects
I'm working on for now...if NeXT does come out on top after this, I want to
be sure I'm there! Let's also hope that the small start up software
companies survive the upcoming lack of sales which might (_might_ == NOT
guaranteed, but it wouldn't surprise me if it happened) happen. Even if
NeXT survives, I would not want to see companies like Srone Design,
Lighthouse, and Right Brain disappear. They are a much, much better
alternative to MicroSnot and Word(im)Perfect by a long shot!

Folks, let's lighten up a bit, "nightmare" or not! If we look beyond
the next 6 months, things don't look quite so bad.

Well, gotta get to work. :-)

Later,
-Don Yacktman
ya...@alaska.et.byu.edu

Yamanari

unread,
Feb 10, 1993, 2:55:05 PM2/10/93
to
In article <56#@byu.edu> ya...@idaho.et.byu.edu (Don Yacktman) writes:
>
>I think the timing is perhaps the biggest problem. At this point, it
>would be silly to go and buy a NeXT, since it's to be discontinued. But
>there's no other way to get NeXTSTEP! That means that until the 486 version


At least this takes care of the Fatbinaries/cross-compiling
thread quite nicely. It's importance, in the light of NeXT's
decision, is minor.


>local software store... then there would be an alternative to move to!
>I'll bet NeXT sales drop to near zero starting now and will stay that way
>until NS486 has shipped. And how much longer will it be delayed? Perhaps
>only way NeXT will be able to get any sales at all is to drop the
>prices on the hardware. (Can you say fire sale?) If that happens, I'll


Maybe we finally hit upon a strategy to get NeXT to maintain
price parity with the rest of the industry.

> After the next 6 months, things should be looking up.


Yep. the price of a '486 machine should be down to about $1800
for a nice one. And the Pentium mightbe out.


--
"The waste.. The waste sometimes brings tears to my eyes. Just the
thought that all of those animals are being put into garbage bags and
thrown in lakes or dumpsters when they could be gloves or shoe laces,
or feeding the homeless." (kill file suggestion: "/ :rsrodger/h :j")

Meinhard E. Mayer (Hardy)

unread,
Feb 10, 1993, 5:48:49 PM2/10/93
to
You should have shown him SAM, and taught him to alias:
lprm -> cancel
lpq -> lpstat
lpr already exists as a script.
I used sam 36 times last week to set up the 36 incarnations of my NeXT
on SLIP (randomly assigned hostnames, every time I log in) to be able
to print on my NeXT_Printer from HP-UX.
I found HP-UX system administration completely painless -- about the
same as using Installer.app on the NeXT (provided you read the manual,
of course).
--
Hardy
-----
Meinhard E. Mayer, Department of Physics, UC Irvine
e-mail: ha...@golem.ps.uci.edu (preferred) or MMA...@UCI.BITNET
!!!! NO NEXTMAIL TO THESE ADDRESSES, PLEASE !!!!!

Ron Pomeroy x(Coop)

unread,
Feb 11, 1993, 10:06:31 AM2/11/93
to
In article <justin.729400080@hp750> jus...@hp750.me.engr.uky.edu (Justin
Sullivan) writes:

>>rpom...@aunext1.amd.com (Ron Pomeroy x(Coop)) writes:
>>
>>>RUN - don't walk - to your nearest CRAY representative. Don't let your HP
>>>do it: It crawls like a crippled sea slug. You'll be old and gray by the
>>>time it gets anything done.
>>
>>>As Einstein said...It's all relative.
>>
>>Considering that we have a respectable fraction of a Cray's performance at
>>a terrific fraction of the cost, and given that performance will more than
>>double when it's upgraded around March to a 755 for a pittance, I would
>>have to think that a Cray rep would try to sell me something I don't need
>>for a price that would bankrupt our already-bankrupt institution.

>>--
>>* Justin P. Sullivan * jus...@engr.uky.edu (Internet) *
>>* System Administrator * jus...@mik.uky.edu (NeXTMail) *
>>* Computational Fluid Dynamics * JUSTIN@UKCC (BITNET) *
>>* (606) 257-2268 * ukma!engr.uky.edu!justin (UUCP) *

Well, say what you will but every person in my office (with NeXTs) was offered
an HP for his desk (to replace their turbo color slabs). They *AL*L said "NO
THANKS". With Co-Xist I can tap the HP's cpu no trouble. Heck, with Co-Xist
my machine feels every bit as fast as running local on the HP. Also, every
person with an HP one his/her desk also has a Mac and or a PC on their desk.
The NeXT users have one (granted - orphaned now) computer on their desk. I
will continue to be productive on my slab.

Now, as soon as I can have NeXTSTEP/HP I'll tell my boss I'll take that machine
whenever he's ready!!

Cheers!

Justin Sullivan

unread,
Feb 10, 1993, 10:02:08 PM2/10/93
to
kgn...@cs.concordia.ca (MATIS stephane) writes:

> I usually stay out of these "mini" flames, but Justin, most of
>us don't do "Computational Fluid Dynamics" or our NeXT.... and trully, a
>HP 9000/750 is $$$ more than my lowlly NeXTstataion color.

Well, you get what you pay for.. And you got screwed. It would really piss me
off if the company that created the center of my life just pulled the rug
right out from under me. If no one has looked lately, that black slug of yours
has been quietly beaten by everthing that's new and in.

> I'm sure my DSP can crank some horsepower ... but not for what

DSP wasn't put to very effective use, and it's a shame, but you won't get
a DSP on a stock 486.. Something else NeXT quit on? Perhaps they get their
new 'quitter' philosophy from a previous board member who goes by HRP..

Justin Sullivan

unread,
Feb 10, 1993, 10:08:00 PM2/10/93
to
rpom...@aunext1.amd.com (Ron Pomeroy x(Coop)) writes:

>RUN - don't walk - to your nearest CRAY representative. Don't let your HP
>do it: It crawls like a crippled sea slug. You'll be old and gray by the
>time it gets anything done.

>As Einstein said...It's all relative.

Considering that we have a respectable fraction of a Cray's performance at


a terrific fraction of the cost, and given that performance will more than
double when it's upgraded around March to a 755 for a pittance, I would
have to think that a Cray rep would try to sell me something I don't need
for a price that would bankrupt our already-bankrupt institution.

Gordon Van Huizen

unread,
Feb 11, 1993, 12:41:26 PM2/11/93
to
In article <1993Feb10.0...@netcom.com> v...@netcom.com (Vin
Locke) writes:
> Do people realize that as of this "no more hardware" leak,
> NeXT no longer has a product? NS 486, oh yeah. er, when did
> they say it would be out?
>
As someone who makes a living selling software products for the machine
it has crossed my mind...

Also as a company who has spent a fair amount of time and money
developing a software product that requires a NeXT box (not a PC
running NeXTSTEP) we have had our fair share of interesting internal
discussions over the last few days...

Gordon
--
----------------------------------------------------------------
Gordon Van Huizen vox: 619.488.9411 fax: 619.488.3045
Metrosoft g...@metrosoft.com [NeXTmail welcome]
----------------------------------------------------------------

Barry Merriman

unread,
Feb 10, 1993, 8:20:25 PM2/10/93
to
Essentially this would be a good thing: with NeXT
out of the HW biz, the other workstation manufactures would
not view them as a competitor, and would line up to license
NeXTSTEP.


Then we would have our real goal, which is to see NeXTSTEP
running all over the place.

--
Barry Merriman
UCLA Dept. of Math
UCLA Inst. for Fusion and Plasma Research
ba...@math.ucla.edu (Internet; NeXTMail is welcome)


Nathan F. Janette

unread,
Feb 11, 1993, 2:07:39 AM2/11/93
to
In article <1993Feb11....@math.ucla.edu>
ba...@arnold.math.ucla.edu (Barry Merriman) writes:
> Essentially this would be a good thing: with NeXT
> out of the HW biz, the other workstation manufactures would
> not view them as a competitor, and would line up to license
> NeXTSTEP.
>
>
> Then we would have our real goal, which is to see NeXTSTEP
> running all over the place.

But for some (many?) of us, that wasnt' the *only* goal.

For all the pricing problems of NeXT hardware (thanks a lot
to all the NeXT "marketing" people), they built a good
machine that offered plenty of performance even when
compared to supposedly faster RISC systems. I was
personally attracted to the sharp 17" megapixel display,
and the keyboard and mouse. I prefer them to any other
user-interface gear I've seen, and I've seen plenty.

I feel like a violinist that was just told that my favorite
strings and bow wouldn't be sold any more, but it was a good
thing because now I can play my music in more concert halls,
I just have to use rubber bands for strings and a baseball
bat for a bow.

I truly hope Canon will place it's name on the hardware line,
especially the almost-completed Power-PC machine, sell
them at competitive prices, and keep them painted BLACK.

--
Nathan Janette
PPP link from hilbert.csb.yale.edu

Please reply to: nat...@laplace.csb.yale.edu (NeXT)

Michael Brill

unread,
Feb 10, 1993, 4:47:59 PM2/10/93
to
> ya...@idaho.et.byu.edu (Don Yacktman) writes

>I think the timing is perhaps the biggest problem. At this point, it
>would be silly to go and buy a NeXT, since it's to be discontinued. But
>there's no other way to get NeXTSTEP! That means that until the 486 version
>ships, the NeXTSTEP market _will_not_grow_!

Agreed. Timing isn't great - say about a week too early. I had it
that NeXT planned a slew of announcements for February 16th including
the status of hardware, strategic alliances, etc. Given the fact that
they're not hemoraging cash (attaining profitability last quarter), I
find it hard to believe that there isn't a good reason for making the
announcement now (er, the 16th). My guess is that there is a whole lot of
very positive news vis-a-vis strategic h/w alliances (HP beyond rumor stage,
McNealy seen by several folks at NeXT last week) and this is a good time to
announce the complete restructuring. So, the net effect of the announcements
is that NeXT is truly becoming a s/w company and has signed up h/w vendors and
*SOLD* [_not_ just dumping] off their h/w division. I think timing is truly
unfortunate to have the only perceivable (to some) bad news being announced by
itself. I also thing it would be a rather grusome event to have NeXT announce
that it's dropping h/w in the middle of NWE! Note that this is speculation...
reasonably well-informed speculation, but speculation nonetheless.


>If you're a small start up
>software company hoping to make money off your products, this will not
>make you happy. It would have been much better (for the NeXT developer)
>if the announcement were to come _after_ NS486 is on the shelves in the
>local software store... then there would be an alternative to move to!

Yeah, I think that the news for immediately increasing market size is not
promising. Let's say sales drop by 75% until NWE... that's what... maybe
5,000 machines. I don't think that'll tank any s/w companies given the
fact that most of them are doing custom development anyway. I think if
anything, this will give the supply of able NeXTSTEP programmers a chance
to catch up with the great demand! :) :) The upside for s/w developers is
that there is now a 100% focus on making NeXTSTEP run on many platforms
which has orders of magnitude more impact than NeXT hanging on to h/w for
a few more months.


>Perhaps only way NeXT will be able to get any sales at all is to drop the
>prices on the hardware. (Can you say fire sale?) If that happens, I'll

>snap up a few slabs, though. It's good hardware, and will be good
>for several years. At a reduced price... well, everyone is this group
>has been telling NeXT to drop their prices anyway. Wish granted? :-)

They probably don't need all of the _used_ machines they have, so maybe
those will go away. New machines... I don't think they have warehouses
full of them, but the ones they do have might be going with Canon?


>At any rate, I'm trying to speed up the projects
>I'm working on for now...if NeXT does come out on top after this, I want to
>be sure I'm there!

Change is odd. It's something that everyone clamors for and when it happens
it really scares people. This is change. If you keep your eyes open and stay
the course... who knows, you could be the next MicroSnot!

...Michael Brill (yi...@netcom.com)

BTW, anyone want to buy a slightly used ColorStation? <just kidding>


--
----

Peter Karbaliotis

unread,
Feb 11, 1993, 4:30:41 PM2/11/93
to
In article <c663f=g...@rpi.edu> g...@eclipse.its.rpi.edu (Garance A. Drosehn)
writes:
[munch]

>
>While this move is pretty much what many people have pestered NeXT to do,
>the timing of it is odd (to say the least). If this happened after
NS-Intel
>was a real, shipping product (and people were confident that it works as
>well or better than NS 3.0 on NeXT hardware), then this move would seem
>almost inevitable.
>
[munch]

The timing is odd because the story was leaked. :-)

As a matter of fact, NeXT is still making their computers.
They will be filling back-orders for black boxes until May 25th
(is that date a coincidence or what? :-).

Canon has not said anything about what they are going to do.
There will probably make some statement next week.

For those who are worried that they suddenly have a black
paperweight on their desk, take it easy. NeXT will NOT drop
software support for their hardware for quite some time yet.
According to a UPI news item on ClariNet, NeXT sold about 69,000
units in 1992. Together with previous years, there are probably
about 90,000 NeXT computers out there. It would be complete
madness to abandon that many users.

Take a deep breath. Exhale slowly. Repeat as necessary.
Feel better?
:-)

--
Peter Karbaliotis - pe...@beast.math.ualberta.ca - NeXTmail ok
When a particle you don't even know gives you momentum, that's impulse.

Henry McGilton

unread,
Feb 11, 1993, 6:28:45 PM2/11/93
to
In article <1993Feb10.1...@netcom.com> yi...@netcom.com (Michael Brill)
writes:
* [miscellaneous doom and gloom happily deleted]

* Apart from the occasional "NeXT should only make X terminals"
* type comments, it seems that the overwhelming view of csna
* has been that NeXT should drop hardware and concentrate on
* making NeXTSTEP ubiquitous.

* Hmmmm... lessee... NeXT drops hardware and IS concentrating
* on NeXTSTEP. I expected to hear the sound of champagne
* corks popping, not whines about the market value of their slabs!
*
* I, for one, am more optimistic than ever (and I've nothing
* vested in the whole game).
Well said Mike. We had a talk about this yesterday. We decided
to keep our heads down and continue working on the stuff on which
we're working. To let events such as these jerk us around now is
foolish. As I said to a friend this morning:

``NeXT have pulled the plug on the hardware line, and announced
they're now a software only company, with their value added
being NeXTSTEP, Object Oriented development, and Display
PostScript. We decided not to let this sea change deflect us,
since we're writing to a software platform that won't change,
as opposed to a specific machine (amazing how many people can't
figure this out).

They have NeXTSTEP running on the 486 and the Pentium. Market
watchers estimate Intel will sell 23 million Pentium chips in
the first year. Assume NeXTSTEP appears on only one percent
of those Pentium chips -- that's 230,000 systems, for a market
four times bigger than NeXT's current installed base.''

........ Henry

Richard Myers

unread,
Feb 11, 1993, 4:14:20 PM2/11/93
to
OK, the hardware's going... I want to know what the deal with software
is!! When(?) NS 3.1, 4.0... Comes out, do I get my *FULL DEVELOPER COPY*
for the usual upgrade price for whatever machine I happen to be running
on? i.e. When the Pentium machine comes out, and I want to 'upgrade to
high speed color' am I gonna have to shell out the 2.7G's for the system
software (after I shell out for the hardware)?

I want software/hardware upgrade, I supply the hardware, you can have the
old, but I want the software cheap!
Concerned 2+year old Mono Slab owner.
Kurt

Ron Pomeroy x(Coop)

unread,
Feb 12, 1993, 2:10:13 PM2/12/93
to
In article <1993Feb11....@trilithon.mpk.ca.us>


Bam!!! Hit the nail right on the head. NeXT is about to jump into a much
bigger ocean. An ocean where, yes, MicroScrote dominates, but an ocean where
lots of different fish are free to swim, an ocean where capturing only a small
percentage of the market would represent signifigant growth for NeXT!

Does everyone remember the first time you swam in the ocean? A bit scary at
first :-)

Sharks - they live in the ocean too.

--
Ronald Pomeroy "We're about to have a lot of fun"
Advanced Micro Devices -Bruce Henderson

Jonathan Hendry

unread,
Feb 11, 1993, 5:32:39 PM2/11/93
to
pe...@beast.math.ualberta.ca (Peter Karbaliotis) writes:

: The timing is odd because the story was leaked. :-)


:
: As a matter of fact, NeXT is still making their computers.
: They will be filling back-orders for black boxes until May 25th
: (is that date a coincidence or what? :-).

Nope. The plant is closed. Orders are being filled from existing
inventory. (From the official press release)

:
: Canon has not said anything about what they are going to do.


: There will probably make some statement next week.
:
: For those who are worried that they suddenly have a black
: paperweight on their desk, take it easy. NeXT will NOT drop
: software support for their hardware for quite some time yet.
: According to a UPI news item on ClariNet, NeXT sold about 69,000
: units in 1992. Together with previous years, there are probably
: about 90,000 NeXT computers out there. It would be complete
: madness to abandon that many users.

Nope again. That is the number which NeXT has sold altogether.
NeXT made about 140 Million last year. It doesn't add up.

NeXT also said they would support existing warranties and service contracts.

--
Jonathan W. Hendry
Drexel University College Of Info. Studies
tjhe...@queen.mcs.drexel.edu

"The experience of programming Windows vs. the experience of
programming NeXTStep is like going to the dentist and having a root canal
without anaesthetic vs. going to the dentist and having your gums cleaned
w/some nitrous-oxide thrown in for the entertainment side of things."
bb...@stone.co

Yamanari

unread,
Feb 12, 1993, 2:58:38 PM2/12/93
to
In article <1993Feb12.1...@adobe.com> ca...@adobe.UUCP (Perry A. Caro) writes:
>In article <56#@byu.edu> ya...@idaho.et.byu.edu (Don Yacktman) writes:
>>I think the timing is perhaps the biggest problem. At this point, it
>>would be silly to go and buy a NeXT, since it's to be discontinued. But
>>there's no other way to get NeXTSTEP! That means that until the 486 version
>>ships, the NeXTSTEP market _will_not_grow_!


Actually, it will likely gradually shrink as people buy
second computers, *knowing* that there are no apps coming
for their NeXT, and while a lot of NeXTites are rather used to
waiting, very few are going to continue to wait now that NeXT
no longer makes a selling product.


>But consider the alternative. NeXT knows they are going to drop
>hardware, but they don't say anything. Rumors fester. There is
>turmoil within NeXT itself as all engineers start to wonder if they are
>going to have a job.


They can always go work for Sun like the others.


>Negotiations with third parties (like Dell,
>Compaq, HP) won't say that they are dropping hardware. Most important
>of all, hardware sales decline ANYWAY because there is so much
>uncertainty, and so many people waiting for a RISC announcement.


Waiting.. and waiting.. and waiting--how long? NeXTies have
been screaming the praises of the New RISC NeXT! for as long as
I've read this group, which much be more than two years now.

Kinda like how Amiga users sing of the glory of the Amiga in
Europe, where it is "taken seriously as a business machine"
(guffaw).


>I think it was a matter of cutting losses.


I agree.


>>Folks, let's lighten up a bit, "nightmare" or not! If we look beyond
>>the next 6 months, things don't look quite so bad.
>

>I'd rephrase that as, "If NeXT survives the next 6 months, things might
>not be so bad."


I'd rephrase that as:

"If NeXT learned their lesson about pricing, and has learned
that no matter how pretty something is the world is not going
to beat a path to their door until they get more than two real
apps, and has learned that their PR sucks, AND they survive
the next few months AND get NS486 out before or at the same
time as Windows NT AND the price is similar AND they include
DOS/Windows compatibility AND they are as supportive as
IBM has learned the hard way that they have to be--NeXT
might come out of this OK."

A lot of ands, though. That's a few too many lessons for
NeXT to have learned so quickly, given their past performance.


>I used to laugh at the yahoos that would regularly predict the doom of
>NeXT. Now I'm thinking the odds are on the side of the yahoos, and it
>galls me.


A friend reminded me that I predicted that NeXT would stop making
hardware and in one sudden convulsion convert to a software
only company--which will float up until just releasing it's
first product, only to flounder and sink, to be bought up by
another company like MS or IBM.

That was ~8 months ago. We both thought it was sad, because
were are both in love with NeXTstep (but not the NeXT hardware,
which takes a stretch to imagine that it had a good price-performance
ratio)

I still get laughs looking at my collection of "Death of Sun is
Nigh!" type posts from this group. Almost sad, in a pathetic
sort of way.

<This coming from someone who sincerely hopes that if anyone
buys NeXT, that it is Apple--because at least they would offer
buyable systems with it, as opposed to HP (who would only offer
it on HP machines) or IBM (who is committed to AIX).>

God forbid we end up with "Microsoft N(eX)T Windows for Intel" and
"Microsoft N(eX)T Windows for Alpha" and so on.. Though
I guess NeXT-NT would work better than NeXT-on-AIX.

Joe Freeman

unread,
Feb 12, 1993, 7:30:13 PM2/12/93
to
In article <1993Feb12.1...@adobe.com> ca...@adobe.UUCP (Perry A. Caro) writes:
>In article <56#@byu.edu> ya...@idaho.et.byu.edu (Don Yacktman) writes:
>>I think the timing is perhaps the biggest problem. At this point, it
>>would be silly to go and buy a NeXT, since it's to be discontinued. But
>>there's no other way to get NeXTSTEP! That means that until the 486 version
>>ships, the NeXTSTEP market _will_not_grow_!
>
>But consider the alternative. NeXT knows they are going to drop
>hardware, but they don't say anything. Rumors fester. There is
>turmoil within NeXT itself as all engineers start to wonder if they are
>going to have a job. Negotiations with third parties (like Dell,

>Compaq, HP) won't say that they are dropping hardware. Most important
>of all, hardware sales decline ANYWAY because there is so much
>uncertainty, and so many people waiting for a RISC announcement.
A well phrased resonse. I think the analysis is quite good. The
only quibble I have with it is the people waiting for the RISC
announcement. I don't think there would have been any reason
to wait.

The 486 box is faster than a turbo and faster than what would have been
the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
which would make the box as fast as any "low cost" RISC box
NeXT would have introduced. Why would anyone want a RISC box?


--
Joe Freeman jfr...@uncecs.edu
The opinions espressed here are my own and are not shared by my employer or
anyone else I know of. Flames can be mailed to me or routed to /dev/null.

Bruce F. Webster

unread,
Feb 12, 1993, 9:25:27 PM2/12/93
to
In article <1993Feb10.1...@wam.umd.edu> rsro...@wam.umd.edu (Yamanari)
writes:

> In article <56#@byu.edu> ya...@idaho.et.byu.edu (Don Yacktman) writes:
> >
> >I think the timing is perhaps the biggest problem. At this point, it
> >would be silly to go and buy a NeXT, since it's to be discontinued. But
> >there's no other way to get NeXTSTEP! That means that until the 486 version
>
>
> At least this takes care of the Fatbinaries/cross-compiling
> thread quite nicely. It's importance, in the light of NeXT's
> decision, is minor.
>

Not really. There's currently an installed base of about 70,000 NeXT systems.
It's going to be a while before there's that many installed NS-486 systems, so
fat binaries are going to be _really_ important for the rest of '93 and
probably on into '94. Then they're going to be even more important when
NeXTSTEP comes out for other processors. :-) ..bruce..

-------------------------------------------------------------------------------
Bruce F. Webster | Am a flamer, goateed, pallid, overweight,
Chief Technical Officer | Willing to pull two shifts, then (hell) a third,
Pages Software Inc | To save a session from a deadlocked state;
bweb...@pages.com | At times, (to put it mildly) unrestrained--
#import <pages/disclaimer.h> | Almost, at times, a nerd. -- Jeff Duntemann
-------------------------------------------------------------------------------

Nathan F. Janette

unread,
Feb 12, 1993, 10:53:40 PM2/12/93
to
In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu
(Yamanari) writes:
> In article <1993Feb13....@cs.yale.edu>
nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
> >In article <1993Feb13.0...@ecsvax.uncecs.edu>
> >jfr...@ecsvax.uncecs.edu (Joe Freeman) writes:
> >
> >> The 486 box is faster than a turbo and faster than what would have
been
> >> the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
> >> which would make the box as fast as any "low cost" RISC box
> >> NeXT would have introduced. Why would anyone want a RISC box?
> >
> >The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50
> >DX2), did not feel nearly as fast (or good, but that's moot
> >now) as our NeXTstation Color Turbo machines. It wasn't even
> >close to the response, and it had the JAWS video card too.
>
> All this really says is that NeXT has done a crummy job
> porting NeXTStep. The Dell is a lot faster than the
> Turbo slab, and the obvious explanation for the mediocre
> performance is NeXTStep, not some mysterious stuff
> like the "great engineering" of the slabs.

Would you care to justify that opinion, or should I just take
it on the weight of your belief? Don't tell me you think MIPs
numbers are a reasonable indication of this...

> > Would anyone who knows what they are talking
> >about even care to suggest, even in humor, that any PC memory
> >subsystem can hold a candle to the NeXT Turbo's?
>
> What subsystems might those be? The dumb frame buffer?
> The (guffaw) SCSI2?

The MEMORY subsystem, just as I stated. Your getting ahead of
yourself and not reading the post you are having so much fun
trying to criticize. How about responding to the question?

> > What good
> >are a few more MIPs if that CPU is sitting around stroking
> >itself while waiting for data?
>
> Or maybe the Dell is an ISA machine--in that case,
> nothing to be done. Simple--don't buy ISA if you've moving
> lots of data.

That has nothing to do with the memory subsystem.

Perry A. Caro

unread,
Feb 12, 1993, 2:15:50 PM2/12/93
to
In article <56#@byu.edu> ya...@idaho.et.byu.edu (Don Yacktman) writes:
>I think the timing is perhaps the biggest problem. At this point, it
>would be silly to go and buy a NeXT, since it's to be discontinued. But
>there's no other way to get NeXTSTEP! That means that until the 486 version
>ships, the NeXTSTEP market _will_not_grow_!

But consider the alternative. NeXT knows they are going to drop


hardware, but they don't say anything. Rumors fester. There is
turmoil within NeXT itself as all engineers start to wonder if they are
going to have a job. Negotiations with third parties (like Dell,
Compaq, HP) won't say that they are dropping hardware. Most important
of all, hardware sales decline ANYWAY because there is so much
uncertainty, and so many people waiting for a RISC announcement.

I think it was a matter of cutting losses.

>Folks, let's lighten up a bit, "nightmare" or not! If we look beyond


>the next 6 months, things don't look quite so bad.

I'd rephrase that as, "If NeXT survives the next 6 months, things might
not be so bad."

I used to laugh at the yahoos that would regularly predict the doom of


NeXT. Now I'm thinking the odds are on the side of the yahoos, and it
galls me.

Perry
--
ca...@mv.us.adobe.com ...!{sun}!adobe!caro Contents: my opinions, no others

David Lemson

unread,
Feb 13, 1993, 12:33:18 AM2/13/93
to
rsro...@wam.umd.edu (Yamanari) writes:

>In article <1993Feb13....@cs.yale.edu> nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:

>>In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu
>>(Yamanari) writes:

>>The MEMORY subsystem, just as I stated. Your getting ahead of
>>yourself and not reading the post you are having so much fun
>>trying to criticize. How about responding to the question?


> Wanna be a little more specific on how the NeXT memory
> is somehow mysteriously better than that in a DX-50?
Although the NeXT may only have a 33 MHz cpu bus, innovations in the
Turbo units made memory accesses much faster. The memory bus in the
Turbo NeXTs is simply better engineered than most PC's.

>>> > What good
>>> >are a few more MIPs if that CPU is sitting around stroking
>>> >itself while waiting for data?
>>>
>>> Or maybe the Dell is an ISA machine--in that case,
>>> nothing to be done. Simple--don't buy ISA if you've moving
>>> lots of data.
>>
>>That has nothing to do with the memory subsystem.


> On the other hand, it has -a lot- to do with percieved performance,
> since NeXTStep is bitmap heavy, and shuffling lots of data to
> an ISA video card is a sure-fire way to generate wait states.

Yes, running NeXTSTEP on an ISA machine if you want to have
NeXTStation-like performance is probably a mistake. On the other
hand, running it on a DX2/66 or DX/50 with local bus Mach 32 video
will deliver faster video, according to an NeXTSTEP 486 engineer
I have talked to. (Ugh, I hope that guy is still there...he is a
great guy) He told me that video performance on a localbus DX2/66
system is perhaps as much as twice as fast as the same video card on
an EISA bus.
I suspect that the Dell with JAWS that Nathan saw (I think it was
Nathan) was some time ago, and the JAWS card was an EISA card. Note
that EISA is restricted to an 8 MHz bus speed. Notwithstanding the
fact that the JAWS card probably has less performance than the
Wingine or Mach 32 chip (I'm speculating there), if you go with
Local Bus, you're bound to be doing better.

Myself, I was poised to buy a Gateway 486/66V system in January. I
was persuaded to wait... The only thing about that machine is that
it has an integrated 32-bit IDE controller on the motherboard. I
checked with NeXT.. that is just fine for the hard disk. HOWEVER,
you absolutely will need the proper SCSI card and CD-ROM drive for
the initial install, at least for the pre-release. Hopefully, NeXT
will include drivers in the 'real' release for standard MPC-compliant
CD-ROM systems like Gateway sells for <$300. (I think $200 or $250!)

--
David Lemson (217) 244-1205
University of Illinois NeXT Campus Consultant / CCSO NeXT Lab System Admin
Internet : lem...@uiuc.edu UUCP :...!uiucuxc!uiucux1!lemson
NeXTMail & MIME accepted BITNET : LEMSON@UIUCVMD

Yamanari

unread,
Feb 12, 1993, 11:06:40 PM2/12/93
to
In article <1993Feb13....@cs.yale.edu> nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
>In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu
>(Yamanari) writes:
>> In article <1993Feb13....@cs.yale.edu>
>> >The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50
>> >DX2), did not feel nearly as fast (or good, but that's moot
>> >now) as our NeXTstation Color Turbo machines. It wasn't even
>> >close to the response, and it had the JAWS video card too.
>>
>> All this really says is that NeXT has done a crummy job
>> porting NeXTStep. The Dell is a lot faster than the
>> Turbo slab, and the obvious explanation for the mediocre
>> performance is NeXTStep, not some mysterious stuff
>> like the "great engineering" of the slabs.
>
>Would you care to justify that opinion, or should I just take
>it on the weight of your belief? Don't tell me you think MIPs
>numbers are a reasonable indication of this...


Faster machine, running new software software.
Slower performance.
Draw the obvious conclusion.


>> > Would anyone who knows what they are talking
>> >about even care to suggest, even in humor, that any PC memory
>> >subsystem can hold a candle to the NeXT Turbo's?
>>
>> What subsystems might those be? The dumb frame buffer?
>> The (guffaw) SCSI2?
>
>The MEMORY subsystem, just as I stated. Your getting ahead of
>yourself and not reading the post you are having so much fun
>trying to criticize. How about responding to the question?

Wanna be a little more specific on how the NeXT memory
is somehow mysteriously better than that in a DX-50?

>> > What good


>> >are a few more MIPs if that CPU is sitting around stroking
>> >itself while waiting for data?
>>
>> Or maybe the Dell is an ISA machine--in that case,
>> nothing to be done. Simple--don't buy ISA if you've moving
>> lots of data.
>
>That has nothing to do with the memory subsystem.

On the other hand, it has -a lot- to do with percieved performance,
since NeXTStep is bitmap heavy, and shuffling lots of data to
an ISA video card is a sure-fire way to generate wait states.

--

David Lemson

unread,
Feb 13, 1993, 2:33:02 AM2/13/93
to
lem...@ux1.cso.uiuc.edu (David Lemson) writes:

>I suspect that the Dell with JAWS that Nathan saw (I think it was
>Nathan) was some time ago, and the JAWS card was an EISA card. Note
>that EISA is restricted to an 8 MHz bus speed. Notwithstanding the
>fact that the JAWS card probably has less performance than the
>Wingine or Mach 32 chip (I'm speculating there), if you go with
>Local Bus, you're bound to be doing better.

I have just looked at the NS/INTEL matrix and it seems that
the supported Dell JAWS machine has it on the motherboard. So, it
would seem that maybe JAWS is the most powerful graphics platform
after all. Since JAWS is getting no press in the PC rags, I haven't
seen any benchmarks. Anyone know how its graphics processor rates
against the Mach 32 for important things like BitBlt's etc.?

Nathan F. Janette

unread,
Feb 12, 1993, 8:08:59 PM2/12/93
to

> The 486 box is faster than a turbo and faster than what would have been
> the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
> which would make the box as fast as any "low cost" RISC box
> NeXT would have introduced. Why would anyone want a RISC box?

The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50


DX2), did not feel nearly as fast (or good, but that's moot
now) as our NeXTstation Color Turbo machines. It wasn't even
close to the response, and it had the JAWS video card too.

It takes more than a faster CPU to beat the well-engineered
NeXT hardware. Would anyone who knows what they are talking


about even care to suggest, even in humor, that any PC memory

subsystem can hold a candle to the NeXT Turbo's? What good


are a few more MIPs if that CPU is sitting around stroking
itself while waiting for data?

I've also heard the Pentium won't be cheap; $1,200 was the
figure I heard from someone at NeXT.

Yamanari

unread,
Feb 12, 1993, 8:41:18 PM2/12/93
to
In article <1993Feb13....@cs.yale.edu> nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
>In article <1993Feb13.0...@ecsvax.uncecs.edu>
>jfr...@ecsvax.uncecs.edu (Joe Freeman) writes:
>
>> The 486 box is faster than a turbo and faster than what would have been
>> the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
>> which would make the box as fast as any "low cost" RISC box
>> NeXT would have introduced. Why would anyone want a RISC box?
>
>The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50
>DX2), did not feel nearly as fast (or good, but that's moot
>now) as our NeXTstation Color Turbo machines. It wasn't even
>close to the response, and it had the JAWS video card too.

All this really says is that NeXT has done a crummy job
porting NeXTStep. The Dell is a lot faster than the
Turbo slab, and the obvious explanation for the mediocre
performance is NeXTStep, not some mysterious stuff
like the "great engineering" of the slabs.

>It takes more than a faster CPU to beat the well-engineered
>NeXT hardware.


Whoops, there it is.


> Would anyone who knows what they are talking
>about even care to suggest, even in humor, that any PC memory
>subsystem can hold a candle to the NeXT Turbo's?

What subsystems might those be? The dumb frame buffer?
The (guffaw) SCSI2?

> What good
>are a few more MIPs if that CPU is sitting around stroking
>itself while waiting for data?

Or maybe the Dell is an ISA machine--in that case,
nothing to be done. Simple--don't buy ISA if you've moving
lots of data.

>I've also heard the Pentium won't be cheap; $1,200 was the
>figure I heard from someone at NeXT.


That sounds about right--remember what a 486 was at release?
Or even a 386? I remember the cost of a '387 alone was
almost $900 at one point.

Joe Freeman

unread,
Feb 13, 1993, 8:40:44 AM2/13/93
to
In article <1993Feb13....@cs.yale.edu> nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
>In article <1993Feb13.0...@ecsvax.uncecs.edu>
>jfr...@ecsvax.uncecs.edu (Joe Freeman) writes:
>
>> The 486 box is faster than a turbo and faster than what would have been
>> the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
>> which would make the box as fast as any "low cost" RISC box
>> NeXT would have introduced. Why would anyone want a RISC box?
>
>The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50
>DX2), did not feel nearly as fast (or good, but that's moot
>now) as our NeXTstation Color Turbo machines. It wasn't even
>close to the response, and it had the JAWS video card too.
You probably saw 2.0 at a trade show. The version that
was demonstrated for the last year did not have shared libraries
which dramaticly degraded app launch performance and increased
the amount of swapping the x86 boxes did.

>
>It takes more than a faster CPU to beat the well-engineered
>NeXT hardware. Would anyone who knows what they are talking
>about even care to suggest, even in humor, that any PC memory
>subsystem can hold a candle to the NeXT Turbo's? What good
>are a few more MIPs if that CPU is sitting around stroking
>itself while waiting for data?

I've seen them both and the Intel box is either as fast or
faster with the right configuration.. Now, an intel box
with ISA graphics or a cheap disk controller will be slower.
But, you can build an intel box ithat is faster in all
benchmarks than an NS Turbo.

The real problem is going to be when people cut corners and
the performance is not as high as they expected. The
configuration of an intel based machine will have a direct
and measurable impact on performance.

Joe Freeman

unread,
Feb 13, 1993, 8:48:32 AM2/13/93
to
In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu (Yamanari) writes:
>In article <1993Feb13....@cs.yale.edu> nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
>>In article <1993Feb13.0...@ecsvax.uncecs.edu>
>>jfr...@ecsvax.uncecs.edu (Joe Freeman) writes:
>>
>>> The 486 box is faster than a turbo and faster than what would have been
>>> the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
>>> which would make the box as fast as any "low cost" RISC box
>>> NeXT would have introduced. Why would anyone want a RISC box?
>>
>>The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50
>>DX2), did not feel nearly as fast (or good, but that's moot
>>now) as our NeXTstation Color Turbo machines. It wasn't even
>>close to the response, and it had the JAWS video card too.
>
>
> All this really says is that NeXT has done a crummy job
> porting NeXTStep. The Dell is a lot faster than the
> Turbo slab, and the obvious explanation for the mediocre
> performance is NeXTStep, not some mysterious stuff
> like the "great engineering" of the slabs.

All this says is that people haven't seen the release code
or run it on the right box. Please read my response Nathan's
message for possible explanations of what he saw.


>
>
>> Would anyone who knows what they are talking
>>about even care to suggest, even in humor, that any PC memory
>>subsystem can hold a candle to the NeXT Turbo's?
>
>
> What subsystems might those be? The dumb frame buffer?
> The (guffaw) SCSI2?

The dumb frame buffer is the way to go for PS graphics.
It is the same way many of the workstations have their
graphics layed out and the same way many of the newer
high end PC graphics are going. An accellerated graphcis
card only does you any good if it accellerates the operations
you can sand it. With faster CPUs, you get good performance
with direct mapped video memory with the added benefit that
the video will be fast for other operationg systems.

I don't believe NeXT ever claimed that the disk was SCSI2.
It just uses the SCSI2 connector. Better noise immunity
while retaining a small footprint.

>> What good
>>are a few more MIPs if that CPU is sitting around stroking
>>itself while waiting for data?
>
>
> Or maybe the Dell is an ISA machine--in that case,
> nothing to be done. Simple--don't buy ISA if you've moving
> lots of data.

The configuration while have a direct impact on performance.
The demonstration machines all were running without the
shared libraries and certain caches. This affected launch and
swap time.

ja...@skatter.usask.ca

unread,
Feb 13, 1993, 8:56:02 PM2/13/93
to
From article <justin.729400080@hp750>, by jus...@hp750.me.engr.uky.edu (Justin Sullivan):

> rpom...@aunext1.amd.com (Ron Pomeroy x(Coop)) writes:
>
>>RUN - don't walk - to your nearest CRAY representative. Don't let your HP
>>do it: It crawls like a crippled sea slug. You'll be old and gray by the
>>time it gets anything done.
>
>>As Einstein said...It's all relative.
>
> Considering that we have a respectable fraction of a Cray's performance at
^^^^^^^^^^^^^^^^^^^^
Ha ha HA ho ho hee hee hee hah ha hooo hee HAAAA ha ha har hoof
ha ha har har he hee ho HAA (I can't breathe!) ha hee hee ha ha HAAA
ha har ha ha ha .. cough .. hack ...


My apologies, but I couldn't help myself.

jake


> --
> * Justin P. Sullivan * jus...@engr.uky.edu (Internet) *
> * System Administrator * jus...@mik.uky.edu (NeXTMail) *
> * Computational Fluid Dynamics * JUSTIN@UKCC (BITNET) *
> * (606) 257-2268 * ukma!engr.uky.edu!justin (UUCP) *

--
Jason C. Breckenridge ja...@skatter.usask.ca

Why you wanna be readin' them long-haired books for, boy? - Foghorn Leghorn

Yamanari

unread,
Feb 13, 1993, 8:32:15 PM2/13/93
to
In article <1993Feb13....@pencom.com> ricardo@chuy (Ricardo Parada) writes:
>Don't forget that people will still be able to run DOS/Windows apps. The
>question is, will this capability be bundled with NS? This will be important
>for single users of PCs.


Yes indeedy. An apt statement.


>It won't be that important for companies since they
>canafford to buy SoftPC, get a site license and put in on the network so that
>their users can run DOS/Windows software at native speeds and coexist with
>NeXTSTEP. Users will be able to also run X apps. You have more options than
>if you don't have NeXTSTEP.


Hm. What options are those? I don't see anything above that
couldn't be done with Univel, and Interactive UNIX ads claim
the same thing: X, DOS&Win, easy networking.

Oh, the option to run NeXTprograms? <not even going to bother>

NeXTStep is the only really selling point of NeXTStep, and
a pretty interface is one thing if it's not going to cost you
a fortune, and another if you need to spend $2500 to get a
copy you can program on.


>NeXT can say: "Waiting for Cairo and Taligent to take advantage of OO
>technology? Call NeXT NOW for a free demonstration!"


It backfires if you advertise your competitors in your
own ads. Better would be:

-------------

Object <big text>
Oriented
Operating

Today.

Call NeXTsoft for a free demonstration. <small>
phone

The NeXTStep/Intel operating system runs on your 80486 <tiny>
and above hardware. <trademarks>

---------------


But when was the last time you saw eye-catching
NeXT advertising? Maybe Apple ("This is making it
easier?") or Dell ("BANG BANG BANG" dead-piggybank)
will loan them their ad team.


>to have used these. I also love the Workspace Manager and Mail.app, that
>alone makes me feel in wonderland. Not true with any other operating
>environments!


It would be nice if you could do more from the keyboard, if
Mail had built in (not added) automatic quoting..


>I think the only problem will be the transition period since even though they
>are taking orders for NS486 right now, the operating system will not ship
>until May 25.


Didn't someone mention--on this group--a few months back
that GM or someone elsehad already ordered 100k copies?


--
--- Attention: All private email *must* go to rsro...@wam.umd.edu ---

Looks good on the outside but..
intel inside (stolen)

Ricardo Parada

unread,
Feb 13, 1993, 5:53:58 PM2/13/93
to
In article Vin Locke writes:
>
> Also, NeXT, to make it as a software company, will have to have software to
> sell. I could see some pretty interesting discussion coming up on the
> net . . .
>

Don't forget that people will still be able to run DOS/Windows apps. The
question is, will this capability be bundled with NS? This will be important

for single users of PCs. It won't be that important for companies since they
can afford to buy SoftPC, get a site license and put in on the network so that

their users can run DOS/Windows software at native speeds and coexist with
NeXTSTEP. Users will be able to also run X apps. You have more options than
if you don't have NeXTSTEP.

NeXT can say: "Waiting for Cairo and Taligent to take advantage of OO

technology? Call NeXT NOW for a free demonstration!"

Also there are already some very nice NeXTSTEP apps out there: I love
Diagram!, Notebook and Mesa. I'm sure people love other apps, I just happen

to have used these. I also love the Workspace Manager and Mail.app, that
alone makes me feel in wonderland. Not true with any other operating
environments!

I think the only problem will be the transition period since even though they

are taking orders for NS486 right now, the operating system will not ship
until May 25.

--
+ Ricardo J. Parada

Edward Jung

unread,
Feb 14, 1993, 4:27:56 AM2/14/93
to

A couple of observations, IMHO, of course:

One nice thing about the refocusing of the company is that it places
greater emphasis on alliances as a competitive advantage. This is
something that arguably is more tuned to Job's skill sets than that
of managing the company's day to day tactics.

On the negative side, for hardware innovation on the PCs (and even
many of the RISC platforms), NeXT will be competing against possibly
higher-volume products. As an example, as the hardware industry gets
more competitive, they are spending a bit more time looking at ways
to optimize their hardware designs for the software environment.
Examples are PCI, Localbus, the PCMCIA design, power management,
video acceleration and capture, sound, etc. Today most of this is
focused on Windows (e.g. there is a regular Windows hardware
conference for exactly this reason). This means it is more likely
that NeXT will have to track h/w optimizations done by h/w vendors
for their highest-volume systems, regardless of whether or not an
option exists to bundle NeXTstep.

NeXT should not be distracted by this refocusing. Nobody should
think that it gives NeXT so much more breathing room to remain
at low volumes. What this represents is a FOCUS. A gathering of
breath for the finishing sprint. NeXT should clearly understand
that it must become a strategic volume platform, and its move to
s/w only is Job's way of focusing his business on exactly this
goal.

Too bad. The hardware was clearly cool.

Another thing that nobody has seemed to note (but is interesting
to me). In PC Week (and other places), Jobs dismissed Taligent and
IBM as his competition. He clearly identified Microsoft's Cairo
project as his competition, stating that if his company could not
best Taligent, IBM, and others, that they would be "incompetent".

At least it's nice to be taken seriously for a change...

--
Edward Jung, Software Architect edw...@microsoft.com
Advanced Systems, Microsoft Corp.

Christopher Songer

unread,
Feb 14, 1993, 4:05:51 PM2/14/93
to
lem...@uiuc.edu writes:
>
>Although the NeXT may only have a 33 MHz cpu bus, innovations in the
>Turbo units made memory accesses much faster. The memory bus in the
>Turbo NeXTs is simply better engineered than most PC's.
>

Actually, NeXT's memory subsystem was not as zowie as all that.
While it was probably better than most PC's -- it needed to be --
the Next did not have a secondary cache and the 8K on board
cache was not that large.

It would be better to say that the "old" memory subsystem was
substandard by todays standards (the original being the same that
was on the "original Cube") and the Turbo subsystem brought Next
back in line with the market.

All of which is water over the bridge... I find this to be the most
positive move Next has made and am quite pleased Next is going to
software only. I own two NeXT's. I would much rather have two PC's
with Nextstep486. Why? Because a PC w/ NS486 could be booted into
another OS. Sure NS is cool, but it has some glaring software holes.
There will always be applications which are only available for some
"Other" OS. With Next's dead hardware, one could run Nextstep only.
With Next's new scheme one's Nextstep hardware can run Dos and OS/2
and NT and ...

The "new" Next hardware solution offers the consumer greater
flexibility andfaster hardware. Next has made the right decision.

-Chris

rpom...@aunext1.amd.com

unread,
Feb 14, 1993, 12:35:29 PM2/14/93
to
In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu
(Yamanari) writes:
> In article <1993Feb13....@cs.yale.edu>
nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
> >In article <1993Feb13.0...@ecsvax.uncecs.edu>
[munch munch]

>
>
> What subsystems might those be? The dumb frame buffer?
> The (guffaw) SCSI2?

Howabout the 9 channel DMA channel processor with 40 Mb/s bandwidth
per channel!! Oh yea, I've seen lots of PeeCees with ASICs like
that...not! ;-)

--
Ronald Pomeroy "NeXT has just gone virtual"
CAM Applications Group
Advanced Micro Devices
Austin, Texas
rpom...@aunext1.amd.com

rpom...@aunext1.amd.com

unread,
Feb 14, 1993, 12:36:14 PM2/14/93
to
In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu
(Yamanari) writes:
> In article <1993Feb13....@cs.yale.edu>
nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
> >In article <1993Feb13.0...@ecsvax.uncecs.edu>
[munch munch]

>
>
> What subsystems might those be? The dumb frame buffer?
> The (guffaw) SCSI2?

Howabout the 9 channel DMA channel processor with 40 Mb/s bandwidth

Peter Kron

unread,
Feb 15, 1993, 8:15:36 AM2/15/93
to
> In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu (Yamanari) writes:
> >In article <1993Feb13....@cs.yale.edu> nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
> >>In article <1993Feb13.0...@ecsvax.uncecs.edu>
> >>jfr...@ecsvax.uncecs.edu (Joe Freeman) writes:
> >>
> >>> The 486 box is faster than a turbo and faster than what would have been
> >>> the NeXT 680x0 box. They come with a Pentium socket (some do anyway)
> >>> which would make the box as fast as any "low cost" RISC box
> >>> NeXT would have introduced. Why would anyone want a RISC box?
> >>
> >>The 486 I've seen, a Dell 50 MHz model (true 50 MHz, not a 25/50
> >>DX2), did not feel nearly as fast (or good, but that's moot
> >>now) as our NeXTstation Color Turbo machines. It wasn't even
> >>close to the response, and it had the JAWS video card too.
> >
> [point-counterpoint deleted]

My impression at NextWorld 92 was that NS/Intel was not as fast as a
NextStation color. A colleague felt just the opposite. It's pretty
subjective at this stage.

I think the important issue is not an absolute comparison at this point
in time, but a look at the trends. In 4 years NeXT has gone from
25Mz 030 to 33Mz 040. PC's have gone from 16Mz 386 to 66Mz 486, with
Pentium around the corner. IBM is playing with 99Mz DX3 chips. I doubt
whatever advantage the black hardware may (or may not) have now will
be maintainable.

This is just raw speed. It doesn't measure speed/$ in which price
competition has also boosted the Intel value tremendously.

Disks and memory systems contribute measurably to performance, and I
don't claim to be able to analyze them. My experience has been though
that they fine tune performance of the processor. You won't beef up
the memory on a 386 to compete with Pentium. If so, the Intel line
presents a pretty bright future.

(RISC may have similar advantages, but nothing matches Intel for the
market economic factors.)
---------------
Peter Kron P.O. Box 51022
Corona Design, Inc. Seattle, WA 98115-1022
Peter...@corona.wa.com

Lars Peter Fischer

unread,
Feb 15, 1993, 5:31:07 PM2/15/93
to

>>>>> "Peter" == Peter Kron (pk...@corona.wa.com)

Peter> I think the important issue is not an absolute comparison at this point
Peter> in time, but a look at the trends. In 4 years NeXT has gone from
Peter> 25Mz 030 to 33Mz 040. PC's have gone from 16Mz 386 to 66Mz 486, with
Peter> Pentium around the corner. IBM is playing with 99Mz DX3 chips. I doubt
Peter> whatever advantage the black hardware may (or may not) have now will
Peter> be maintainable.

True.

The same hold when compared to other hardware. Consider Sun. When the
Cube was announced, it was pretty good compared to Sun's offering,
especially given the price. The slab was reasonable if not good
compared to the SPARCstations at the time of introduction. Now,
compare the Turbo to the SS Classic -- no, it simply does not hold up.

NeXT had to move on, one way or another. Either new hardware of their
own, with a new CPU family, or this. One way or another, it's pretty
much the same to those with black boxes. One might even guess that
NeXT would support the 040 more this way, out of nostalgica towards
"our own hardware".

Now, let's have NeXTSTEP/Classic.

/Lars
--
Lars Fischer, fis...@iesd.auc.dk | It takes an uncommon mind to think of
CS Dept., Aalborg Univ., DENMARK. | these things. -- Calvin

Matthew Dillon

unread,
Feb 21, 1993, 1:09:48 PM2/21/93
to
In article <86...@ut-emx.uucp> rpom...@aunext1.amd.com writes:
>In article <1993Feb13.0...@wam.umd.edu> rsro...@wam.umd.edu
>(Yamanari) writes:
>> In article <1993Feb13....@cs.yale.edu>
>nat...@laplace.csb.yale.edu (Nathan F. Janette) writes:
>> >In article <1993Feb13.0...@ecsvax.uncecs.edu>
>[munch munch]
>>
>>
>> What subsystems might those be? The dumb frame buffer?
>> The (guffaw) SCSI2?
>
>Howabout the 9 channel DMA channel processor with 40 Mb/s bandwidth
>per channel!! Oh yea, I've seen lots of PeeCees with ASICs like
>that...not! ;-)

Maybe not PC's, but the Amiga has at least 10... Four audio channels,
sprite, blitter, copper, floppy, video (which you can ignore if you
like), and SCSI. More if you count the sub-channels for the blitter,
add another if you have an ethernet board.

There is nothing special about having a lot of DMA channels. What
counts is what you get out of all those channels.

-Matt

>
>--
>Ronald Pomeroy "NeXT has just gone virtual"
>CAM Applications Group
>Advanced Micro Devices
>Austin, Texas
>rpom...@aunext1.amd.com

--

Matthew Dillon dil...@Overload.Berkeley.CA.US
1005 Apollo Way uunet.uu.net!overload!dillon
Incline Village, NV. 89451 ham: KC6LVW (no mail drop)
USA Sandel-Avery Engineering (702)831-8000

Matthew Dillon

unread,
Feb 21, 1993, 2:34:38 PM2/21/93
to
>In article <C2DGr...@ux1.cso.uiuc.edu> lem...@ux1.cso.uiuc.edu (David Lemson) writes:

>rsro...@wam.umd.edu (Yamanari) writes:
>
>> Wanna be a little more specific on how the NeXT memory
>> is somehow mysteriously better than that in a DX-50?
>Although the NeXT may only have a 33 MHz cpu bus, innovations in the
>Turbo units made memory accesses much faster. The memory bus in the
>Turbo NeXTs is simply better engineered than most PC's.

I truely wish I could agree with you, but unfortunately I cannot. Most
medium to high-end PC's have 128KB+ caches OR 0 wait state interleaved
memory. 486 Cache controllers are a commodity item and 4-way
set-associative chips are common.

The NeXT's memory subsystem may be better engineered, but you can't get
faster then 0 wait states and PC's are already there.

>great guy) He told me that video performance on a localbus DX2/66
>system is perhaps as much as twice as fast as the same video card on
>an EISA bus.

Anything on localbus will be faster then EISA because localbus is
really nothing more then an extension of the processor's pins. In
fact, you can theoretically reach 0 wait states by placing a cache with
write-buffering capabilities on the video board itself.

-Matt

>--
>David Lemson (217) 244-1205
>University of Illinois NeXT Campus Consultant / CCSO NeXT Lab System Admin
>Internet : lem...@uiuc.edu UUCP :...!uiucuxc!uiucux1!lemson
>NeXTMail & MIME accepted BITNET : LEMSON@UIUCVMD

--

Ron Christian x1545

unread,
Feb 24, 1993, 2:21:57 PM2/24/93
to
In article <1993Feb12....@wam.umd.edu> rsro...@wam.umd.edu (Yamanari) writes:
>> There is
>>turmoil within NeXT itself as all engineers start to wonder if they are
>>going to have a job.
>
> They can always go work for Sun like the others.

No kidding. Sun just announced a Display Postscript engine for the next
major release (two releases from now, or something like that) of their own
windowing system. I wonder where they got the engineers? :-)


Ron

0 new messages