[comp.sys.next] NeXTstation

pclark@SRC.Honeywell.COM (Peter Clark) (10/11/90)

I've been blessed with the opportunity to play with a demo NeXTStation for the
past two days, and felt I owed it to the net to summarize my experiences. This
will be somewhat stream-of-consciousness, so bear with me.

In a word- Outstanding! If this is the 'future of NeXT', NeXT should be pretty
well off. The '040 slab seems to be around 4 times as fast as the '030 cube
for cpu-intensive work (it can count to 1^7 4 times as fast)- the '040 manages
to stay ahead of the DSP in Mandelbrot for quite a while, and finishes way
ahead of the '030.

The 2.0 software is also nice. The Workspace manager is now the 'file viewer',
and it has a 'shelf' to put commonly-used directories and files. Clicking on
the shelf will jump you to the location of the file/directory in the file
system. It also lets you run things like copying & removing as background
processes- the workspace manager isn't tied up until the copy is finished.

All programs launch *much* faster.

The 3.5 IBM-Compatible drive really is. It shows up just like a directory in
your home directory, and you can open text files with Edit, & read & edit
them. The MS-DOS file system mounts flawlessly into the Unix file system.

Most 1.0 binaries work- NX_VOID and FrameMaker 2.0 work fine, but Scott Hess'
TimeMon program doesn't work. I don't know why.

Terminal & Shell have been replaced with a re-worked version of Stuart.
Congrats, Scott!

The Admin tools have been reworked, and they all seem to be better.

The Librarian is a *hell* of a lot faster, but it doesn't work with old index
files- be prepared to re-index all your non-standard stuff. It's worth it.

You can print to a fax-modem as well as the printer. It shows up in the print
panel, almost like a printer.

WriteNow does indeed now do underlining.

The Preferences app has a panel for choosing monitors. Evidently, you can
indeed have a multiply-headed cube.

There are a few new toys in the NextDeveloper/Demos directory- a Datebook, a
couple of fractal programs, the Gourmet calculator (mentioned in the NeXT on
campus magazine), a spiffy PendulumLab program (which does very smooth
animation), and a program called ShowAndTell, which records all the things you
do with the computer & can play them back (great for developing training
scripts).

The ScorePlayer program now has two types of score files- one which is
'compiled' or 'optimized' for playback, and the old text format. It seems to
be a lot better at playing without skipping or slowing down with the system is
loaded.

Improv is neat, but I haven't had enough time to get a feel for it. It looks
a lot better then Excel (the only spreadsheet I've used a lot).

PowerStep and WorkPerfect are there, and they work. I haven't played with them
much either, but they seem to work as advertised.

The 'Request' Menu has been renamed to Services, and it extensible. Various
applications can install menus in here- Mail shows up (i.e., mail the
selection), and there's been some others as well.

The IB has some new palettes, and a better Text object. It's supposedly easier
to add custom palettes now, but I haven't tried it.

The C compiler now supports both Objective-C and C++. All the libraries are
still in Obj-C. Emacs is ver. 18.55.

These are all the things that I can think of right now- I'm going to be sad
when I have to give it back to NeXT tomorrow, but I'm looking forward to
having my very own soon! Feel free to e-mail any questions that I might be
able to answer.


	Pete Clark
	
p.s. I have no affiliation with NeXT, other than as a satisfied and impressed
customer.

neil@ms.uky.edu (Neil Greene) (11/21/90)

Have the nextstations already been shipped?  Just curious, I have sen one and
used it for awhile, very impressed with it all the same, but wondering has
anyone got one?

blumoose@mrfreeze.MIT.EDU (David W Gotthold) (12/19/90)

Well, my station is finally here and installed.  Now I need to get 
serious about purchasing a large hard drive.  If anyone has a list
with sizes, prices, etc. I would appreciate if you could email it
(or post it).  Also any other lists of things a new NeXT owner should 
know.  Thanks in advance.

blumoose@athena.mit.edu

holmer@ucbarpa.Berkeley.EDU (Bruce K. Holmer) (01/18/91)

[]

I copied a number crunching application of mine from an '030 Next cube
to my new '040 Nextstation, and was shocked with the relatively poor
performance.  After some experimentation, I've found the cause---the
'040 floating point does not support in hardware the entire 68882
instruction set, so the unimplemented instructions must be done by
software.  That's fine with me, since Motorola assured us (IEEE Micro,
February 1990, p. 77) that:

	A software emulator of all unimplemented instruction si available
	from Motorola....  Execution time of the software emulation for
	elementary functions including all trap overhead (running on a
	25 MHz 68040) is 13 to 130 percent faster than the equivalent
	instructions on the 68882 running at 25 MHz.

However, whatever turned up in the Nextstation is certainly not what
Motorola was promising (whether the blame is Motorola's or Next's I
don't know).  For your amusement here is a small assembly language
program:
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% tmp.s %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
#NO_APP
gcc_compiled.:
.text
LC0:
	.even
.globl _main
_main:
	link a6,#0
	clrl d1
L61:
	fmovex #0r0.5,fp0
	fQQQx fp0,fp0

	addql #1,d1
	cmpl #999999,d1
	jle L61
	unlk a6
	rts
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Here are the timings (using /bin/time) for different QQQ's:

Next cube (25 MHz '030)
-----------------------

QQQ		user (sec.)	sys (sec.)

<none>		 1.8		 0.0		(instruction removed)
sqrt		 4.5		 0.1		(square root)
cos		13.9		 0.4		(cosine)
etox		19.8		 0.7		(e to the x power)
int		 2.8		 0.0		(integer part)
intrz		 2.8		 0.1		(integer part/round to zero)


Nextstation (25 MHz '040)
-------------------------

QQQ		user (sec.)	sys (sec.)

<none>		 0.4		 0.0		(instruction removed)
sqrt		 4.4		 0.0		(square root)
cos		 0.8		27.6		(cosine)
etox		 0.9		27.0		(e to the x power)
int		 0.9		82.9		(integer part)
intrz		 1.0		81.9		(integer part/round to zero)




Note that sqrt is implemented in hardware on the '040 (I threw it in
for a reality check).  Also, I ran each once, so I didn't average out
the variations in the timing.

However, the numbers do make the point that the emulation is done at
great expense on the Nextstation.  The real shock is the integer to
float conversion (2000 cycles!).  That's the one that hurt my application.

I do not know if the Nextstep 2.0 C compiler still uses fetoxx,
fintrzx, etc.  (it may use subroutine calls to faster emulation
software), but my alarm is still valid for programs that are copied
over as binaries or Sun executables that are converted using atom.

Can someone clarify this situation?  Will it be fixed soon?

--Bruce Holmer
holmer@ucbarpa.berkeley.edu

melling@cs.psu.edu (Michael D Mellinger) (01/19/91)

In article <1693@marlin.NOSC.MIL> aburto@marlin.NOSC.MIL (Alfred A. Aburto) writes:

   Could you please run the following program on the 040.  It is a very
   simple program but numerous folks over in 'comp.benchmarks' have run
   it on their systems and it would be interesting to see where the 
   NEXT 040 fits.   

   Perhaps NeXt did not use the Motorola math function library ?  The math
   function library from Motorola I am assured is faster than a 33 MHz
   68882.  The FP operations like +,-,*,/ are many times faster than the
   68882 (I understand but I am not sure).

handel> cc -O bench.c
handel> time a.out
r = 16.695311
RunTime = 26.133333
26.137u 0.171s 0:26.78 98% 0+0k 0+0io 0pf+0w

-Mike

datran2 (01/20/91)

In article <1693@marlin.NOSC.MIL> aburto@marlin.nosc.mil.UUCP (Alfred A. Aburto) writes:
>
>Could you please run the following program on the 040.  It is a very
>simple program but numerous folks over in 'comp.benchmarks' have run
>it on their systems and it would be interesting to see where the 
>NEXT 040 fits.   

I compiled and ran it on a relatively quiet 68040 NeXT.

datran2% cc bench1.c
datran2% a.out
r = 16.695311
RunTime = 29.116667

I haven't been following comp.benchmarks, so I don't know how this fits
in.  Would you enlighten me.

Steve.

-- 
 #====#====#====#====#====#====#====#====#====#====#====#====#====#====#====#
 #  Steve Boker           #  "Badgers, we don't have no stinking badgers"   #
 #  datran2!smb@csn.org   #    -from Treasure of the Sierra Madre Zoo       #
 #====#====#====#====#====#====#====#====#====#====#====#====#====#====#====#

masaru@media-lab.MEDIA.MIT.EDU (Masaru Sugai) (02/10/91)

 This might be one of FAQs, but I'm so curious about NeXTstation I just got
yesterday after waiting nearly over 4 months.

 - How can I get Mathematica ? I took it granted for customers through higher 
educatinal resellers, or they branded me as an incapable guy ? :) I read RSVP
but I couldn't find any words on Mathematica. And several demos assume the 
installation of it. 
 
 - NeXTstep snapped and restarted all of sudden while I opened up several 
demos including Webster, Librarian, SoundPlayer, and so on. I don't remember
the exact situation, so I cannot redo the same sequence again, but it happended 
twice in a few hours. One of the problems seems to have something to do with
SoundPlayer, as I experienced unusual slow responce when I clicked on its 
buttons.

  I am satisfied with NeXTstation due to quick responce and rich documentation.
Any tips are appreciated.  Thanks in advance.

-- Masaru Sugai