[comp.sys.nsc.32k] AUTHOR SPEAKS: '532 Manifesto

scwilk@sdrc.UUCP (Ken_Wilkinson) (11/16/88)

   HOLY MACHEREL BATMAN!

   Six or seven months ago I started using Usenet and found to my
   pleasent surprise a neat newgroup devoted to the NSC procressor.
   At that time, a discussion developed around a home-brew '532 based
   system.  To try and settle the noise abit, I suggested the now
   famous :-) '532 Manifesto.   UNFORTUNATLY I had to drop put of the 
   discussion due to work demands.  So imagine my surprise, when this
   week I received a mail message from someone willing to help with 
   the project!  Puzzled, I logged on and started reading the mail.

   YIKES! I was as if I hadn't left!
    
   Message after message about the '532 appears and cheap systems based
   on the '532 desired.

   So I'll dust off the 'ol '532 Manifesto...and...

   Herein is the last version from some months ago...
   ---------------------------------------------------------------------
   
   Hello Folks!

     Since my last posting I have recieved some very interesting
   responses.  It now looks like we have sufficient workers and
   a potential person to do a port (System V though...). I have 
   collected the votes and thoughts of responders and have formulated
   "THE '532 MANIFESTO" which is enclosed in draft form.  If some of
   the words seem familiar don't be alarmed, I stole them from your
   e-mail :-}.



>In article <235@sdrc.UUCP> I wrote:
>     o   IBM PC/AT  system board replacement, uses present PC cheap 
>         and availble parts for enclosure , HD and graphics contrlr etc.
>
>     o   Classic SBC design with SCSI on board memory. Uses a terminal.
>
>     o   Above with AT type (EGA?) graphics controller chip set.
>
>     o   design a SBC but have slots for expansion.
>
>

      8 -   votes for IBM PC/AT system board replacement.
      2 -   votes for SBC with SCSI and terminal.
      1 -   vote for SBC with EGA graphics chips.
      1 -   vote for SBC to place into old Macintosh case.
      1-    vote passive AT style bus (ala Zenith) with plug-in cpu.
 
>Software:
>
>     o   REAL UN*X (BSD, SYS V) 
>
>     o   MINIX                      (yecchhh, sorry :) )
>
>     o   MACH
>
>     o   Other PD OS                (check out os.research)


      7  - votes for BSD Un*x.
      1  - vote for Sys V.
      1  - vote for Minix
      1  - vote for GNU   

      Not everyone voted for an OS, (hence the numbers don't add up)
      Gee guys, some of you just program to bare metal? :-).
     
> 
>     I have a friend who is a dedicated hacker and owns a business
>     with the ablility to make PCB's from schematics.  I would be 
>     willing to help translate a schematic to artwork if others are
>     will to help design it.  Any takers?                  
> 

YES! Enough hardware and software have offered.




   -DRAFT DOCUMENT--DRAFT DOCUMENT--DRAFT DOCUMENT--DRAFT DOCUMENT-


                    The  '532 Manifesto
                (Appologies to R. Stallman)


Purpose:

          Design and build a computer based on the NSC '532 chip set.
   The computer will use presently available mass-produced componets to
   keep cost as low as possible.  Exploit cheap available finished IBM bus
   I/O boards, to take advantage of everybody else's volume.  

           Design will be a modified KISS approach (KISS Plus),
            "Keep It Simple Stupid" but as fast as possible!


CPU Design:

          Based upon a hardware compatable PC/AT motherboard (size,and
   mounting hardware NOT CPU!) to allow use of the IBM type AT enclosures
   and power supply.


Features:

 o on board SCSI bus for hard disk and tape drive support. 

 o IBM PC 101 keyboard connection

 o 4 to 8 IBM PC periphial slots half XT and half AT style. (depends on 
   available board area).

 o uses at a minimum a Hercules clone display adapter. (at $41 cheaper than
   a terminal. CRT about $82, keyboard $61. Total $184).  Later, one could
   use one of the Super VGA type cards with a NEC M.S. II for X-windows,
   PHIGS (which I'm working on now), etc.

 o Ethernet , parallel ports, serial ports would use the bus. We might if
   enough room put the serial and parallel ports on the motherboard.

 o memory will be MAC II style SIMM chips (the fastest we can buy :-})
   for 4 to 8 megs (maybe 4 meg with a daughter board for 8 megs)
   NO memory expansion bus. If you need more than 8 megs tuff sh*t.



Operating System:

 SYS V    - First choice.
 Minix    - Possible, maybe a parallel effort?  Race?

 BSD was the choice of most respondants, but no one has come forward to 
 do the port.


Concerns:

 o  Do we use instruction cache? Is there enough room on the cpu motherboard?
    Do we really need it, as the 532 has a burst mode.
    (no)

 o  use same serial uarts as ICM and preserve some of the driver code?
    (no)

 o  how much memory can we fit?
    ( 8 megs max easily, but design for 1 meg min.)

 o  bring out internal bus lines to a connector unbuffered?
    (nahhhh)

 o  Clock speed, multi layers 4,6,8? Cost?
    (?)

 o  Memory speed, wait states.
    (?)

 -END OF DRAFT--END OF DRAFT--END OF DRAFT--END OF DRAFT--END OF DRAFT-


 Please comment by e-mail on the concerns and design. I will firm up 
 "THE '532 MANIFESTO" and then we can begin the real work!

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


 SOOOOO, there it is again.  Still interested?  Changes? 
 A few very qualified people have contacted me to help and I must
 applogise that I have not followed up to their mail. If these folks
 are still interested (I am!) and NSC is still willing to help we may 
 yet bring this beast to light.





 Ken Wilkinson                            (uunet!sdrc!scwilk)
 SDRC
 2000 Eastman Dr.
 Milford, Ohio  45014 (Cincinnati)
 (513)-576-2569



 Above words are mine and others not the company's. Not responsible
 for anything you might think of now or ever.

"We now have the ultimate in Office Automation, all six printers are down"
                                                           Ken W.

dtynan@sultra.UUCP (Der Tynan) (11/17/88)

In article <433@sdrc.UUCP>, scwilk@sdrc.UUCP (Ken_Wilkinson) writes:
> 
>  o Ethernet , parallel ports, serial ports would use the bus. We might if
>    enough room put the serial and parallel ports on the motherboard.

Nah!  The best way to handle serial ports is with a separate processor.  Let's
keep the '532 for *real work*, and let some pissant 68K handle the serial
lines :-)

>  o memory will be MAC II style SIMM chips (the fastest we can buy :-})
>    for 4 to 8 megs (maybe 4 meg with a daughter board for 8 megs)
>    NO memory expansion bus. If you need more than 8 megs tuff sh*t.

Hmm.  I wonder.  Right now 4MB is loads, but what about next year?  Why not
make it so that one can use 1Mbx9 SIMMS or 256Kx9 SIMMS depending on choice
(read budget).  With 1MB SIMMS, we *should* be able to put 16 meg. on the
board.  Don't forget also, that the unfortunate (?) thing about a 32-bit
bus, is you have to buy SIMMs four-at-a-time.  Another concern, here, is
parity.  What is the overall concensus?  Yay or nay?  I don't even want
to *think* about such budget killers as ECC.

> Operating System:
> 
>  SYS V    - First choice.
>  Minix    - Possible, maybe a parallel effort?  Race?
> 
>  BSD was the choice of most respondants, but no one has come forward to 
>  do the port.

If you use a '532 as a CPU, then Minix is a dead issue here (sorry, AST).
Minix was never designed for something like that.  Adding all the code to
handle a screamer like the '532, would mean major work to Minix.  Better to
spend the time working on a BSD port.  Yes, I said BSD.  I writing this
under SysV.  PLEEZ PLEEZ PLEEZ, don't use SysV!!!  It may be just dandy for
people who want the stability (?) of AT&T, but they buy IBM anyway.  The
only reasonable SysV will probably be 4.0, which ain't available.  Moreover,
4MB won't hack it (probably).  I'll do the port (with help, of course :-).
The only problem is, I don't have a source licence.  The way I see it, there's
the same problem  with SysV.  I don't think we can collectively put together
~$75K for a source-licence either, so unless we can find a PD version, or
someone willing to allow us to use *their* licence, then we're stuck with
Minix.  Comments anyone?

>  o  Do we use instruction cache? Is there enough room on the cpu motherboard?
>     Do we really need it, as the 532 has a burst mode.
>     (no)

Assuming something like a 20MHz '532, then the clock cycle is 50ns.  Hmm.  The
memory cycle will probably be something close.  I'm not about to buy 4Mb of
50ns SIMMS!.  Where's Steve Wilson when ya need him?  How about an *optional*
64K cache?  Too much?  Oh well.

>  o  use same serial uarts as ICM and preserve some of the driver code?
>     (no)

Why not?  As I understand it, the ICM uses 16450's.  These are *nice* chips.
They even have a version with a fifo.  What's more, NSC makes them (does the
word 'discount' sound familiar?).  What did you have in mind instead? The
6850?  The nice thing about the '450 (not counting FULL modem control, and
interrupts on "change-of-status"), is the internal baud-rate generator.

>  o  how much memory can we fit?
>     ( 8 megs max easily, but design for 1 meg min.)

See above.  Don't forget, the GCC binary is >512K.  What's the OS going to
burn?  Don't forget too, that TCP/IP is now available (in SOURCE).  What
does that mean?  Twice as much memory.

>  o  bring out internal bus lines to a connector unbuffered?
>     (nahhhh)

If there are PC-bus slots, to heck with it.  HOWEVER, let's be a little more
far-reaching than the great Blue Engineer.  Make the *address* bus
tri-stateable.  The world of co-processors would be a *lot* better today, if
IBM could think.  With all that memory on the motherboard, you don't want to
have to wait for a slow old DMA.  Do it yourself.

>  o  Clock speed, multi layers 4,6,8? Cost?
>     (?)

Somehow, a CPU speed of 20MHz, sounds nice.  It has a nice ring to it (I HOPE
NOT :-).  I would see the design as "three-tiered".  A *fast* CPU, with cache.
A not-so-fast main memory (120 -> 150ns), and a dead-slow PC/AT-bus.  I would
like to see the thing confined to six layers, but I'm not sure that is
possible.  With some parts running at 40MHz, there *have* to be power & ground
planes.

>  o  Memory speed, wait states.
>     (?)

Go with 120 -> 150ns SIMM's.  I think this is probably the best price/perfor-
mance ratio.

>  Please comment by e-mail on the concerns and design. I will firm up 

Using email unfortunately, closes the subject to debate.  I don't think the
net bandwidth is wasted, when you consider we're plotting the overthrow of
IBM, Apple and Sun in one fell swoop :-)

>  "THE '532 MANIFESTO" and then we can begin the real work!
> 
>  Ken Wilkinson                            (uunet!sdrc!scwilk)

The real work has already begun.  The actual architecture discussion is one
of the most important considerations.  I mean, you don't want to be two years
down the road thinking "I wish I had more memory".

The thought strikes me, that there may be too many chiefs at this stage.  I
have been collecting names of interested people, as have others.  What we
need here, is a little "Law and Order".  Anyone want to volunteer as the
"Gatekeeper".  On a related point, in my estimation we need 20 hard-core
check-book in-hand people, before any of this is reasonable.  Of course, more
is good too :-)  What is the "head-count" right now?  Anyone know?
						- Der
PS:
OK, one last point.  Does anyone have a nice, catchy name for such a system?
I'm sick of calling it the '532 el-cheapo-thing'.  How about Merlin?
-- 
	dtynan@zorba.Tynan.COM  (Dermot Tynan @ Tynan Computers)
	{apple,mips,pyramid,uunet}!Tynan.COM!dtynan

 ---  If the Law is for the People, then why do we need Lawyers? ---

bga@raspail.UUCP (Bruce Albrecht) (11/18/88)

In article <2659@sultra.UUCP>, dtynan@sultra.UUCP (Der Tynan) writes:
> The thought strikes me, that there may be too many chiefs at this stage.  I
> have been collecting names of interested people, as have others.  What we
> need here, is a little "Law and Order".  Anyone want to volunteer as the
> "Gatekeeper".  On a related point, in my estimation we need 20 hard-core
> check-book in-hand people, before any of this is reasonable.  Of course, more
> is good too :-)  What is the "head-count" right now?  Anyone know?

I'm not quite ready to count myself as a checkbook-in-hand person just yet, but
I might be, if someone can provide a rough estimate of what it will cost to
build one.

stevew@nsc.nsc.com (Steve Wilson) (11/18/88)

In article <2659@sultra.UUCP> dtynan@sultra.UUCP (Der Tynan) writes:
>Assuming something like a 20MHz '532, then the clock cycle is 50ns.  Hmm.  The
>memory cycle will probably be something close.  I'm not about to buy 4Mb of
>50ns SIMMS!.  Where's Steve Wilson when ya need him?  How about an *optional*
>64K cache?  Too much?  Oh well.

Huh, Whaa..Who Me! Some one call my name?.....Guess not, guess I'll
go back to sleep.  ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ

Seriously,  you should be able to put an external cache on the 532 at 20Mhz
with no great shakes.  Just going though the numbers in my head I'd guess
you would have to use 50ns static SRAMS for the cache at 20Mhz. I'd 
think it might be easier to build a Page mode or Static Column based
DRAM system that supported burst. This might be a useful alternative
to a cache AND drams.  A local gentleman named George Scolaro was 
yelling to me yesterday about such a fantasy.  (Oh my, another George! ;-)
Hey Mr. Scolaro, do you read this net?

Steve Wilson
32K Architecture Group
National Semiconductor, Santa Clara
 
 
 

rfg@nsc.nsc.com (Ron Guilmette) (11/18/88)

In article <2659@sultra.UUCP> dtynan@sultra.UUCP (Der Tynan) writes:
>... so unless we can find a PD version, or
>someone willing to allow us to use *their* licence, then we're stuck with
>Minix.  Comments anyone?

Well, there already a port of XINU to the 3200's series available.  I have it
here in my hot little hands.  Actually, I think that the best thing to do
would be to use the XINU kernel, and then for the utilities use the sum of
MINIX, XINU, and GNU, taking the best pieces where there are overlaps.

>OK, one last point.  Does anyone have a nice, catchy name for such a system?
>I'm sick of calling it the '532 el-cheapo-thing'.  How about Merlin?

I suggest picking a non-propritary name that won't get you into trouble with
anybody.  AT&T already uses the name Merlin for some of its business phone
systems.  How about the legendary "foobar"?  If you were to copyright that
then in 6 months you could sue everybody on the net over their examples ;-)

-- 
Ron Guilmette
National SemiConductor, 1135 Kern Ave. M/S 7C-266; Sunnyvale, CA 94086
Internet: rfg@nsc.nsc.com   or   amdahl!nsc!rfg@ames.arc.nasa.gov
Uucp: ...{pyramid,sun,amdahl,apple}!nsc!rfg

adh@anumb.UUCP (a.d.hay) (11/18/88)

In article <2659@sultra.UUCP> dtynan@sultra.UUCP (Der Tynan) writes:
[]
-->PS:
-->OK, one last point.  Does anyone have a nice, catchy name for such a system?
-->I'm sick of calling it the '532 el-cheapo-thing'.  How about Merlin?

Swift, Thunderbolt, Mercury, Bullet, Cheetah

-- 
Andrew Hay		+------------------------------------------------------+
Holistic Specialist	| I will design a computer for you, so powerful that   |
AT&T-BL Ward Hill MA	| organic life will form part of its operational matrix|
mvuxq.att.com!adh	+------------------------------------------------------+

jack@cwi.nl (Jack Jansen) (11/18/88)

In article <7885@nsc.nsc.com> rfg@nsc.nsc.com.UUCP (Ron Guilmette) writes:
>
>Well, there already a port of XINU to the 3200's series available.  I have it
>here in my hot little hands.  Actually, I think that the best thing to do
>would be to use the XINU kernel, and then for the utilities use the sum of
>MINIX, XINU, and GNU, taking the best pieces where there are overlaps.

Uhm.... Either I'm missing something, or somebody else is.
Two points:
- Minix is a complete, full-blown operating system. I would not
  be surprised if it ran *faster* than either BSD or sysV.
  The only drawbacks are that it doesn't have the fast filesystem
  (but neither has sysV) and  that it doesn't have VM (but that makes
  it faster, and who needs VM anyway:-)
- XINU, on the other hand, was a toy operating system the last time
  I looked. Actually, the term 'monitor' was probably better than
  'operating system'. Fine for teaching classes, not something you
  would want to do serious work on.
  (Note however that I've only looked at the old PDP-11 XINU. Ignore
  this comment if it has substantially improved since then).
--
Fight war, not wars			| Jack Jansen, jack@cwi.nl
Destroy power, not people! -- Crass	| (or mcvax!jack)

mlewis@unocss.UUCP (Marcus S. Lewis) (11/22/88)

In article <392@anumb.UUCP>, adh@anumb.UUCP (a.d.hay) writes:
> In article <2659@sultra.UUCP> dtynan@sultra.UUCP (Der Tynan) writes:
> []
> -->PS:
> -->OK, one last point.  Does anyone have a nice, catchy name for such a system?
> -->I'm sick of calling it the '532 el-cheapo-thing'.  How about Merlin?
> 

Some discussion in the newsletter brought about a selection of names for the
PD OS we are attempting to put together.  My favorite was Visigoth, so named
for the guys that wasted Rome, and guess who is the roman empire here?

We haven't actually settled on the name, so Visigoth is still open.

Marc Lewis

jpdres10@usl-pc.usl.edu (Green Eric Lee) (11/22/88)

All those people advocating Sys V and BSD4.x are forgetting one thing:
COST. Source licenses ain't cheap, folks, and I sure the hell don't
want to have an experimental homebrew-style system WITHOUT source. 

Someone mentioned that Comer's Xinu kernal was running on the 32K.
Note that Xinu is ONLY a kernal, no utilities (he assumed that you'd
just compile standard Unix utilities onto it, apparently).

On the other hand, someone else mentioned Minix. Minix includes all
the utilities necessary to do minimal Unix, but its kernal is very
eighty-eighty-sux specific and quite minimal (does no paging,
implements subset of V7, etc.). It would need a LOT of work to run on
a high-end 32K system.

But: has anybody ever thought of combining the two?

Don't ask me, I own an Amiga (the only multitasking computer with a
window system that I can buy for under $1k... all that poor college
student me can afford).

===
Eric Lee Green                            P.O. Box 92191, Lafayette, LA 70509
     {ames,mit-eddie,osu-cis,...}!killer!elg, killer!usl!elg, etc.

dtynan@sultra.UUCP (Der Tynan) (11/22/88)

In article <7718@boring.cwi.nl>, jack@cwi.nl (Jack Jansen) writes:
> In article <7885@nsc.nsc.com> rfg@nsc.nsc.com.UUCP (Ron Guilmette) writes:
> 
> Uhm.... Either I'm missing something, or somebody else is.
> - Minix is a complete, full-blown operating system. I would not
>   be surprised if it ran *faster* than either BSD or sysV.
>   The only drawbacks are that it doesn't have the fast filesystem
>   (but neither has sysV) and  that it doesn't have VM (but that makes
>   it faster, and who needs VM anyway:-)

I think you forgot to add all the smileys.  In fact, I find this one of the
funniest postings in a while.  What you should realize, is some of us out
there are actually running Minix.  As for it being complete, and full-blown,
is indeed laughable.  Minix is great if you want to learn about OS's, however,
if you want to do 'real work', forget it.  My system crashes on the average
of twice a day.  Most software off the net won't work with it, unless seriously
modified, and then if it does, it will usually crash the system or die.  I've
been running PD uucp implementations for a while now, and for the most part,
I get a good connection once out of every ten tries.  Also, it *never* has
enough room, and so it bounces mail left, right and centre, when the fork()
fails.  I'm not trying to put down Minix, but don't try and tell us it is
complete.  As for it running faster than BSD or SysV, again this is amusing.
Obviously, you've never tried unsharing something.  This kind of activity
can literally take hours.  Compiles are similar.  The fact that it doesn't
support swapping or paging is a bug, not a feature.  It doesn't even support
shared-text.  As for who needs VM?  Anyone trying to get 'real work' done.
I've had this argument with people around here, for a while now.  One thing
we *do* agree on, is that Minix is not for serious work.  I would suggest,
that rather than going to a '532 system, you stick with the '386, because
judging by your posting, you're looking for a 'faster version of Minix',
not something to do real computing.  BTW;  I'm not typing this from the
Minix machine.  That's why you're seeing this posting :-)

> - XINU, on the other hand, was a toy operating system the last time
>   I looked. Actually, the term 'monitor' was probably better than
>   'operating system'. Fine for teaching classes, not something you
>   would want to do serious work on.
>   (Note however that I've only looked at the old PDP-11 XINU. Ignore
>   this comment if it has substantially improved since then).

> Fight war, not wars			| Jack Jansen, jack@cwi.nl

Perhaps you got Minix and XINU mixed up, here.  This comment belongs with
the above :-)
						- Der
-- 
	dtynan@zorba.Tynan.COM  (Dermot Tynan @ Tynan Computers)
	{apple,mips,pyramid,uunet}!Tynan.COM!dtynan

 ---  If the Law is for the People, then why do we need Lawyers? ---

dean@mars.Berkeley.EDU (R. Drew Dean) (11/23/88)

In article <115@usl-pc.usl.edu> elg@killer.UUCP writes:
>On the other hand, someone else mentioned Minix. Minix includes all
>the utilities necessary to do minimal Unix, but its kernal is very
>eighty-eighty-sux specific and quite minimal (does no paging,
>implements subset of V7, etc.). It would need a LOT of work to run on
>a high-end 32K system.
>
>===
>Eric Lee Green                            P.O. Box 92191, Lafayette, LA 70509
>     {ames,mit-eddie,osu-cis,...}!killer!elg, killer!usl!elg, etc.

Uh...I guess you haven't been following the net too closely....
A few weeks ago Andrew Tannenbaum (sp?), the author of Minix, announced the
avaiability of Minix on the Atari-ST, (for those who don't know), a 68000
machine....As for virtual memory, I think that an MMU to provide hardware
memory protection is needed, but with 4Mb RAM on a _single_-user system,
I should hope that there wouldn't be TOO much paging going on, actually
there really shouldn't be much of ANY paging....(Of course some really
large jobs will page, and if you're running Lisp under X Windows (anyone
want to volunteer to do the port ? :-) (100 MB of source + docs for X11R3), 
and Macsyma at the same time, you're going to be close to thrashing ....)

Minix remains an alternative, but I'd rather see 4.3 BSD...(But look where
I'm coming from...:-))

Drew Dean
dean@xcssun.berkeley.edu

dtynan@sultra.UUCP (Der Tynan) (11/23/88)

In article <511@unocss.UUCP>, mlewis@unocss.UUCP (Marcus S. Lewis) writes:
> 
> We haven't actually settled on the name, so Visigoth is still open.
> 
> Marc Lewis

OK, I'll vote for that.  Only one problem though, it needs a 'series number'.
I mean, "Visigoth" is OK and everything, but how about the "Visigoth I".
In fact, why not start out ahead of the bunch, and call it the Visigoth IV?
						- Der
-- 
	dtynan@zorba.Tynan.COM  (Dermot Tynan @ Tynan Computers)
	{apple,mips,pyramid,uunet}!Tynan.COM!dtynan

 ---  If the Law is for the People, then why do we need Lawyers? ---

dclemans.falcon@mntgfx.mentor.com (Dave Clemans) (11/29/88)

From article <115@usl-pc.usl.edu>, by jpdres10@usl-pc.usl.edu (Green Eric Lee):
> On the other hand, someone else mentioned Minix. Minix includes all
> the utilities necessary to do minimal Unix, but its kernal is very
> eighty-eighty-sux specific and quite minimal (does no paging,
> implements subset of V7, etc.). It would need a LOT of work to run on
> a high-end 32K system.
> 

Just a side note: Minix is now available from Prentice-Hall for the
Atari ST (a mc68000 system); thus there is a starting point that doesn't
have built-in Intel dependencies.

dgc