From owner-riscy@pyramid.com  Tue Aug  3 02:54:17 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA27815; Tue, 3 Aug 93 02:54:17 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA29775; Mon, 2 Aug 93 23:26:46 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA19733; Mon, 2 Aug 93 23:22:59 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA19725; Mon, 2 Aug 93 23:22:56 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA29387; Mon, 2 Aug 93 23:23:47 -0700
Message-Id: <9308030623.AA29387@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA21843; Tue, 3 Aug 93 02:22:09 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Mips project Go or Wait?
To: riscy@pyramid.com (Mips 3000)
Date: Tue, 3 Aug 1993 02:22:09 -0500 (EDT)
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 1292      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Seems like all we need to decide on what to do about the project is
the price of the 4000 PC and the 4200 PC to decide on which cpu to use,
and the price of parts for the arc 100/150 kits.

I heard the the approximate price of the 4000 PC is about $300.  But I don't
have any distributor contacts to check on a price.

Assuming us and Waldorf can get 50 board purchases together seems like a
good guess would be.  

$100		$5000 for the production kits, cpu's, and plans / 50 boards
		(can we get the gerber files for free as mentioned on the list?)
$300		mips 4000 PC or less for a 4200
$100		appropiate size motherboard, drilled etc.
$300		vram, sims slots, support chips (MCT-ADR, 2*MCT-DP), scsi
		support, video support, etc.
----------------------------------
$800  for 64 bit 50 Mhz external 100 Mhz internal workstation on a board.

Hopefully it could be alot cheaper.

If we can pin these numbers down we can make a decision an send our checks
to Neil, or Waldorf ,and start the process of getting the boards.

I hate seeing no discusion on the list, I'd hate to see it die.

-- 
Bill					1st>	Broadley@neurocog.lrdc.pitt.edu
Broadley@schneider3.lrdc.pitt.edu <2nd 	3rd> 	             Broadley+@pitt.edu
Linux is great.         Bike to live, live to bike.                      PGP-ok
 
From owner-riscy@pyramid.com  Tue Aug  3 04:08:34 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA29640; Tue, 3 Aug 93 04:08:34 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA11858; Tue, 3 Aug 93 01:07:30 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA27126; Tue, 3 Aug 93 01:06:19 -0700
Received: from bert.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA27118; Tue, 3 Aug 93 01:06:15 -0700
Received: from mail.Germany.EU.net 
	by bertha.pyramid.com (5.67/OSx5.1a Pyramid-Internet-Gateway)
	id AA00783; Tue, 3 Aug 93 08:06:06 GMT
Received: by mail.Germany.EU.net(EUnetD-2.3.0.g) via EUnet
	id gE17096; Tue, 3 Aug 1993 10:01:37 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA07295; Tue, 3 Aug 93 09:54:08 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA24346; Tue, 3 Aug 93 09:56:35 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Tue, 3 Aug 93 09:55:37 +0200
Message-Id: <9308030755.AA04886@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA04886; Tue, 3 Aug 93 09:55:37 +0200
To: riscy@pyramid.com
Subject: Re:  Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


> Seems like all we need to decide on what to do about the project is
> the price of the 4000 PC and the 4200 PC to decide on which cpu to use,
> and the price of parts for the arc 100/150 kits.

Yes, that's right, but it seems to be *very* difficult. I never had
problems to get prices from IDT, but dealing with NEC is a hard, hard,
hard job. I have still no answer to my question wether a production kit
is available now or not, nor could they tell me prices. All I know is that
NEC works on an application note, which is currently available in japanese
only. However, I'm working on getting more informations, and it would be
very helpful if someone can tell me from where the information about the
production kit came. The more information *I* have, the more I can make
the NEC people work for me !

> I heard the the approximate price of the 4000 PC is about $300.  But I don't
> have any distributor contacts to check on a price.

I think this price is a bit too optimistic, at least for now. A few weeks
ago I got a price of around us$600 for a R4000PC, but this *will* change
as soon the Orion and the R4200/VRX is available and well-known.
IDT suggested me to use the Orion, because it's pin-compatibel with
the R4000PC. That means that we could start the development with R4000 CPUs
and later plug in the Orion. Assuming that it will take some months
to get the hardware working, this might be worth a discussion. Only
a few people would need to pay for the R4000, and all others can (or will)
wait for the Orion.
However, the NEC VRX might be good choice too, but we need more informations.

> Assuming us and Waldorf can get 50 board purchases together seems like a
> good guess would be.  

Here's the corrected guess.

$100		$5000 for the production kits, cpu's, and plans / 50 boards
 		(can we get the gerber files for free as mentioned on the list?)
$600		mips 4000 PC or less for a 4200
$100		appropiate size motherboard, drilled etc.
$300		vram, sims slots, support chips (MCT-ADR, 2*MCT-DP), scsi
 		support, video support, etc.
----------------------------------
$1100  for 64 bit 50 Mhz external 100 Mhz internal workstation on a board.

or
$700   for a NEC VRX solution (estimated VRX price: $200)
or
$700   for an IDT Orion solution (estimated Orion price: $200)

> Hopefully it could be alot cheaper.

Hmm. I don't think that we can make it cheaper than $700.
That would be a miracle :-)

> If we can pin these numbers down we can make a decision an send our checks
> to Neil, or Waldorf ,and start the process of getting the boards.

> I hate seeing no discusion on the list, I'd hate to see it die.

Yes, this is what I feel. I understand that people are confused by
the discussion about what Mips-CPU we should use. But I have the
feeling that we come closer to a decision. We have currently three
choices:

- R3081/R3041
- NEC R4200 and ARCset
- IDT Orion and ARCset

I hope the discussion will awake again !

Bye,
Andy

-------------------------------------------------------------------------------
        Waldorf Electronics GmbH        | Phone:  +49 (0)2636-80294
              R&D Department            | Fax:    +49 (0)2636-80188
Neustrasse 9-12, 53498 Waldorf, Germany | email:  andy@resi.waldorf-gmbh.de
-------------------------------------------------------------------------------
 
From owner-riscy@pyramid.com  Tue Aug  3 04:44:41 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA00663; Tue, 3 Aug 93 04:44:41 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16718; Tue, 3 Aug 93 01:43:39 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA01789; Tue, 3 Aug 93 01:42:32 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA01722; Tue, 3 Aug 93 01:42:24 -0700
Received: from hp750a.csc.cuhk.hk 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16655; Tue, 3 Aug 93 01:42:44 -0700
Message-Id: <9308030842.AA16655@gossip.pyramid.com>
Received: by hp750a.csc.cuhk.hk
	(16.8/16.2) id AA22371; Tue, 3 Aug 93 16:40:55 +0800
From: Stephen S M Wong <a080700@hp750a.csc.cuhk.hk>
Subject: Re:  Mips project Go or Wait?
To: riscy@pyramid.com
Date: Tue, 3 Aug 93 16:40:54 HKT
In-Reply-To: <9308030755.AA04886@resi.waldorf-gmbh.de>; from "Andreas Busse" at Aug 3, 93 9:55 am
Mailer: Elm [revision: 70.30]
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Hi risciers,

>- R3081/R3041
>- NEC R4200 and ARCset
>- IDT Orion and ARCset

My personal preference is IDT Orion, since

1) It is R4000PC pin compatible.

2) The price of Orion will be similar to NEC R4200, and R4000PC will
   out perform R4200, can we deduce that Orion will out perform R4200?

3) Even point 2) may not hold, a superior upgrade path exists if we
   use Orion, ie. replace the CPU on the to be made board by a R4000PC, 
   and get better (at least better than R4200) performance.  WITH THE
   SAME PRICE (on CPU), why not opt for better upgrade path!

--
*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*
v Stephen Wong Siu-ming    v internet: stephenwong@cuhk.hk  v
^ Computer Services Center ^ bitnet  : a080700@cucsc.bitnet ^
* Chinese University of    * phone   : (852) 609-8904       *
v Hong Kong                v fax     : (852) 603-5001       v
*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*
 
From owner-riscy@pyramid.com  Tue Aug  3 08:35:13 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA06850; Tue, 3 Aug 93 08:35:13 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15659; Tue, 3 Aug 93 05:33:39 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA21083; Tue, 3 Aug 93 05:32:26 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA21076; Tue, 3 Aug 93 05:32:23 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15644; Tue, 3 Aug 93 05:33:13 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.g) via EUnet
	id gM29669; Tue, 3 Aug 1993 14:29:06 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA09205; Tue, 3 Aug 93 14:22:25 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA25564; Tue, 3 Aug 93 14:22:24 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Tue, 3 Aug 93 14:21:23 +0200
Message-Id: <9308031221.AA05562@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA05562; Tue, 3 Aug 93 14:21:23 +0200
To: riscy@pyramid.com
Subject: Re:  Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


> >- R3081/R3041
> >- NEC R4200 and ARCset
> >- IDT Orion and ARCset
>
> My personal preference is IDT Orion, since
>
> 1) It is R4000PC pin compatible.
> 
> 2) The price of Orion will be similar to NEC R4200, and R4000PC will
>    out perform R4200, can we deduce that Orion will out perform R4200?
>
> 3) Even point 2) may not hold, a superior upgrade path exists if we
>    use Orion, ie. replace the CPU on the to be made board by a R4000PC, 
>    and get better (at least better than R4200) performance.  WITH THE
>    SAME PRICE (on CPU), why not opt for better upgrade path!

Ok, we *can* decide not to go with the R3081/3041 solution, but
we *cannot* decide now between R4200 and Orion, because we don't
know enough.
I would prefer the Orion too, but it won't be available until
end of the year. That means (I said that already), that only few
of us can start with the expensive R4000PC.

We should try tro decide between R3xxx and R4xxx solutions. If
we all agree over a R4xxx machine things will be clearer.
When we have enough information about the different flavours
of R4xxx CPUs we can decide which one to use.

Agreed ?

-------------------------------------------------------------------------------
        Waldorf Electronics GmbH        | Phone:  +49 (0)2636-80294
              R&D Department            | Fax:    +49 (0)2636-80188
Neustrasse 9-12, 53498 Waldorf, Germany | email:  andy@resi.waldorf-gmbh.de
-------------------------------------------------------------------------------
 
From owner-riscy@pyramid.com  Tue Aug  3 09:55:39 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA10526; Tue, 3 Aug 93 09:55:39 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA22050; Tue, 3 Aug 93 06:52:44 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA24715; Tue, 3 Aug 93 06:50:49 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA24697; Tue, 3 Aug 93 06:50:47 -0700
Received: from benoni.Uit.No 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA21976; Tue, 3 Aug 93 06:51:38 -0700
Received: from benoni by ppenoni.uit.no with SMTP (PP) 
          id <18242-0@ppenoni.uit.no>; Tue, 3 Aug 1993 15:50:35 +0000
Received: from unas.tss.no 
          by benoni.uit.no (5.65+IDA/Babel-1.15/ABaa-1.2/Ultrix) 
          id AAbenoni18238; Tue, 3 Aug 1993 15:50:31 +0200
Received: by unas.tss.no (4.0/ABaa-1.3mini) id AA06121;
          Tue, 3 Aug 93 15:43:56 +0200
Message-Id: <9308031343.AA06121@unas.tss.no>
From: tor@tss.no (Tor Arntsen)
Date: Tue, 3 Aug 1993 15:43:54 +0200
In-Reply-To: Andreas Busse <andy@resi.waldorf-gmbh.de> "Re: Mips project Go or Wait?" (Aug 3, 2:21pm)
X-Mailer: Mail User's Shell (7.2.5 10/14/92)
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

On Aug 3,  2:21pm, Andreas Busse wrote:

>We should try tro decide between R3xxx and R4xxx solutions. If
>we all agree over a R4xxx machine things will be clearer.
>When we have enough information about the different flavours
>of R4xxx CPUs we can decide which one to use.
>
>Agreed ?

I would vote for R4xxx.

Tor (tor@tss.no)

(I have been on vacation for a while (still am), not following the 
 discussions.  I guess others are as well, causing the quietness on the list..)


 
From owner-riscy@pyramid.com  Tue Aug  3 10:51:35 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA13453; Tue, 3 Aug 93 10:51:35 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA29985; Tue, 3 Aug 93 07:48:35 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA27881; Tue, 3 Aug 93 07:47:21 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA27870; Tue, 3 Aug 93 07:47:19 -0700
Received: from giverny.daimi.aau.dk 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA29785; Tue, 3 Aug 93 07:48:09 -0700
Received: by giverny.daimi.aau.dk id AA10224
  (5.65c8/IDA-1.4.4 for riscy@pyramid.com); Tue, 3 Aug 1993 16:47:08 +0200
Date: Tue, 3 Aug 1993 16:47:08 +0200
From: Tommy Thorn <tthorn@daimi.aau.dk>
Message-Id: <199308031447.AA10224@giverny.daimi.aau.dk>
To: riscy@pyramid.com
Subject: Re:  Mips project Go or Wait?
References: <9308031221.AA05562@resi.waldorf-gmbh.de>
Reply-To: Tommy.Thorn@daimi.aau.dk
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Andreas Busse writes:
 > Ok, we *can* decide not to go with the R3081/3041 solution, but
 > we *cannot* decide now between R4200 and Orion, because we don't
 > know enough.
 > I would prefer the Orion too, but it won't be available until
 > end of the year. That means (I said that already), that only few
 > of us can start with the expensive R4000PC.

With the goal of a fast and (fairly) cheap MIPS based workstation,
I'd prefer the Orion chip + ARCset too, even though it might not
arrive soon.

 > We should try tro decide between R3xxx and R4xxx solutions. If
 > we all agree over a R4xxx machine things will be clearer.
 > When we have enough information about the different flavours
 > of R4xxx CPUs we can decide which one to use.
 > 
 > Agreed ?

A design based on R3xxx is going to be obsolete soon, so let
concentrate on R4xxx.

/Tommy Thorn
 
From owner-riscy@pyramid.com  Wed Aug  4 08:24:53 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA00663; Wed, 4 Aug 93 08:24:53 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA11045; Wed, 4 Aug 93 05:24:16 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08481; Wed, 4 Aug 93 05:23:08 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08474; Wed, 4 Aug 93 05:23:06 -0700
Received: from [128.6.21.9] 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10885; Wed, 4 Aug 93 05:24:00 -0700
Received: from eerie.acsu.buffalo.edu by rutgers.edu (5.59/SMI4.0/RU1.5/3.08) 
	id AA12737; Wed, 4 Aug 93 08:21:41 EDT
Received: from uucp@localhost by eerie (8.5/8.1)
	 via UUCP. id IAA28285; Wed, 4 Aug 1993 08:21:38 -0400
Received: by toz.buffalo.ny.us (1.65/waf)
	via UUCP; Tue, 03 Aug 93 23:47:00 EDT
	for eerie!pyramid.com!riscy
To: riscy@pyramid.com
Subject: Re:  Mips project Go or Wait?
From: cyberman@toz.buffalo.ny.us (Cyberman)
Comments:             
Message-Id: <X0Nq8B1w165w@toz.buffalo.ny.us>
Date: Tue, 03 Aug 93 23:45:08 EDT
In-Reply-To: <9308030842.AA16655@gossip.pyramid.com>
Organization: The Tower of Zot - (716)8338-616
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Stephen S M Wong <eerie!hp750a.csc.cuhk.hk!a080700> writes:
> 1) It is R4000PC pin compatible.
> 
> 2) The price of Orion will be similar to NEC R4200, and R4000PC will
>    out perform R4200, can we deduce that Orion will out perform R4200?
> 
> 3) Even point 2) may not hold, a superior upgrade path exists if we
>    use Orion, ie. replace the CPU on the to be made board by a R4000PC, 
>    and get better (at least better than R4200) performance.  WITH THE
>    SAME PRICE (on CPU), why not opt for better upgrade path!

These are all valid reasons.  I suggest  we determine if the "price" is 
right.  $700 is NOT bad for what one would be getting however there is
always Something that one misses :)

Cyberman

 "Everything is vanity ..."
"Even a fool looks wise if he keeps his mouth shut"
"A man to his own, is a person left to his own devices"
"Bewary of thinking you are always right, Hitler did"

Cyberman - this is mysig 
#include <std_disclaimer.h>
#include <explained.h>
main(int argc, char **argv) { FILE *mess; mess = fopen("life.doc"r); 
 
From owner-riscy@pyramid.com  Wed Aug  4 08:37:52 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA01225; Wed, 4 Aug 93 08:37:52 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA02664; Wed, 4 Aug 93 03:38:35 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA02944; Wed, 4 Aug 93 03:37:30 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA02935; Wed, 4 Aug 93 03:37:28 -0700
Received: from boulder.Colorado.EDU 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA02652; Wed, 4 Aug 93 03:38:22 -0700
Received: from agua.Colorado.EDU by boulder.Colorado.EDU with SMTP id AA02294
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Wed, 4 Aug 1993 04:36:09 -0600
Received: by agua.colorado.edu (cu.generic.890828)
Date: Wed, 4 Aug 93 04:33:26 MDT
From: Alan Krantz <atk@agua.Colorado.EDU>
Message-Id: <9308041033.AA24379@agua.colorado.edu>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

I'm just a customer and not a designer but i would vote for an R4200. The
other chips mainly enhance floating point and cost significantly more.
 
From owner-riscy@pyramid.com  Wed Aug  4 08:38:03 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA01273; Wed, 4 Aug 93 08:38:03 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA05531; Wed, 4 Aug 93 04:15:27 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA05754; Wed, 4 Aug 93 04:14:19 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA05747; Wed, 4 Aug 93 04:14:16 -0700
Received: from rrzs1.rz.uni-regensburg.de 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA05506; Wed, 4 Aug 93 04:15:09 -0700
Received: from rrzc1.rz.uni-regensburg.de (rrzc3.rz.uni-regensburg.de) by rrzs1.rz.uni-regensburg.de (4.1/URRZ-relay (1.3))
	id AA18878; Wed, 4 Aug 93 13:12:46 +0200
Received: by rrzc1.rz.uni-regensburg.de (4.1/URRZ-sub (1.3))
	id AA20220; Wed, 4 Aug 93 13:13:06 +0200
Date: Wed, 4 Aug 93 13:13:06 +0200
From: Jens Decker <c5093@rrzc1.rz.uni-regensburg.de>
Message-Id: <9308041113.AA20220@rrzc1.rz.uni-regensburg.de>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

IMHO enhanced floating point performance is the main reason for such a 
project.

Jens Decker, University of Regensburg

 
From owner-riscy@pyramid.com  Wed Aug  4 08:38:07 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA01303; Wed, 4 Aug 93 08:38:07 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA28774; Wed, 4 Aug 93 02:49:13 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28769; Wed, 4 Aug 93 02:47:31 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28759; Wed, 4 Aug 93 02:47:24 -0700
Received: from cs.huji.ac.il 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA28515; Wed, 4 Aug 93 02:48:04 -0700
Received: from picton.cs.huji.ac.il by cs.huji.ac.il with SMTP id AA19542
  (5.65b/HUJI 4.114); Wed, 4 Aug 93 12:45:43 +0300
Received: from localhost by picton.cs.huji.ac.il with SMTP id AA13170
  (5.65c/HUJI 4.121 for <riscy@pyramid.com>); Wed, 4 Aug 1993 12:46:47 +0300
Message-Id: <199308040946.AA13170@picton.cs.huji.ac.il>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait? 
In-Reply-To: Your message of Tue, 3 Aug 93 14:21:23 +0200 .
             <9308031221.AA05562@resi.waldorf-gmbh.de> 
From: Amos Shapira <amoss@cs.huji.ac.il>
Date: Wed, 04 Aug 1993 12:46:44 +0300
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

In message <9308031221.AA05562@resi.waldorf-gmbh.de> you write:
|We should try tro decide between R3xxx and R4xxx solutions. If
|we all agree over a R4xxx machine things will be clearer.
|When we have enough information about the different flavours
|of R4xxx CPUs we can decide which one to use.
|
|Agreed ?

You allready had my vote for an R4xxx solution and from what I read since then
I might even support the Orion too.  Certainly not the R3xxx as it seems quite
clear that it is going to be "out" soon.

Cheers,

--Amos

--Amos Shapira (Jumper Extraordinaire) |  "It is true that power corrupts,
C.S. System Group, Hebrew University,  |   but absolute power is better!"
Jerusalem 91904, ISRAEL                |
amoss@cs.huji.ac.il                    |          -- the Demon to his son
 
From owner-riscy@pyramid.com  Wed Aug  4 09:52:26 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA04866; Wed, 4 Aug 93 09:52:26 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15704; Wed, 4 Aug 93 06:51:38 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA12779; Wed, 4 Aug 93 06:50:13 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA12772; Wed, 4 Aug 93 06:50:11 -0700
Received: from eos03a.eos.ncsu.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15686; Wed, 4 Aug 93 06:51:05 -0700
Received: from homer.ece.ncsu.edu by eos03a.eos.ncsu.edu (5.65b/Eos/S-U-09Jun93)
	id AA04992; Wed, 4 Aug 93 09:50:03 -0400
Posted-Date: Wed, 04 Aug 93 09:50:00 EDT
Received: by homer.eos.ncsu.edu (5.57/eos-client.920428)
	id AA06273; Wed, 4 Aug 93 09:50:02 -0400
From: jlnance@eos.ncsu.edu
Message-Id: <9308041350.AA06273@homer.eos.ncsu.edu>
To: riscy@pyramid.com
Subject: Re:  Mips project Go or Wait?
Date: Wed, 04 Aug 93 09:50:00 EDT
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Hello,
     I wanted to throw out some thoughs about this project and see what
people thought.  It seems to me that two things are certain:

	1) The price/performance ratio of RISC processors will only get
	   better in the future.
	
	2) It will take a (fairly) long to design and debug this thing.

I know that it would be difficult, but given the above, I believe that it
would be good to try to predict how long it will take to design the board
and its operating system, and to try to predict what the cost of the components
will be at that time, and design on the basis of that.  I realize that this
approach is full of risks, and would be difficult to do, but I am afraid that
if this approach is not taken, the board my be obsolete by the time it is
usable.

Jim Nance
 
From owner-riscy@pyramid.com  Wed Aug  4 10:54:24 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA08190; Wed, 4 Aug 93 10:54:24 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA20491; Wed, 4 Aug 93 07:53:32 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA16042; Wed, 4 Aug 93 07:52:22 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA16003; Wed, 4 Aug 93 07:52:20 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA20473; Wed, 4 Aug 93 07:53:09 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.g) via EUnet
	id iA19182; Wed, 4 Aug 1993 16:48:54 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA18307; Wed, 4 Aug 93 16:46:58 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA00674; Wed, 4 Aug 93 16:46:48 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Wed, 4 Aug 93 16:45:47 +0200
Message-Id: <9308041445.AA08844@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA08844; Wed, 4 Aug 93 16:45:47 +0200
To: riscy@pyramid.com
Subject: Re:  Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


> Hello,
>      I wanted to throw out some thoughs about this project and see what
> people thought.  It seems to me that two things are certain:
> 
> 	1) The price/performance ratio of RISC processors will only get
> 	   better in the future.

Correct. Although nobody can pay the latest developments.
Look at the R4000PC. Only *one* year ago it was around $2000,
now it's at about $600...

>	2) It will take a (fairly) long to design and debug this thing.

Don't know. This depends on how complete the NEC ARCset is. So far I
know solves it most of our problems. In case it is true that the
ARCset is a *complete* design, we would only have to make some
changes to satisfy the requirement of the group. Perhaps we have to
make no changes at all !

> I know that it would be difficult, but given the above, I believe that it
> would be good to try to predict how long it will take to design the board
> and its operating system, and to try to predict what the cost of the components
> will be at that time, and design on the basis of that.  I realize that this
> approach is full of risks, and would be difficult to do, but I am afraid that
> if this approach is not taken, the board my be obsolete by the time it is
> usable.

Neil made the assumption that it will take a bit more than half a man-year
to port the kernel and drivers. This depends on how many filesystems we will
support. For me, ext2fs and the ISO fs would be enough.
Everything else *should* compile out of the box, except the X11-server stuff.
That sounds not too bad, isn't it ?

I can't say how long it takes to make the hardware working, because it
depends on what we want.

Andy

-------------------------------------------------------------------------------
        Waldorf Electronics GmbH        | Phone:  +49 (0)2636-80294
              R&D Department            | Fax:    +49 (0)2636-80188
Neustrasse 9-12, 53498 Waldorf, Germany | email:  andy@resi.waldorf-gmbh.de
-------------------------------------------------------------------------------
 
From owner-riscy@pyramid.com  Wed Aug  4 14:40:49 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA21168; Wed, 4 Aug 93 14:40:49 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA21563; Wed, 4 Aug 93 11:40:25 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA10143; Wed, 4 Aug 93 11:39:06 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA10108; Wed, 4 Aug 93 11:39:00 -0700
Received: from hydra.Helsinki.FI 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA21337; Wed, 4 Aug 93 11:39:54 -0700
Received: from kypros (kypros.Helsinki.FI) by hydra.Helsinki.FI (4.1/SMI-4.1/39)
	id AA01138; Wed, 4 Aug 93 21:37:40 +0300
From: Hannu.Savolainen@cs.Helsinki.FI (Hannu Savolainen)
Message-Id: <9308041837.AA01138@hydra.Helsinki.FI>
Subject: Re:  Mips project Go or Wait?
To: riscy@pyramid.com
Date: Wed, 4 Aug 93 21:37:39 EET DST
In-Reply-To: <9308041445.AA08844@resi.waldorf-gmbh.de>; from "Andreas Busse" at Aug 4, 93 4:45 pm
X-Mailer: ELM [version 2.3 PL11]
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> Neil made the assumption that it will take a bit more than half a man-year
> to port the kernel and drivers. This depends on how many filesystems we will
> support. For me, ext2fs and the ISO fs would be enough.
> Everything else *should* compile out of the box, except the X11-server stuff.
> That sounds not too bad, isn't it ?
It should not be difficult to port the filesystems since they don't
access hardware. Also most of the SCSI driver code should be easy to
port since we have to write just a low level driver for one SCSI
chip/host adapter. If we have the ISA bus, porting the ISA device
drivers could be possible by just rewriting the inline assembler macros
used by the drivers. A well written device driver is not CPU dependent.
There could be some problems if the new CPU is big endian.

Hannu
 
From owner-riscy@pyramid.com  Wed Aug  4 16:02:53 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA26015; Wed, 4 Aug 93 16:02:53 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA04580; Wed, 4 Aug 93 13:01:51 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA19006; Wed, 4 Aug 93 13:00:28 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA18994; Wed, 4 Aug 93 13:00:25 -0700
Received: from relay1.UU.NET 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA04488; Wed, 4 Aug 93 13:01:19 -0700
Received: from spool.uu.net (via LOCALHOST) by relay1.UU.NET with SMTP 
	(5.61/UUNET-internet-primary) id AA26169; Wed, 4 Aug 93 15:59:04 -0400
Received: from rei2.UUCP by uucp3.uu.net with UUCP/RMAIL
	(queueing-rmail) id 155731.25934; Wed, 4 Aug 1993 15:57:31 EDT
Received: by rei.com (5.4/5.40/1.0)
	id AA11141; Wed, 4 Aug 1993 13:36:27 -0500
From: rei2!tsprad@uunet.UU.NET (6692)
Message-Id: <9308041836.AA11141@rei.com>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Wed, 4 Aug 1993 13:36:25 -0500 (CDT)
In-Reply-To: <9308041421.AA04255@rei.com> from "UUCP Login" at Aug 4, 93 09:21:39 am
X-Mailer: ELM [version 2.4 PL13]
Content-Type: text
Content-Length: 3954      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> ... good to try to predict how long it will take to design the board and
> its operating system, and to try to predict what the cost of the components
> will be at that time, and design on the basis of that.  I realize that this
> approach is full of risks, and would be difficult to do, but I am afraid that
> if this approach is not taken, the board my be obsolete by the time it is
> usable.
> 
> Jim Nance

Hear, Hear!  Not to be a wet blanket, but it might also be enlightening
to try to predict what the cost of competitive products will be at that
time.  I understand that DEC Alphas are already available for ~US$5K,
at least to selected WindowsNT developers.  And that's complete, with
memory, tube, keyboard, mouse, disk, and CD-ROM drive.

> Personal computer firms plan to use the RISC chips in personal 
> computers when Microsoft's Japanese Windows NT begins to 
> gain popularity. Taiwan's Acer is planning to ship a PC 
> based on the VR4400 as early as this fall.

How much do you think this Taiwanese machine will be selling for by
the time our board is able to boot Linux?  Particularly when the
prediction of that time has to be adjusted for time spent bickering
about fundamental design decisions, like 'dumb frame buffer vs. VESA
local bus connector' and 'IDT R3051 vs. IDT R3081 vs. NEC VR4000PC
vs. NEC VR4200' and 'complete complement of I/O on board vs. ISA
expansion bus'.

I was really disappointed when the IDT 3730 didn't materialize.  I
think that probably set the timetable back _several_ months.  And it
looks to me like we're even more caught up in pursuing Vaporware (tm)
chips now, rather than accepting the lesson of the 3730 and
designing with parts that are available (in small quantities) now.

What I mean about quantities is that we could probably have the 3730
in three months if we could convince IDT that we'll be buying 50K
per month in 1994 and 1995.  On the other hand, it's entirely
possible that the PowerPC will really take off and the VR4200, even
the whole MIPS 4000 line, will be abandoned because there's no
market.

At the risk of repeating myself, let me say again that I don't think
that this project will lead to a state-of-the-art workstation at a
bargain price.  We just can't compete with the large-volume
manufacturers.  But I still think it's worth doing, and I still
think there's a (small) chance there could eventually be many
thousands, even tens of thousands of the things built over the next
several years.

Does anyone have any usable estimates of how many
Linux/386BSD/NetBSD  machines there are?  A month or so ago Bill
Jollitz was throwing around the number 'a million' for 386BSD alone
(but of course I take that with a grain of salt).  But say there's
something on the order of a million of Linux & 386BSD & NetBSD
combined, worldwide.  Obviously, all of these people are adventurous
tinkers, or they would stick with 'supported' software.  Do you
think that 1% of those people might buy our board, if they could get
system software that boots up and basically runs, and if they knew
they could get compatible motherboards from at least two or three
manufacturers around the world, and all their other hardware needs
from their local discount computer retailer, and all the 'support'
they could possibly want from net news and mailing lists?

One percent of a million is ten thousand.  Even if I'm off by an
order of magnitude it's still one thousand to one hundred thousand.
Enough to support a pretty active news group.  And if every one of
them has the 'sources' of the hardware design, enough to deal
compatibly with obsolete components for a long, long time.

-- 
Ted Spradley   Recognition International, Inc.  Opinions are mine, not theirs.
2701 E Grauwyler Rd. |Your productivity is not enhanced when you're staring
Irving TX 75061      |at that thing.  Your productivity is enhanced when
214-579-6692         |the computer is working and you're doing something else.
 
From owner-riscy@pyramid.com  Wed Aug  4 18:12:31 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA02852; Wed, 4 Aug 93 18:12:31 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA19725; Wed, 4 Aug 93 15:11:54 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA04234; Wed, 4 Aug 93 15:09:34 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA04214; Wed, 4 Aug 93 15:09:31 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA19698; Wed, 4 Aug 93 15:10:25 -0700
Message-Id: <9308042210.AA19698@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA03943; Wed, 4 Aug 93 18:08:43 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Wed, 4 Aug 1993 18:08:43 -0500 (EDT)
In-Reply-To: <9308041836.AA11141@rei.com> from "6692" at Aug 4, 93 01:36:25 pm
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 4730      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

>> ... good to try to predict how long it will take to design the board and
>> its operating system, and to try to predict what the cost of the components
>> will be at that time, and design on the basis of that.  I realize that this
>> approach is full of risks, and would be difficult to do, but I am afraid that
>> if this approach is not taken, the board my be obsolete by the time it is
>> usable.

Well current state of the art in the pc world is a 486/66 (32.4 specint 16.1 fp)
In six months the 66 Mhz pentium will be state of the art (64.5 specint 56.9 fp)
and a good motherboard like micronices VESA is over $1000 with a 486 66 in it.
Last year a 486/50 ISA was over $1000.  Don't forget that the 486 motherboard 
doesn't have video, ethernet or scsi.

So if we can get the 4200 (55int/30 fp) 4000 (62int/63fp) going it seems
that even in 6 months we would be competitive if it's under $1000.
If we use the arc 100 chipset we have nothing to design, all we 
have to do is order the parts, and contract out the design of the motherboards.
Oh yeah and port linux ;-)

> Hear, Hear!  Not to be a wet blanket, but it might also be enlightening
> to try to predict what the cost of competitive products will be at that
> time.  I understand that DEC Alphas are already available for ~US$5K,
> at least to selected WindowsNT developers.  And that's complete, with
> memory, tube, keyboard, mouse, disk, and CD-ROM drive.

Well I don't expect anyone to sell 486 motherboard ugprade over the next
6 months because of problems with software/integrations/hardware compatibility.
Complete systems will compete with complete pentium/486 systems, but I don't
expect any direct competition with 486 motherboards.

> How much do you think this Taiwanese machine will be selling for by
> the time our board is able to boot Linux?  

ACER isn't selling mothebroards, and ACER needs to make a profit, and
ACER's machine won't run linux.  

> Particularly when the
> prediction of that time has to be adjusted for time spent bickering
> about fundamental design decisions, like 'dumb frame buffer vs. VESA
> local bus connector' and 'IDT R3051 vs. IDT R3081 vs. NEC VR4000PC
> vs. NEC VR4200' and 'complete complement of I/O on board vs. ISA
> expansion bus'.

Well hopefully everyone will agree the arc 100 design is better and we
will have to do NO design, just use the NEC design. 

> I was really disappointed when the IDT 3730 didn't materialize.  I
> think that probably set the timetable back _several_ months.  And it

Well the arc 100/150 look even better. (MCT-ADR MCT-DP)

> possible that the PowerPC will really take off and the VR4200, even
> the whole MIPS 4000 line, will be abandoned because there's no
> market.

Well I personally prefer the powerpc.  It run's gcc, gets more done per Mhz,
very elegant, reasonably priced, and has the largest onchip cache.  But the 
simple fact that there exists a mips 4000 board design outweighs any 
advantages the powerpc has over the mips 4000 series.  Unless we find
a ready to use powerpc motherboard design of course.

I'd really doubt the mips chips will have any problems, they are faster
then the pentium, run windows nt, and have SGI/MIPS and various pc companies
supporting them.

> At the risk of repeating myself, let me say again that I don't think
> that this project will lead to a state-of-the-art workstation at a
> bargain price.  We just can't compete with the large-volume
> manufacturers.  But I still think it's worth doing, and I still
> think there's a (small) chance there could eventually be many
> thousands, even tens of thousands of the things built over the next
> several years.

Well the 486 is an ugly architecture, and is slow especially in fp.  (not
to mention I'd bet that the 486 specmarks are more optimized then anyone elses.)
If I can get better price/performance on a riscy motherboard with everything
on board, then I can get with a 486 motherboard+video+scsi I'll buy it.

[ deleted long speculation on the number of free unix users and extrapolating
market for the riscy board]

There are no doubt that there are MANY free unix users, but MANY of those
are running with <= 4 MB ram, <= 386-33, IDE disks, tident or tseng video
cards, etc.

Most free unix users are college age or so, and don't have the better part
of $1000 to upgrade there system that doesn't run winNT or dos software.

But I hate the 486/586 design, and microsoft software, and hope to
get a riscy motherboard to run linux.  So I can have a workstation at home.

-- 
Bill					1st>	Broadley@neurocog.lrdc.pitt.edu
Broadley@schneider3.lrdc.pitt.edu <2nd 	3rd> 	             Broadley+@pitt.edu
Linux is great.         Bike to live, live to bike.                      PGP-ok
 
From owner-riscy@pyramid.com  Wed Aug  4 20:03:10 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA06663; Wed, 4 Aug 93 20:03:10 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA29035; Wed, 4 Aug 93 17:02:35 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA20122; Wed, 4 Aug 93 17:01:22 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA20107; Wed, 4 Aug 93 17:01:16 -0700
Received: from kinglear.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA29023; Wed, 4 Aug 93 17:02:10 -0700
Received: from localhost by kinglear.cs.Colorado.EDU with SMTP id AA10724
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Wed, 4 Aug 1993 18:01:09 -0600
Message-Id: <199308050001.AA10724@kinglear.cs.Colorado.EDU>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait? 
In-Reply-To: Your message of "Wed, 04 Aug 1993 16:45:47 +0200."
             <9308041445.AA08844@resi.waldorf-gmbh.de> 
Date: Wed, 04 Aug 1993 18:01:08 -0600
From: Drew Eckhardt <drew@kinglear.cs.Colorado.EDU>
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

    > would be good to try to predict how long it will take to design the board
    > and its operating system, and to try to predict what the cost of the comp
   onents
    > will be at that time, and design on the basis of that.  I realize that th
   is
    > approach is full of risks, and would be difficult to do, but I am afraid 
   that
    > if this approach is not taken, the board my be obsolete by the time it is
    > usable.
    
    Neil made the assumption that it will take a bit more than half a man-year
    to port the kernel and drivers. This depends on how many filesystems we wil
   l
    support. For me, ext2fs and the ISO fs would be enough.

It has little to do with how many filesystems we support, as the filesystems 
should be devoid of any system dependant code.  Endianess is pretty easy to 
deal with where we need binary compatability with existing filesystems (ie
MSDOS since you may want to read DOS floppies). 

As far as how long device drivers take :

IDE, serial, parallel, etc. where we've got the same parts as in the PC - 
we simply add appropriate #ifdef LINUX_MIPS or whatever to the kernel sources
around code that does memory mapped I/O instead of using the port commands.

SCSI - it looks like the SCSI driver I'm writing right now (NCR5380) 
will have taken about 20 man-hours (including me reading the NCR5380 
databook, designing the driver architecture, writing the code, etc) by 
the time it goes into ALPHA test.  The 5390's are even easier to program. 

Ie : device drivers are not a significant problem.

I think the hard parts will be taking care of the major 386isms, ie the 
scheduler and VM which is really prone to race conditions.  Since people have 
access to other MIPS architecture machines (ie, Decstations) and sample MIPS 
code to work with (BSD4.4 runs on the Decstations) this can happen in parallel 
with the hardware development.

    Everything else *should* compile out of the box, except the X11-server stuf
   f.
    That sounds not too bad, isn't it ?

I don't think there will be any significant technical problems getting 
the software up. 

Speaking of kernel development for the MIPS : we need to start a discussion
on how to best make the kernel sources multi-architecture, so that one 
release of Linux runs on MIPS, Intel, and Motorola platforms.

The AMIGA people didn't bother to do this, so they need a different 
kernel distribution and can never take advantage of new features 
when they're added.
 
From owner-riscy@pyramid.com  Wed Aug  4 23:30:35 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA12789; Wed, 4 Aug 93 23:30:35 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA17925; Wed, 4 Aug 93 20:29:22 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA09352; Wed, 4 Aug 93 20:27:14 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA09345; Wed, 4 Aug 93 20:27:11 -0700
Received: from munnari.OZ.AU 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA17465; Wed, 4 Aug 93 20:27:52 -0700
Received: from sw.oz.au (via basser.cs.su.oz.au) by munnari.oz.au with MHSnet (5.83--+1.3.1+0.50)
	id AA15177; Thu, 5 Aug 1993 13:24:07 +1000 (from jeremy@sw.oz.au)
Received: from chao.sw.oz.au by swift.sw.oz.au with SMTP
	id AA06649; Thu, 5 Aug 93 13:19:59 AES (5.59)
	(from jeremy@sw.oz.au for riscy%pyramid.com@munnari.oz.au)
Received: by chao.sw.oz.au (4.1/SMI-4.1)
	id AA18703; Thu, 5 Aug 93 13:19:45 EST
From: jeremy@sw.oz.au (Jeremy Fitzhardinge)
Message-Id: <9308050319.AA18703@chao.sw.oz.au>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Thu, 5 Aug 1993 13:19:44 +1000 (EST)
In-Reply-To: <9308041445.AA08844@resi.waldorf-gmbh.de> from "Andreas Busse" at Aug 4, 93 04:45:47 pm
Organization: Softway Pty Ltd
X-Face: 
	 '6U=%Tv\k1<Ek%ql%PN^v`Db4bakr[v~y]\u7"GbO#I=]N{l1=#P,glz$9q>l-:?\$C[D@G
	 7(vl~w8&y}!f\bh#w<Y*S~bEBTI:s&.QR>L#n,TGKh>T.c7eT5-y)Hl'i;A1z$9?*lD.k}yqshddFb
	 l[EC}c=;uc%x'}uh3E91p&oE<q$w1r&U0yw.Sb3V&uw
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 2550      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Andreas Busse bubbles:

> Neil made the assumption that it will take a bit more than half a man-year
> to port the kernel and drivers. This depends on how many filesystems we will
> support. For me, ext2fs and the ISO fs would be enough.

The filesystems are entirely in the machine-independant code.
I don't think many changes at all will be needed for porting all of
the filesystems.  The MMU, DMA, interrupt handling and simplest device
drivers (console, serial, naif SCSI/ether) will probably be the longest to
get going because the machine is useless until they do.

> Everything else *should* compile out of the box, except the X11-server stuff.
> That sounds not too bad, isn't it ?

With a dumb frame buffer there will be very little work getting X11 going.
There's already lots of support in the MIT code for 8 bit deep dumb
frame buffers.  It'll just be a matter of reading the keyboard and mouse
properly (I think - I could be completely wrong).

Libraries could be a challenge to port.  Not because its very difficult
or machine dependant, but all kinds of correct decisions need to be made.
For example, if I were to port linux to an 4000, should off_t etc be
64 bit?  If so, should the library and compile environment have support
for a 32 bit environment and a 64 bit one (by having the 32 bit syscall
lib functions convert their args and returns to 64 bit).

> I can't say how long it takes to make the hardware working, because it
> depends on what we want.

Bearing in mind that 4000 based commercial machines are becoming available
*now* at reasonable or nearly reasonable prices.  For example, a brocure
from a company called "Tangent" turned up here selling an R4000 based
machine (can't remember which variant) running at 100MHz, 16MB ram,
16k cache (cannot sure if its I+D combined or 16k each), 200MB disk, hi
res graphics (~1000x1000x8) with 64 bit path to CPU + S3 accellerator,
E?ISA slots, monitor and Windows NT (SVR4 port available) for $4000 US.
That was the base machine with others up to 150MHz, more cache and memory
and so on.  A similar machine is $10000 here, but that's predictable.
Whether or not thay can actually deliver is another question.  Its also
possible that they'd sell motherboards alone.

I'm not cash starved and I'm keen to port linux to a Mips machine
(I've started already, mainly by looking at the i386 kernel for machine
dependancies and isolating them), so this kind of thing is an option
for me.  It also means there's some chance of a Mips kernel waiting for
any board that gets built.

	J
 
From owner-riscy@pyramid.com  Wed Aug  4 23:38:53 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA13097; Wed, 4 Aug 93 23:38:53 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18567; Wed, 4 Aug 93 20:38:01 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA10103; Wed, 4 Aug 93 20:35:55 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA10095; Wed, 4 Aug 93 20:35:52 -0700
Received: from kinglear.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18552; Wed, 4 Aug 93 20:36:47 -0700
Received: from localhost by kinglear.cs.Colorado.EDU with SMTP id AA13729
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Wed, 4 Aug 1993 21:35:45 -0600
Message-Id: <199308050335.AA13729@kinglear.cs.Colorado.EDU>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait? 
In-Reply-To: Your message of "Thu, 05 Aug 1993 13:19:44 +1000."
             <9308050319.AA18703@chao.sw.oz.au> 
Date: Wed, 04 Aug 1993 21:35:43 -0600
From: Drew Eckhardt <drew@kinglear.cs.Colorado.EDU>
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

    With a dumb frame buffer there will be very little work getting X11 going.
    There's already lots of support in the MIT code for 8 bit deep dumb
    frame buffers.  It'll just be a matter of reading the keyboard and mouse
    properly (I think - I could be completely wrong).

There's also 32 bit server code (from DEC) in the contrib hierchy.
    
    Libraries could be a challenge to port.  Not because its very difficult
    or machine dependant, but all kinds of correct decisions need to be made.
    For example, if I were to port linux to an 4000, should off_t etc be
    64 bit?  If so, should the library and compile environment have support
    for a 32 bit environment and a 64 bit one (by having the 32 bit syscall
    lib functions convert their args and returns to 64 bit).

Take a look at the various papers from Convex ('92 Winter USENIX proceedings),
BSD CSRG, etc. on supporting Big Files.
    
 
From owner-riscy@pyramid.com  Thu Aug  5 02:16:27 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA17252; Thu, 5 Aug 93 02:16:27 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09042; Wed, 4 Aug 93 23:15:25 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA25037; Wed, 4 Aug 93 23:13:40 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA25030; Wed, 4 Aug 93 23:13:37 -0700
Received: from munnari.OZ.AU 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09018; Wed, 4 Aug 93 23:13:42 -0700
Received: from sw.oz.au (via basser.cs.su.oz.au) by munnari.oz.au with MHSnet (5.83--+1.3.1+0.50)
	id AA21983; Thu, 5 Aug 1993 16:09:22 +1000 (from jeremy@sw.oz.au)
Received: from chao.sw.oz.au by swift.sw.oz.au with SMTP
	id AA09677; Thu, 5 Aug 93 15:48:30 AES (5.59)
	(from jeremy@sw.oz.au for riscy%pyramid.com@munnari.oz.au)
Received: by chao.sw.oz.au (4.1/SMI-4.1)
	id AA00638; Thu, 5 Aug 93 15:48:22 EST
From: jeremy@sw.oz.au (Jeremy Fitzhardinge)
Message-Id: <9308050548.AA00638@chao.sw.oz.au>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Thu, 5 Aug 1993 15:48:21 +1000 (EST)
In-Reply-To: <199308050001.AA10724@kinglear.cs.Colorado.EDU> from "Drew Eckhardt" at Aug 4, 93 06:01:08 pm
Organization: Softway Pty Ltd
X-Face: 
	 '6U=%Tv\k1<Ek%ql%PN^v`Db4bakr[v~y]\u7"GbO#I=]N{l1=#P,glz$9q>l-:?\$C[D@G
	 7(vl~w8&y}!f\bh#w<Y*S~bEBTI:s&.QR>L#n,TGKh>T.c7eT5-y)Hl'i;A1z$9?*lD.k}yqshddFb
	 l[EC}c=;uc%x'}uh3E91p&oE<q$w1r&U0yw.Sb3V&uw
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 2879      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Drew Eckhardt bubbles:

> Speaking of kernel development for the MIPS : we need to start a discussion
> on how to best make the kernel sources multi-architecture, so that one 
> release of Linux runs on MIPS, Intel, and Motorola platforms.

I've already started this.  I have the root source dir.  Under it is
common/ with everything that's machine independant (filesystems, most
syscalls, etc).  There's another tree with cpu/arch, eg, i386/isa,
m68k/common, m68k/amiga, mips/common, mips/riscy, mips/idt and so on.
Currently I have a i386/isa with all the obviously intel specific stuff
in it (most things with __asm__(), dma, interrupt device drivers and so
on), and common with everything else.  It almost compiles, but I haven't
upgraded it to pl11.  Once it compiles, I want to try and move some
of the more gratuitous dependancies into portable code (like the asm()
in printf for doing a divide).

The include dir is also split, with the machine dependant stuff being
looked at before the common.  Of course, common code will include machine
dependancies from headers.

I think the port should avoid unnecessary changes to linux itself - it
should be a port and nothing more.  That doesn't exclude architecture
specific optimisations, but they are not necessary for the initial port.
Once its done, then other changes can happen.  This means that there
should be as much shared code as possible (which is why each cpu has a
common directory).  Also, the port will be for little endian mips, and
the tlb stuff will look like the intel mmu, because its easy to do so,
and there's no efficiency problems from doing so.  Using this approach
there can be a maximum of sharing with the intel linux.

I'm mostly interested in coordinating the initial port; that is getting
core of linux running on a mips chip.  I think it should be done by a very
small number of people in close communication, with a strong direction.
After that, it's easy to get more people working at once on porting device
drivers, essential usermode stuff (libraries), and so on.  From there,
it's everyone go.

I'd like to get a mips platform to start porting with before any hardware
is built, so there is something solid to start with -- there's little
enough time to make this compeditive as it is.

> The AMIGA people didn't bother to do this, so they need a different 
> kernel distribution and can never take advantage of new features 
> when they're added.

Yes, I was dissapointed by this.  The whole port looked a quite hackish.
It wasn't even very useful to find out where they had found intel
specific stuff, because they didn't localize it but just replaced it
with Moto things.  But it does work, I gather, and that's a start.
Perhaps when we have a good multi-architecture tree the moto stuff will
slot in easily.  I notice that they are trying to make it work on the
Amiga, ST and Mac.

	J
 
From owner-riscy@pyramid.com  Thu Aug  5 03:13:43 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA18773; Thu, 5 Aug 93 03:13:43 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18953; Thu, 5 Aug 93 00:13:01 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA01987; Thu, 5 Aug 93 00:11:52 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA01980; Thu, 5 Aug 93 00:11:50 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18839; Thu, 5 Aug 93 00:12:40 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.g) via EUnet
	id AX25910; Thu, 5 Aug 1993 09:08:27 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA25294; Thu, 5 Aug 93 09:01:45 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA03220; Thu, 5 Aug 93 09:01:18 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Thu, 5 Aug 93 09:00:22 +0200
Message-Id: <9308050700.AA09895@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA09895; Thu, 5 Aug 93 09:00:22 +0200
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> [lots of stuff deleted]
>
> Speaking of kernel development for the MIPS : we need to start a discussion
> on how to best make the kernel sources multi-architecture, so that one 
> release of Linux runs on MIPS, Intel, and Motorola platforms.
> 
> The AMIGA people didn't bother to do this, so they need a different 
> kernel distribution and can never take advantage of new features 
> when they're added.

Yes, this is right and I was already thinking about. The simplest
way would be a lot of #ifdef's, but that would make the code really
ugly. We should talk to Linus if it's possible to separate the
CPU-dependent stuff into several files, so that the CPU decision
can be made by compiling and linking different files. I'm sure
this is not always possible, but at least for things like the kernel boot
code it should not make any problem (and is necessary there).

Andy

-------------------------------------------------------------------------------
        Waldorf Electronics GmbH        | Phone:  +49 (0)2636-80294
              R&D Department            | Fax:    +49 (0)2636-80188
Neustrasse 9-12, 53498 Waldorf, Germany | email:  andy@resi.waldorf-gmbh.de
-------------------------------------------------------------------------------
 
From owner-riscy@pyramid.com  Thu Aug  5 03:24:31 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA19187; Thu, 5 Aug 93 03:24:31 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA20475; Thu, 5 Aug 93 00:23:47 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA02162; Thu, 5 Aug 93 00:22:34 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA02155; Thu, 5 Aug 93 00:22:31 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA20411; Thu, 5 Aug 93 00:23:26 -0700
Message-Id: <9308050723.AA20411@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA06397; Thu, 5 Aug 93 03:21:34 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Thu, 5 Aug 1993 03:21:33 -0500 (EDT)
In-Reply-To: <9308050319.AA18703@chao.sw.oz.au> from "Jeremy Fitzhardinge" at Aug 5, 93 01:19:44 pm
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 1220      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> Bearing in mind that 4000 based commercial machines are becoming available
> *now* at reasonable or nearly reasonable prices.  For example, a brocure
> from a company called "Tangent" turned up here selling an R4000 based
> machine (can't remember which variant) running at 100MHz, 16MB ram,
> 16k cache (cannot sure if its I+D combined or 16k each), 200MB disk, hi
> res graphics (~1000x1000x8) with 64 bit path to CPU + S3 accellerator,
> E?ISA slots, monitor and Windows NT (SVR4 port available) for $4000 US.
> That was the base machine with others up to 150MHz, more cache and memory
> and so on.  A similar machine is $10000 here, but that's predictable.
> Whether or not thay can actually deliver is another question.  Its also
> possible that they'd sell motherboards alone.

Well 16 MB ram is about $400, 340 MB IDE disk is around $298.  Sounds like
they are charging $3300 for a motherboard+video+case+monitor sounds alot
more then the planned under $1k for video+scsi+ethernet+motherboard to me.


-- 
Bill					1st>	Broadley@neurocog.lrdc.pitt.edu
Broadley@schneider3.lrdc.pitt.edu <2nd 	3rd> 	             Broadley+@pitt.edu
Linux is great.         Bike to live, live to bike.                      PGP-ok

 
From owner-riscy@pyramid.com  Thu Aug  5 05:01:00 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA21991; Thu, 5 Aug 93 05:01:00 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA00596; Thu, 5 Aug 93 02:00:16 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08814; Thu, 5 Aug 93 01:58:58 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08807; Thu, 5 Aug 93 01:58:55 -0700
Received: from svin02.info.win.tue.nl 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA00551; Thu, 5 Aug 93 01:59:50 -0700
Received: from wsinis01.info.win.tue.nl by svin02.info.win.tue.nl (4.1/1.45)
    id AA29423; Thu, 5 Aug 93 10:57:32 +0200
Received: by wsinis01.info.win.tue.nl (4.1/1.45)
    id AA26878; Thu, 5 Aug 93 10:57:29 +0200
From: monty@info.win.tue.nl (Guido Leenders)
Message-Id: <9308050857.AA26878@wsinis01.info.win.tue.nl>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Thu, 5 Aug 93 10:57:24 MET DST
In-Reply-To: <9308050548.AA00638@chao.sw.oz.au>; from "Jeremy Fitzhardinge" at Aug 5, 93 3:48 pm
Organization: The Timewasters
X-Mailer: ELM [version 2.3 PL11]
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> 
> I've already started this.  I have the root source dir.  Under it is
> common/ with everything that's machine independant (filesystems, most
> syscalls, etc).  There's another tree with cpu/arch, eg, i386/isa,
> m68k/common, m68k/amiga, mips/common, mips/riscy, mips/idt and so on.
> Currently I have a i386/isa with all the obviously intel specific stuff
> in it (most things with __asm__(), dma, interrupt device drivers and so
> on), and common with everything else.  It almost compiles, but I haven't
> upgraded it to pl11.  Once it compiles, I want to try and move some
> of the more gratuitous dependancies into portable code (like the asm()
> in printf for doing a divide).
> 

That was a great idea. I just want to make two suggestions:

1. Maybe you could contact Linus about this. You might try to
   convince him to incorporate this source tree into the official
   releases, so we don't have to catch up with new releases 
   all the time.
   Of course it's not at all totally sure that this project will
   ever get done, but it's a general good thing, since maybe
   someone in the future will try to make a port to (say) Alpha.
2. Maybe you could move the common-directory, to get more object-oriented,
   e.g.
	  i386  (common i386-stuff)
		 --- isa (isa-specific)
	  m68k  (common 68000-stuff)
		 --- amiga
		 --- mac
		 --- atari


Just a few thoughts.

Guido

--
Guido Leenders
monty@win.tue.nl
Eindhoven University of Technology
the Netherlands
 
From owner-riscy@pyramid.com  Thu Aug  5 12:40:18 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA08916; Thu, 5 Aug 93 12:40:18 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09818; Thu, 5 Aug 93 09:39:44 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08054; Thu, 5 Aug 93 09:37:51 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08047; Thu, 5 Aug 93 09:37:49 -0700
Received: from avignon.daimi.aau.dk 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09621; Thu, 5 Aug 93 09:38:44 -0700
Received: by avignon.daimi.aau.dk id AA22134
  (5.65c8/IDA-1.4.4 for riscy@pyramid.com); Thu, 5 Aug 1993 18:37:40 +0200
Date: Thu, 5 Aug 1993 18:37:40 +0200
From: Tommy Thorn <tthorn@daimi.aau.dk>
Message-Id: <199308051637.AA22134@avignon.daimi.aau.dk>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait?
Reply-To: Tommy.Thorn@daimi.aau.dk
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

   From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
   Date: Wed, 4 Aug 1993 18:08:43 -0500 (EDT)
   Sender: owner-riscy@pyramid.com

   Well I personally prefer the powerpc.  It run's gcc, gets more done per Mhz,
   very elegant, reasonably priced, and has the largest onchip cache.  But the 
   simple fact that there exists a mips 4000 board design outweighs any 
   advantages the powerpc has over the mips 4000 series.  Unless we find
   a ready to use powerpc motherboard design of course.

My *personal* *taste* for the processors ranks Alpha, R4000, and
PowerPC, in this order, but that irrelevant of course. Why do you keep
pointing out that gcc exits for the powerpc, when the same is the case
for the mips?

In fact setting up a gas+binutils+gcc cross compilation enviroment +
spim (mips 3000 simulator) makes it possible to experiment with the
mips architechure already today.

I've tried to get information on the R4000 series the last two days,
but to no avail. Whats the difference from a R3000 (from a programming
view, and yes, I know it's 64 bit ;^)? Can anyone point me to
documentation?

Tommy.Thorn@daimi.aau.dk | If you use a PC consider upgrading to a real OS.
------------------------ | Linux is Posix (Unix). It has GCC,Emacs,X11,TCP,NFS,
*L*I*N*U*X* Fanatic (tm) | Doc,Interviews,Prolog...(you name it) and it's free!
 
From owner-riscy@pyramid.com  Thu Aug  5 18:35:44 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA26063; Thu, 5 Aug 93 18:35:44 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10380; Thu, 5 Aug 93 15:35:20 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA19591; Thu, 5 Aug 93 15:33:58 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA19584; Thu, 5 Aug 93 15:33:55 -0700
Received: from ubiserver.ubitrex.mb.ca 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10363; Thu, 5 Aug 93 15:34:29 -0700
Received: from ska.ubitrex.mb.ca ([192.75.16.23]) by ubitrex.mb.ca (4.1/SMI-4.1)
	id AA26476; Thu, 5 Aug 93 17:32:05 CDT
Date: Thu, 5 Aug 93 17:32:05 CDT
From: tim@ubitrex.mb.ca (Tim Braun)
Message-Id: <9308052232.AA26476@ubitrex.mb.ca>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait?
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

My personal opinion is Go, and that means R3081.

It strikes me that if you are saturating a bus with
i/o, it doesn't matter if the CPU is doing it or if
it's DMA, you are bus limited.  So I suggest considering
non-dma IDE and floppy interfaces.  The existing IDE
driver uses cpu controlled transfers, which are to
a sector buffer that can be accessed at full speed.

Consider:
  R3081 CPU                               $80.00
  MOM memory/video/keyboard i/f           $30.00
  smc37c652 floppy/serial/ide/parallel    $20.00

  dual-port ram ethernet (8390) _or_ bus-master ethernet (79900) $20.00
  bus-master SCSI (53cf90) _or_ how big is the fifo on these?    $25.00

  Board $100.00
  Misc  $100.00

  total $375.00

They (IDT) say 35 MIPS, 11 MFlops.  I believe that's on the
order of a 486dx2-66.  Sytem performance would be (because of
the local bus peripherals) on the same order. Correct me if I'm 
wrong.
________________________________________________________________
Tim Braun                          |
Ubitrex Corporation                | Voice: 204-942-2992 ext 228
1900-155 Carlton St                | FAX:   204-942-3001
Winnipeg, Manitoba, Canada R3C 3H8 | Email: tim@ubitrex.mb.ca
 
From owner-riscy@pyramid.com  Thu Aug  5 20:25:47 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA29299; Thu, 5 Aug 93 20:25:47 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA19262; Thu, 5 Aug 93 17:25:35 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA09681; Thu, 5 Aug 93 17:24:19 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA09656; Thu, 5 Aug 93 17:24:16 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA19254; Thu, 5 Aug 93 17:25:12 -0700
Message-Id: <9308060025.AA19254@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA16879; Thu, 5 Aug 93 20:23:35 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Re: Mips project Go or Wait?
To: riscy@pyramid.com
Date: Thu, 5 Aug 1993 20:23:35 -0500 (EDT)
In-Reply-To: <9308052232.AA26476@ubitrex.mb.ca> from "Tim Braun" at Aug 5, 93 05:32:05 pm
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 1002      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> Consider:
>   R3081 CPU                               $80.00

Last I heard this was over $300.

>   MOM memory/video/keyboard i/f           $30.00
Really this cheap?

I seem to remember adding ISA+VESA+SCSI+ethernet on the single available
DMA channel on the mom chip would be difficult.

>   smc37c652 floppy/serial/ide/parallel    $20.00
> 
>   dual-port ram ethernet (8390) _or_ bus-master ethernet (79900) $20.00
>   bus-master SCSI (53cf90) _or_ how big is the fifo on these?    $25.00
> 
>   Board $100.00
>   Misc  $100.00
> 
>   total $375.00

So this is off at least $220.00

> 
> They (IDT) say 35 MIPS, 11 MFlops.  I believe that's on the
> order of a 486dx2-66.  Sytem performance would be (because of
> the local bus peripherals) on the same order. Correct me if I'm 
> wrong.

-- 
Bill					1st>	Broadley@neurocog.lrdc.pitt.edu
Broadley@schneider3.lrdc.pitt.edu <2nd 	3rd> 	             Broadley+@pitt.edu
Linux is great.         Bike to live, live to bike.                      PGP-ok

 
From owner-riscy@pyramid.com  Fri Aug  6 04:42:12 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA12629; Fri, 6 Aug 93 04:42:12 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA02707; Fri, 6 Aug 93 01:42:03 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA02361; Fri, 6 Aug 93 01:40:11 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA02349; Fri, 6 Aug 93 01:40:08 -0700
Received: from kinglear.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA02696; Fri, 6 Aug 93 01:41:05 -0700
Received: from localhost by kinglear.cs.Colorado.EDU with SMTP id AA15047
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Fri, 6 Aug 1993 02:40:02 -0600
Message-Id: <199308060840.AA15047@kinglear.cs.Colorado.EDU>
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait? 
In-Reply-To: Your message of "Thu, 05 Aug 1993 17:32:05 CDT."
             <9308052232.AA26476@ubitrex.mb.ca> 
Date: Fri, 06 Aug 1993 02:40:01 -0600
From: Drew Eckhardt <drew@kinglear.cs.Colorado.EDU>
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


    My personal opinion is Go, and that means R3081.

The 4200, which is both cheaper and faster than the 3081 
is also shipping, albeit NEC is only making 10K 4000PC's and
4200's in a month.
    
    It strikes me that if you are saturating a bus with
    i/o, it doesn't matter if the CPU is doing it or if
    it's DMA, you are bus limited.  So I suggest considering
    non-dma IDE and floppy interfaces.  

You aren't saturating the bus.

You really want a DMA floppy interface (ie, wire the floppy
controller up as a slave device as it was intended to be) because
without it you have to sit in a loop while you read the track 
(ie, 160ms) and can't be doing other things.

DMA transfers will move the same amount of data, but you can
do other things while the actual transfer (9K over 1/6th of a second
for a 1:1 interleaved 1.44M floppy) happens.

The IDE drives do have a sector buffer so it's not as critical,
but you could still gain something since IDE drives use a 16 bit
bus that (if standard) runs 8Mhz, 1 wait.
    
    The existing IDE
    driver uses cpu controlled transfers, which are to
    a sector buffer that can be accessed at full speed.

Full ISA bus speed.
    
    Consider:
      R3081 CPU                               $80.00

Wrong.  All of the quotes we have on the R3081 in quantities
we can use are at least $300, NEC press releases have said $70
for the 4200 but no one has talked to a real distributor who sets
prices for small quantities.

Low quotes on the R3051 (ie, no floating point) were in the 
$120 range.

FYI, I talked to NEC today and they refered me to the local 
distributor for their parts

Compass Marketing +1 (303) 721-9663

I didn't get a chance to call during business hours and there's nothing
to say the distributor will be able to give us prices / parts given the
low (10K/month) production rate at this point, but I finally have a number
and can try tomorrow.

      MOM memory/video/keyboard i/f           $30.00

The MOM really doesn't do what we want - no variable bus sizing for the 
peripherials, no slave mode DMA for the SCSI / ethernet, the bus master
has to have 3051 / 3081 / 3041 compatable signals (ie Mux'd address/data),
etc.  In other words : the simplest way to do it hardware is with something
other than the MOM (ie, arc), the external hardware needed on the MOM makes
life difficult.

The simplest MOM solution (hardware) may be to use a 3041 running real 
time software in the sample design in place of the ILAC ethernet busmaster 
device, which in turn acts as a slave mode DMA controller, etc. but software 
would be tricky.

      smc37c652 floppy/serial/ide/parallel    $20.00
    
      dual-port ram ethernet (8390) _or_ bus-master ethernet (79900) $20.00

Do you have actual quotes on this, or is it an estimate?  If it's the 
former, does it include the price of the memory, arbiter PAL (or some
fraction of some other PAL we use) for the shared memory, tranciever,
etc (ie, everything in the NS app. note).

      bus-master SCSI (53cf90) _or_ how big is the fifo on these?    $25.00

The 53cf90 is *not* a busmaster, it's a slave dma/pio device with a 16 byte 
FIFO.  However, as I have stated in the past, with a *sane* architecture
(ie, the m68k which runs all DMA through the MMU) you get scatter gather
for slave mode devices and you don't need to bus-master.  For the 
typical SCSI command using this chip we'll service *at most* four 
interrupts where we can transmit data immediately without sitting in a busy 
loop of any sort.

      Board $100.00
      Misc  $100.00
    
      total $375.00
    
    They (IDT) say 35 MIPS, 11 MFlops.  I believe that's on the
    order of a 486dx2-66.  
    
    Sytem performance would be (because of the local bus peripherals) on the 
    same order. Correct me if I'm wrong.

It really depends - 

The interleaved VRAM (ie, 64 bit access) will allow far more video 
bandwidth than with the DRAM used on cheap localbus video boards (where
you loose 45-90M/sec to the serializers if you can run at that resolution)
and the 40Mhz R3000 will do graphics primitaves faster than an S3.

While the 486 looks OK under some benchmarks (SpecFP), real 
world and other benchmarks are more favorable towards the R2000/R3000
FPU's even at a much lower clock speeds (not the 40/66 to 100/66 ratios
we'd have with the various proposed solutions).

In the 486's favor is the second level cache, which may have a significant
effect on it's relative performance in real world software.

The various MIPS chips have at least double the internal cache,
separate I & D so locality is maintained better, but you're going
to loose something without having a second level cache like 
the 486 boards have.  How much cache is the important issue here,
since there are diminishing returns and the hit rate depends on the 
design (set associative vs. direct mapped).  Some one needs to 
*research* the issue.

IMHO : there are probably more chips out there than the 
IDT3081/MOM combination, ie the 4200 or 4000PC and NEC ARC chipset,
that should be considered before a design of any kind happens.

Also, there are design issues that *need* resolving which will have 
a very significant effect on the final cost, market size,
and performance of the system.

IMHO the decision to provide an expansion path beyond the local bus
connector for Waldorf's board, memory, and pin-compatable chips, 
second level cache, and use of the NEC manufacturing kit rather than
our own design built arround the ARC chipset are the most significant
of these issues.

 
From owner-riscy@pyramid.com  Fri Aug  6 06:25:11 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA14829; Fri, 6 Aug 93 06:25:11 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10881; Fri, 6 Aug 93 03:25:13 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08563; Fri, 6 Aug 93 03:24:06 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08556; Fri, 6 Aug 93 03:24:04 -0700
Received: from RUTGERS.EDU 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10865; Fri, 6 Aug 93 03:25:01 -0700
Received: from eerie.acsu.buffalo.edu by rutgers.edu (5.59/SMI4.0/RU1.5/3.08) 
	id AA10003; Fri, 6 Aug 93 06:22:44 EDT
Received: from uucp@localhost by eerie (8.5/8.1)
	 via UUCP. id GAA27018; Fri, 6 Aug 1993 06:22:37 -0400
Received: by toz.buffalo.ny.us (1.65/waf)
	via UUCP; Thu, 05 Aug 93 16:36:25 EDT
	for eerie!pyramid.com!riscy
To: riscy@pyramid.com
Subject: Re: Mips project Go or Wait? 
From: cyberman@toz.buffalo.ny.us (Cyberman)
Comments:             
Message-Id: <PiTT8B1w165w@toz.buffalo.ny.us>
Date: Thu, 05 Aug 93 16:31:12 EDT
In-Reply-To: <199308050001.AA10724@kinglear.cs.Colorado.EDU>
Organization: The Tower of Zot - (716)839-0431
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Drew Eckhardt <eerie!kinglear.cs.Colorado.EDU!drew> writes:
> I don't think there will be any significant technical problems getting 
> the software up. 
> 
> Speaking of kernel development for the MIPS : we need to start a discussion
> on how to best make the kernel sources multi-architecture, so that one 
> release of Linux runs on MIPS, Intel, and Motorola platforms.
> 
> The AMIGA people didn't bother to do this, so they need a different 
> kernel distribution and can never take advantage of new features 
> when they're added.

WELL one way to do this is too do what the people who made nethack did.
Have people submit additions and patches and let them corelate 
modifications.  Also they when for system independance by using the
ussual conditional compiling.  They had one config.h for selecting the
OS options and one OS specific config.  In linux case perhaps a general
purpose config.h and a processor specific config and then a machine 
specific configuration.  The only problem I've had compiling nethack 
under this strategy are unimplemented system functions like stty gtty and 
lock.  I hate having to work around such things (grin).  

Cyberman

 "Everything is vanity ..."
"Even a fool looks wise if he keeps his mouth shut"
"A man to his own, is a person left to his own devices"
"Bewary of thinking you are always right, Hitler did"

Computer Engineer - hopeing to get employement
 
