From owner-riscy@pyramid.com  Sun Jul 25 04:41:06 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA19420; Sun, 25 Jul 93 04:41:06 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10167; Sun, 25 Jul 93 01:39:20 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA23286; Sun, 25 Jul 93 01:37:15 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA23245; Sun, 25 Jul 93 01:37:12 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10151; Sun, 25 Jul 93 01:37:50 -0700
Message-Id: <9307250837.AA10151@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA05450; Sun, 25 Jul 93 04:36:15 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Mips on vesa
To: riscy@pyramid.com (Mips 3000)
Date: Sun, 25 Jul 1993 04:36:15 -0500 (EDT)
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 862       
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

The original mips design seemed to be basically mips+simm slots on ISA bus and
seems to have been dropped because of ISA limitations and board space.

(see previous msg about the commercial version for mips 4k on VESA for $1750)

How about one simpler.  Just a mips chip (whatever we can afford)+cache 
on a vesa card.  As I understand it VESA is basically the pc memory bus.

Running on the VESA bus at 25-33 Mhz will give you 100-128 MB/sec bandwdith
to memory, this makes a local cache important but seems workable.

Everything would be reused.  People could buy the ISE/VESA card of there choice.
A 486/25 sx is $215 or so would be used as a host.

-- 
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  Mon Jul 26 03:59:34 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA02897; Mon, 26 Jul 93 03:59:34 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09701; Mon, 26 Jul 93 00:57:31 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA06556; Mon, 26 Jul 93 00:56:14 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA06548; Mon, 26 Jul 93 00:56:11 -0700
Received: from ophelia.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09689; Mon, 26 Jul 93 00:56:51 -0700
Received: from localhost by ophelia.cs.Colorado.EDU with SMTP id AA05290
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Mon, 26 Jul 1993 01:56:07 -0600
Message-Id: <199307260756.AA05290@ophelia.cs.Colorado.EDU>
To: riscy@pyramid.com
Subject: Re: Mips on vesa 
In-Reply-To: Your message of "Sun, 25 Jul 1993 04:36:15 CDT."
             <9307250837.AA10151@gossip.pyramid.com> 
Date: Mon, 26 Jul 1993 01:56:06 -0600
From: Drew Eckhardt <drew@ophelia.cs.Colorado.EDU>
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


    The original mips design seemed to be basically mips+simm slots 

Plus cheap onboard SCSI, serial ports, video, and perhaps ethernet.
    
    on ISA bus and

Not on an ISA bus, but with an ISA bus for people who insist on 
having PC-compatable expansion slots. 

    seems to have been dropped because of ISA limitations and board space.

IMHO, it seems to have been postponed because of

1. No suitable DMA/DRAM/peripherial controller is available - 
	the 3730 isn't shipping and the Visual MOM chipset lacks 
	support for slave mode DMA. (Has anyone looked at the 
	NEC chipset?)

2.  The 4200 would be faster, and should be cheaper than the 
	3081.  Availability at this point is questionable.$

3.  In 6-9 months, IDT ships the Orion, an R4000 compatable chip.
	Also faster and cheaper than the 3081.

    How about one simpler.  Just a mips chip (whatever we can afford)+cache 
    on a vesa card.  

It depends on what our goals are. IMHO, we need to clearly define what these
are before getting furthur into the technical aspects of things.

    As I understand it VESA is basically the pc memory bus.
    
    Running on the VESA bus at 25-33 Mhz will give you 100-128 MB/sec bandwdith
    to memory, this makes a local cache important but seems workable.
    
    Everything would be reused.  People could buy the ISE/VESA card of there ch
   oice.
    A 486/25 sx is $215 or so would be used as a host.

For the people who want a "workstation", the motherboard solution
wins hands down.

+ $200 SCSI (You can't buy large IDE drives at any price)
+ $200 video (VRAM based local bus)
+ $50 16550 serial board
+ ethernet ???

Total : $615 + ???

Compared to a motherboard solution of 

+ $90 main board
+ $25 SCSI
+ $20 (serial ports, either one of the integrated chips or a 16552)
+ $100 video (1M, VRAM based (ie, 1152x900x256) $60 for 2M )
+ ethernet ???

Total : $235 + ???


 
From owner-riscy@pyramid.com  Mon Jul 26 08:41:46 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA08964; Mon, 26 Jul 93 08:41:46 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA13247; Mon, 26 Jul 93 05:39:49 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA25152; Mon, 26 Jul 93 05:38:27 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA25145; Mon, 26 Jul 93 05:38:24 -0700
Received: from quasi.vmars.tuwien.ac.at 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA13227; Mon, 26 Jul 93 05:39:02 -0700
Received: by quasi.vmars.tuwien.ac.at id AA08234
  (5.64+/IDA-1.3.4 for riscy@pyramid.com); Mon, 26 Jul 93 14:36:15 +0200
From: Peter Holzer <hp@quasi.vmars.tuwien.ac.at>
Message-Id: <9307261236.AA08234@quasi.vmars.tuwien.ac.at>
Subject: Re: Mips on vesa
To: riscy@pyramid.com
Date: Mon, 26 Jul 93 14:36:13 MET DST
In-Reply-To: <199307260756.AA05290@ophelia.cs.Colorado.EDU>; from "Drew Eckhardt" at Jul 26, 93 1:56 am
Reply-To: hp@vmars.vmars.tuwien.ac.at
X-Mailer: ELM [version 2.3 PL5]
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

You (Drew Eckhardt) wrote:
> 
> 
>     The original mips design seemed to be basically mips+simm slots 
>     on ISA bus and
> 
> Not on an ISA bus, but with an ISA bus for people who insist on 
> having PC-compatable expansion slots. 

No, the original project was indeed, as Bill wrote, to build a card
with a MIPS processor and some memory to put into a PC. This was soon
changed to build a motherboard, however (But you should know that,
Drew. The first mail I got on this list was by Neil answering (among
other things) your claim that there isn't enough real estate on an isa
board).

> IMHO, it seems to have been postponed because of
> 
> 1. No suitable DMA/DRAM/peripherial controller is available - 
> 	the 3730 isn't shipping and the Visual MOM chipset lacks 
> 	support for slave mode DMA. (Has anyone looked at the 
> 	NEC chipset?)
> 
> 2.  The 4200 would be faster, and should be cheaper than the 
> 	3081.  Availability at this point is questionable.$
> 
> 3.  In 6-9 months, IDT ships the Orion, an R4000 compatable chip.
> 	Also faster and cheaper than the 3081.

It seems so. But this happened only recently, whereas Bill was talking
about a decision made a month ago. But I don't like going back to the
mips+ram-on-a-card idea. I like it even less if its going to be a vesa
card. I have an old 25MHz 386 without vesa. When I buy a new computer
it it will either be a high-end 486 or some RISC workstation. I will
not buy a 486sx just to plug a crippled mips motherboard into it.

	hp

-- 
|    _  | Peter J. Holzer                       | Think of it   |
| |_|_) | Technical University Vienna           | as evolution  |
| | |   | Computer Science/Real-Time Systems    | in action!    |
| __/   | hp@vmars.tuwien.ac.at                 |     Tony Rand |
 
From owner-riscy@pyramid.com  Mon Jul 26 09:27:56 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA10498; Mon, 26 Jul 93 09:27:56 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16376; Mon, 26 Jul 93 06:24:30 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28155; Mon, 26 Jul 93 06:23:37 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28148; Mon, 26 Jul 93 06:23:31 -0700
Received: from nag.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16364; Mon, 26 Jul 93 06:24:10 -0700
Received: from localhost by nag.cs.Colorado.EDU with SMTP id AA25469
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Mon, 26 Jul 1993 07:23:12 -0600
Message-Id: <199307261323.AA25469@nag.cs.Colorado.EDU>
To: hp@vmars.vmars.tuwien.ac.at, riscy@pyramid.com
Subject: Re: Mips on vesa 
In-Reply-To: Your message of "Mon, 26 Jul 1993 14:36:13 MST."
             <9307261236.AA08234@quasi.vmars.tuwien.ac.at> 
Date: Mon, 26 Jul 1993 07:23:11 -0600
From: Drew Eckhardt <drew@nag.cs.Colorado.EDU>
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


--------

    You (Drew Eckhardt) wrote:
    > 
    > 
    >     The original mips design seemed to be basically mips+simm slots 
    >     on ISA bus and
    > 
    > Not on an ISA bus, but with an ISA bus for people who insist on 
    > having PC-compatable expansion slots. 
    
    No, the original project was indeed, as Bill wrote, to build a card
    with a MIPS processor and some memory to put into a PC. 
    changed to build a motherboard, however (But you should know that,
    Drew. The first mail I got on this list was by Neil answering (among
    other things) your claim that there isn't enough real estate on an isa
    board).

If we're being pedantic, yes.  However, Neil quickly conceeded that we 
needed a motherboard if we were going to get any performance out of the
system as a whole, and all recent discussions have been about a system
with a mother board.
    
    > IMHO, it seems to have been postponed because of
    > 
    > 1. No suitable DMA/DRAM/peripherial controller is available - 
    > 	the 3730 isn't shipping and the Visual MOM chipset lacks 
    > 	support for slave mode DMA. (Has anyone looked at the 
    > 	NEC chipset?)
    > 
    > 2.  The 4200 would be faster, and should be cheaper than the 
    > 	3081.  Availability at this point is questionable.$
    > 
    > 3.  In 6-9 months, IDT ships the Orion, an R4000 compatable chip.
    > 	Also faster and cheaper than the 3081.
    
    It seems so. But this happened only recently, whereas Bill was talking
    about a decision made a month ago. But I don't like going back to the
    mips+ram-on-a-card idea. 

Agreed.  Performance and total system price would be lousy for people 
looking for a workstation/complete system, who seem to be the majority
of the riscy list readers.

    I like it even less if its going to be a vesa card. I have an old 
    25MHz 386 without vesa. When I buy a new computer it it will either be 
    a high-end 486 or some RISC workstation. 

Agreed.  

With the VESA solution, you get a PC with a MIPS CPU, not a workstation 
with reasonable I/O and video.

    I will not buy a 486sx just to plug a crippled mips motherboard into it.

Like I said, the motherboard solution makes the most sense (performance
and price wise) for the people who want a "workstation", ie the majority
of the riscy list readers.

For people wanting a *simple* solution *now* and not in six months, the 
VESA board is the only viable choice.

NEC's 4200 and chipset look like they're happening now and not in six
months, so I'll talk to NEC today and see if I can get some information 
on availability of their 4200 and chipset, as well as technical 
specifications (the Visual MOM chipset sounded great until we got the 
specs that showed it had no slave mode DMA, no variable bus sizing for 
peripherials, etc).

--------

 
From owner-riscy@pyramid.com  Mon Jul 26 09:41:46 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA11080; Mon, 26 Jul 93 09:41:46 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16562; Mon, 26 Jul 93 06:39:38 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28294; Mon, 26 Jul 93 06:38:36 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28287; Mon, 26 Jul 93 06:38:34 -0700
Received: from quasi.vmars.tuwien.ac.at 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16550; Mon, 26 Jul 93 06:38:23 -0700
Received: by quasi.vmars.tuwien.ac.at id AA09694
  (5.64+/IDA-1.3.4 for riscy@pyramid.com); Mon, 26 Jul 93 15:35:30 +0200
From: Peter Holzer <hp@quasi.vmars.tuwien.ac.at>
Message-Id: <9307261335.AA09694@quasi.vmars.tuwien.ac.at>
Subject: Re: Mips on vesa
To: riscy@pyramid.com
Date: Mon, 26 Jul 93 15:35:29 MET DST
In-Reply-To: <199307261323.AA25469@nag.cs.Colorado.EDU>; from "Drew Eckhardt" at Jul 26, 93 7:23 am
Reply-To: hp@vmars.vmars.tuwien.ac.at
X-No-Return-Receipt-To:  hp@vmars.tuwien.ac.at
X-Mailer: ELM [version 2.3 PL5]
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

You (Drew Eckhardt) wrote:
> 
>     
>     No, the original project was indeed, as Bill wrote, to build a card
>     with a MIPS processor and some memory to put into a PC. 
> 
> If we're being pedantic, yes.  However, Neil quickly conceeded that we 
> needed a motherboard if we were going to get any performance out of the
> system as a whole, and all recent discussions have been about a system
> with a mother board.

Yup.

> Like I said, the motherboard solution makes the most sense (performance
> and price wise) for the people who want a "workstation", ie the majority
> of the riscy list readers.

Agreed.

> For people wanting a *simple* solution *now* and not in six months, the 
> VESA board is the only viable choice.

Well, `now' is probably somewhat too optimistic, but I agree that it
will be completed in a lot less than 6 months, since all the necessary
parts are available, so all we have to do is design it. However, as you
said the majority of the risciers is interested in a workstation. How
many people could live with a vesa board? I certainly can't. I can
however wait for six months or even a year (of course by then there
might be R4000-based workstations available for similar cost).

> NEC's 4200 and chipset look like they're happening now and not in six
> months, so I'll talk to NEC today and see if I can get some information 
> on availability of their 4200 and chipset, as well as technical 
> specifications .

Hope you get something.

	hp

-- 
|    _  | Peter J. Holzer                       | Think of it   |
| |_|_) | Technical University Vienna           | as evolution  |
| | |   | Computer Science/Real-Time Systems    | in action!    |
| __/   | hp@vmars.tuwien.ac.at                 |     Tony Rand |
 
