From owner-riscy@pyramid.com  Mon Jul 26 21:10:34 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA12674; Mon, 26 Jul 93 21:10:34 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA27233; Mon, 26 Jul 93 18:07:16 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13126; Mon, 26 Jul 93 18:06:07 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13117; Mon, 26 Jul 93 18:06:03 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA27204; Mon, 26 Jul 93 18:06:44 -0700
Message-Id: <9307270106.AA27204@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA07535; Mon, 26 Jul 93 21:05:00 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Mips Chips from NEC
To: riscy@pyramid.com (Mips 3000)
Date: Mon, 26 Jul 1993 21:05:00 -0500 (EDT)
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 1766      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

I received the data sheets on the 4000/4400 PC/SC

Here's a quick summary:  
	Part #       package   Power Cache 
4000PC	PD30400RJ-50 179 pin    5v   8k+8k
	PD30400RP-50 179 pin+HS 5v   8k+8k
4000SC	PD30401RJ-50 447 pin    5v   8k+8k
	PD30401RP-50 447 pin+HS 5v   8k+8k

4400PC	PD30410RJ-(50,67,75)  179 pin    5v 	16k+16k 
	PD30410RP-(50,67,75)  179 pin+HS 5v	16k+16k 
        PD30410LRJ-(50,67,75) 179 pin    3.3v 	16k+16k 
	PD30410LRP-(50,67,75) 179 pin+HS 3.3v	16k+16k 

4400SC	PD30411RJ-(50,67,75)  447 pin    5v 	16k+16k 
	PD30411RP-(50,67,75)  447 pin+HS 5v	16k+16k 
        PD30411LRJ-(50,67,75) 447 pin    3.3v 	16k+16k 
	PD30411LRP-(50,67,75) 447 pin+HS 3.3v	16k+16k 

HS means with heat-sink, all chips are ceramic. The last two digits of the
part number is the external clock speed, internal is twice that.

Main difference between 4000 and 4400 is double the cache and faster clocks.

Main difference between PC and SC is the the the SC supports a 128 bit 
secondary cache interface support.  But based on there diagrams the
PC supports a secondary cache, just not 128 bits wide.

All chips share: 
	64 bit microprocessor, integer, fp, registers, and virtual addresses.
	Dual instruction issue with no restrictions.
	32 64 bit general purpose registers and 16 64 bit fp registers.
	one level deep uncached store buffer.

The 4400 is around 94 specint and 105 specfp (seperate source)

The 4000 is listed as 62 specint, and 63 specfp.

The 4200 is mentioned in passing:  plastic, 1.5 Watts, 80 Mhz internal clock,
55 SPECint92 and 30 SPECfp92.

-- 
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 Jul 27 03:09:24 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA25908; Tue, 27 Jul 93 03:09:24 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA03932; Tue, 27 Jul 93 00:05:44 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08780; Tue, 27 Jul 93 00:04:33 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08759; Tue, 27 Jul 93 00:04:30 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA03913; Tue, 27 Jul 93 00:05:08 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.f) via EUnet
	id VA05170; Tue, 27 Jul 1993 09:01:16 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA01568; Tue, 27 Jul 93 09:48:09 GMT
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA24147; Tue, 27 Jul 93 09:00:35 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Tue, 27 Jul 93 08:59:34 +0200
Message-Id: <9307270659.AA10308@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA10308; Tue, 27 Jul 93 08:59:34 +0200
To: riscy@pyramid.com
Subject: Re:  Mips Chips from NEC
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> I received the data sheets on the 4000/4400 PC/SC
> 
> Here's a quick summary:  
> 	Part #       package   Power Cache 
>4000PC	PD30400RJ-50 179 pin    5v   8k+8k
>  	PD30400RP-50 179 pin+HS 5v   8k+8k
>4000SC	PD30401RJ-50 447 pin    5v   8k+8k
>	PD30401RP-50 447 pin+HS 5v   8k+8k
>
>4400PC	PD30410RJ-(50,67,75)  179 pin    5v 	16k+16k 
>	PD30410RP-(50,67,75)  179 pin+HS 5v	16k+16k 
>        PD30410LRJ-(50,67,75) 179 pin    3.3v 	16k+16k 
>	PD30410LRP-(50,67,75) 179 pin+HS 3.3v	16k+16k 
>
>4400SC	PD30411RJ-(50,67,75)  447 pin    5v 	16k+16k 
>	PD30411RP-(50,67,75)  447 pin+HS 5v	16k+16k 
>       PD30411LRJ-(50,67,75) 447 pin    3.3v 	16k+16k 
>	PD30411LRP-(50,67,75) 447 pin+HS 3.3v	16k+16k 
>
>HS means with heat-sink, all chips are ceramic. The last two digits of the
>part number is the external clock speed, internal is twice that.
>
>Main difference between 4000 and 4400 is double the cache and faster clocks.
>
>Main difference between PC and SC is the the the SC supports a 128 bit 
>secondary cache interface support.  But based on there diagrams the
>PC supports a secondary cache, just not 128 bits wide.
>
>All chips share: 
>	64 bit microprocessor, integer, fp, registers, and virtual addresses.
>	Dual instruction issue with no restrictions.
>	32 64 bit general purpose registers and 16 64 bit fp registers.
>	one level deep uncached store buffer.
>
>The 4400 is around 94 specint and 105 specfp (seperate source)
>
>The 4000 is listed as 62 specint, and 63 specfp.
>
>The 4200 is mentioned in passing:  plastic, 1.5 Watts, 80 Mhz internal clock,
>55 SPECint92 and 30 SPECfp92.

Yes, and the main difference between the R4200 and the other R4x00 CPUs
are that is *not* pincompatibel, that it has the old R3000 pipeline and
a simpler FPU. Note: This is what I've heard. I will receive data sheets
this week. Perhaps things have changed...
However, if the R4200 is no real R4x00 CPU I would suggest to use
one of the others, even if it is more expensive at the moment.
Are we going to build a R4000 machine now ? I would appreciate this !

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 Jul 27 03:47:40 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA26854; Tue, 27 Jul 93 03:47:40 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA06491; Tue, 27 Jul 93 00:40:47 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA10124; Tue, 27 Jul 93 00:39:50 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA10110; Tue, 27 Jul 93 00:39:48 -0700
Received: from csunix.urc.kun.nl 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA06479; Tue, 27 Jul 93 00:40:28 -0700
Received: by csunix.urc.kun.nl (5.64/1.37)
	id AA21809; Tue, 27 Jul 93 09:38:12 +0200
From: ronald%csunix.urc.kun.nl@kunrc1.urc.kun.nl (Ronald Schalk)
Message-Id: <9307270738.AA21809@csunix.urc.kun.nl>
Subject: Re: Mips Chips from NEC
To: riscy@pyramid.com
Date: Tue, 27 Jul 1993 09:38:12 +0200 (CET)
In-Reply-To: <9307270106.AA27204@gossip.pyramid.com> from "Bill Broadley" at Jul 26, 93 09:05:00 pm
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 1310      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> 
> I received the data sheets on the 4000/4400 PC/SC
> 
> Here's a quick summary:  

<specs of chips deleted>

> 
> The 4400 is around 94 specint and 105 specfp (seperate source)
> 
> The 4000 is listed as 62 specint, and 63 specfp.
> 
> The 4200 is mentioned in passing:  plastic, 1.5 Watts, 80 Mhz internal clock,
> 55 SPECint92 and 30 SPECfp92.

Do you also have some priceinformation on these things. Maybe it's also
interesting the keep in mind that it would be nice if there are
pin-compatible second-sourced versions available. This could drop the
price within a couple of months, especially because a LOT of sources are
predicting a HUGE rise in sales of R4K chips.


 ********************************************************************
 * ing. Ronald Schalk                                               *
 * sectie COOS                                                      *
 * Universitair Centrum Informatievoorziening (UCI)                 *
 * Katholieke Universiteit Nijmegen (KUN)                           *
 * e-mail : R.Schalk@uci.kun.nl   snailmail: Geert Grooteplein 41   *
 * tel.   : +31 80 617997                    6525 GA Nijmegen       *
 * fax   :  +31 80 617979                    Nederland              *
 ********************************************************************
 
 
From owner-riscy@pyramid.com  Wed Jul 28 04:17:03 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA18705; Wed, 28 Jul 93 04:17:03 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15095; Wed, 28 Jul 93 01:13:56 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA03905; Wed, 28 Jul 93 01:07:17 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA03897; Wed, 28 Jul 93 01:07:09 -0700
Received: from nag.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA12994; Wed, 28 Jul 93 01:07:30 -0700
Received: from localhost by nag.cs.Colorado.EDU with SMTP id AA02076
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Wed, 28 Jul 1993 02:05:56 -0600
Message-Id: <199307280805.AA02076@nag.cs.Colorado.EDU>
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC 
In-Reply-To: Your message of "Tue, 27 Jul 1993 08:59:34 MDT."
             <9307270659.AA10308@resi.waldorf-gmbh.de> 
Date: Wed, 28 Jul 1993 02:05:43 -0600
From: Drew Eckhardt <drew@nag.cs.Colorado.EDU>
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

    
    Yes, and the main difference between the R4200 and the other R4x00 CPUs
    are that is *not* pincompatibel, that it has the old R3000 pipeline and
    a simpler FPU. Note: This is what I've heard. I will receive data sheets
    this week. Perhaps things have changed...

    However, if the R4200 is no real R4x00 CPU I would suggest to use
    one of the others, even if it is more expensive at the moment.

    Are we going to build a R4000 machine now ? I would appreciate this !

Prices and performance
Notes : 
1.  Quantities were unspecified, in any case these prices are for 
	more chips than we're looking to buy, and for small lots the 
	distributor sets the price.

2.  All prices in US Dollars

3.  The figures for everything but the 4200 are from the July OEM magazine,
	I don't know where the 4200 info originated, but several net.sources
	said the same things.

4.  Clock speed is in Mhz, obviously these things run the internal clock
	faster than external.

Vendor		Clock 	Chip	Price	SPECint92 	SPECfp92
unspecified 	100	R4000SC 640	61.7		63.4
unspecified	150	R4400SC 1120	88(est)		97(est)
NEC 	        80	R4200   70	55		30


And, for comparison
Intel		66	486	542	32.2		16.1
Intel		66	Pentium	965	64.5		56.9

Unquestionably, the R4200 gives the best price/performance of 
any of these chips.  

After we adpoted the motherboard idea and people started calling this
thing a "workstation", a target price of $600-$700 was set.  It might
be a little difficult to meet a price somewhere in that neighborhood 
if we end up using a $640 chip...
 
From owner-riscy@pyramid.com  Wed Jul 28 04:46:48 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA19333; Wed, 28 Jul 93 04:46:48 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18208; Wed, 28 Jul 93 01:44:01 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA05666; Wed, 28 Jul 93 01:42:48 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA05652; Wed, 28 Jul 93 01:42:41 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18188; Wed, 28 Jul 93 01:43:17 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.f) via EUnet
	id WI01370; Wed, 28 Jul 1993 10:39:19 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA00988; Wed, 28 Jul 93 10:37:33 GMT
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA00897; Wed, 28 Jul 93 10:37:37 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Wed, 28 Jul 93 10:36:37 +0200
Message-Id: <9307280836.AA12449@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA12449; Wed, 28 Jul 93 10:36:37 +0200
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

>    Yes, and the main difference between the R4200 and the other R4x00 CPUs
>    are that is *not* pincompatibel, that it has the old R3000 pipeline and
>    a simpler FPU. Note: This is what I've heard. I will receive data sheets
>    this week. Perhaps things have changed...
>
>    However, if the R4200 is no real R4x00 CPU I would suggest to use
>    one of the others, even if it is more expensive at the moment.
>
>    Are we going to build a R4000 machine now ? I would appreciate this !

> Prices and performance
> Notes : 
> 1.  Quantities were unspecified, in any case these prices are for 
> 	more chips than we're looking to buy, and for small lots the 
> 	distributor sets the price.
> 
> 2.  All prices in US Dollars
> 
> 3.  The figures for everything but the 4200 are from the July OEM magazine,
> 	I don't know where the 4200 info originated, but several net.sources
> 	said the same things.
> 
> 4.  Clock speed is in Mhz, obviously these things run the internal clock
> 	faster than external.
> 
> Vendor		Clock 	Chip	Price	SPECint92 	SPECfp92
> unspecified 	100	R4000SC 640	61.7		63.4
> unspecified	150	R4400SC 1120	88(est)		97(est)
> NEC 	        80	R4200   70	55		30

Ok, but you can't compare the SC versions with the R4200, so far I know.
In any case, I would suggest to use the PC version.
What I have heard from IDT ( *OFFICIAL* statement, no net mumble ) is
that the price for the R4400PC will be around $200 in low quantities.
As I said already, I'm waiting for the NEC-Prices. If there's in fact
still a large price difference between the R4200 and others: Yes, we
should then use the R4200.

> And, for comparison
> Intel		66	486	542	32.2		16.1
> Intel		66	Pentium	965	64.5		56.9
> 
> Unquestionably, the R4200 gives the best price/performance of 
> any of these chips.  

> After we adpoted the motherboard idea and people started calling this
> thing a "workstation", a target price of $600-$700 was set.  It might
> be a little difficult to meet a price somewhere in that neighborhood 
> if we end up using a $640 chip...

We should try to get *real* prices, not just estimations.
Let us decide then, and don't let us confuse by unqualified
statements on the net. I totally agree that we should *not* try
to build a $2000 thing. Nobody would buy it, so nobody would
work in it. A price of around us$700 should be ok for a R4000
machine, isn't it ?

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 Jul 28 12:45:26 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA09330; Wed, 28 Jul 93 12:45:26 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA28915; Wed, 28 Jul 93 09:42:44 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA12253; Wed, 28 Jul 93 09:41:45 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA12245; Wed, 28 Jul 93 09:41:43 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA28874; Wed, 28 Jul 93 09:42:25 -0700
Message-Id: <9307281642.AA28874@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA07039; Wed, 28 Jul 93 12:40:44 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Re: Mips Chips from NEC
To: riscy@pyramid.com
Date: Wed, 28 Jul 1993 12:40:44 -0500 (EDT)
In-Reply-To: <199307280805.AA02076@nag.cs.Colorado.EDU> from "Drew Eckhardt" at Jul 28, 93 02:05:43 am
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 955       
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> Vendor	Clock 	Chip	Price	SPECint92 	SPECfp92
> unspecified 	100	R4000SC 640	61.7		63.4
> unspecified	150	R4400SC 1120	88(est)		97(est)
> NEC 	        80	R4200   70	55		30
  IBM		66 	601	350	55		80

Any reason the R4200 is being compared to the SC chips and not the PC?
Based on pin count alone I'd expect the PC at 179 pins to be a fair 
amount cheaper then the SC at 447 pins.  Or is the 4200 a SC??

How does the group feel this could come about?  Purchase the ARC 100 150
kits so that we can use there design (so we aren't limited to 40 Mhz)?

Apparently the powerpc 601 can use the RS/6000 port of GCC, I have several 
statements that the 601 will be supported in the next release.  Byte
magazine benchmarks the 601 with the GCC compiler.

-- 
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 Jul 29 11:09:11 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA01933; Thu, 29 Jul 93 11:09:11 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA04545; Thu, 29 Jul 93 08:04:45 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13834; Thu, 29 Jul 93 08:03:02 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13820; Thu, 29 Jul 93 08:02:59 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA04237; Thu, 29 Jul 93 08:03:41 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.g) via EUnet
	id YA15885; Thu, 29 Jul 1993 16:59:42 +0200
Received: from wegy
	by scotty.waldorf-gmbh.de with SMTP (5.65b/GEN-1.0.10)
	via EUnet for unido
	id AA00746; Thu, 29 Jul 93 16:51:58 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for scotty
	id AA01100; Thu, 29 Jul 93 16:52:00 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Thu, 29 Jul 93 16:51:03 +0200
Message-Id: <9307291451.AA00405@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA00405; Thu, 29 Jul 93 16:51:03 +0200
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com


> Any reason the R4200 is being compared to the SC chips and not the PC?
> Based on pin count alone I'd expect the PC at 179 pins to be a fair 
> amount cheaper then the SC at 447 pins.  Or is the 4200 a SC??

I don't know. So far I know is the R4200 a PC version.
It's *very* difficult to get informations about the R4200.
An engineer at NEC asked "no problem, how many do you want to buy ?"
while a distributor told me that he had never heard about that part
and that NEC doesn't produce it !
Seems we are a bit ahead of the technology...

However, I've got a data sheet for the NEC VRX / uPD30450. It looks
like that this is what we call the R4200. At least the detail that
this CPU has a 5-stage pipeline instead of the 8-stage pipeline used
in the R4000 and R4400 makes it very probably that it is the R4200.

Some Specs quoted from the sheet (Feb. 93):

Input Clock: 		40 MHz
Operating Clock:	80 MHz
I-Cache:		16 kB
D-Cache:		 8 kB
Power Consumption:	 2 W

System interface compatibel with the Vr4000 series.
Instruction set compatibel with the Vr4000 series.

So far, so good. But there's one important thing, probably a great
disadvantage for some of the group: The VRX has ***NO*** FPU !!!

Again, qouted from the data sheet:

All floating point instructions, as defined in the MIPS ISA for
the floating-point coprocessor CP1, are processed by the same
hardware as for the integer instructions. That is to say, there
is no separate floating point coprocessor. However, the execution
of floating point instructions can still be disabled via the
coprocessor usability CU bit defined in the CP0 status register.
[end of qouted material]

Here are some examples of floating point instruction timings:
(in pipeline clock cycles)

Insn		S	D
--------------------------
Add.fmt		3	3
Sub.fmt		3	3
Mul.fmt		11	20
Div.fmt		28	57
Sqrt.fmt	28	57
Abs.fmt		1	1
[...]
S=single prec., D=double prec.

So far I understand the data sheet, these instructions
are executed in parallel to loads, stores and so on.
The CPU stalls when, for example, an int-multiply is
in progress and a second multiply, no matter what format,
is to be executed.
I would say, except of Mul and Div, this is the same
timing as the R4000 and R4400 have. I can't decide
if the performance of Mul and Div satisfies the requirements
of the group. All I can say is: It's ok for *me*.

Any comments ? Please - if we really want to make hardware
instead of vapor-ware, wakeup, guys. Here's something to discuss.

Cheers,
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 Jul 29 11:35:53 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA03393; Thu, 29 Jul 93 11:35:53 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09291; Thu, 29 Jul 93 08:34:23 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA15623; Thu, 29 Jul 93 08:33:23 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA15616; Thu, 29 Jul 93 08:33:20 -0700
Received: from cs.huji.ac.il 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA09215; Thu, 29 Jul 93 08:33:59 -0700
Received: from picton.cs.huji.ac.il by cs.huji.ac.il with SMTP id AA14948
  (5.65b/HUJI 4.114); Thu, 29 Jul 93 18:32:04 +0300
Received: from localhost by picton.cs.huji.ac.il with SMTP id AA05905
  (5.65c/HUJI 4.121 for <riscy@pyramid.com>); Thu, 29 Jul 1993 18:32:21 +0300
Message-Id: <199307291532.AA05905@picton.cs.huji.ac.il>
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC 
In-Reply-To: Your message of Thu, 29 Jul 93 16:51:03 +0200 .
             <9307291451.AA00405@resi.waldorf-gmbh.de> 
From: Amos Shapira <amoss@cs.huji.ac.il>
Date: Thu, 29 Jul 1993 18:32:14 +0300
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

In message <9307291451.AA00405@resi.waldorf-gmbh.de> Andreas Busse writes:

 [ lots of info,  bottom line is that the R4200 doesn't have a FPU ]

|Any comments ? Please - if we really want to make hardware
|instead of vapor-ware, wakeup, guys. Here's something to discuss.

Ok,  so shall we take a vote or what?

I am not a hardware type but the R4200 seems to be enough for me it and it
eats 486/66 raw even as it is (here is the table again:

Vendor          Clock   Chip    Price   SPECint92       SPECfp92
unspecified     100     R4000SC 640     61.7            63.4
unspecified     150     R4400SC 1120    88(est)         97(est)
NEC             80      R4200   70      55              30
                                                        ^^
And, for comparison
Intel           66      486     542     32.2            16.1
                                                        ^^^^
Intel           66      Pentium 965     64.5            56.9
)

and the Pentium is out of the question price-wise as far as I'm concerned,
not to mention that going for it must mean waiting for another year at least.

Especially when you take the price/performance and at what price do we want
the system to end up with, I vote for the R4200 and let's go on with the rest
of the design.

Only one question (not that it is going to change my view): you mention
a possibility of a separate FPU.  Is it possible to leave space for a socket
for it a-la the 386/387 combo?  (again,  I'm not a hardware type so please
bear with me if this is a stupid question).

I hope I'm doing the right thing but I propose that others on the list who
kept quite so far (like me) and followed the discussion for the last couple
of weeks will let us all know what they think (is there an echo out here ? :).

Maybe the "core" of this list should appruve of this before things get out
of hand, ok?

|
|Cheers,
|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
|------------------------------------------------------------------------------
|-

Bye,

--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  Thu Jul 29 11:38:53 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA03595; Thu, 29 Jul 93 11:38:53 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10166; Thu, 29 Jul 93 08:38:05 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA15924; Thu, 29 Jul 93 08:37:07 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA15917; Thu, 29 Jul 93 08:37:05 -0700
Received: from neurocog.lrdc.pitt.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10130; Thu, 29 Jul 93 08:37:50 -0700
Message-Id: <9307291537.AA10130@gossip.pyramid.com>
Received: by neurocog.lrdc.pitt.edu
	(1.37.109.4/16.2) id AA14442; Thu, 29 Jul 93 11:35:46 -0400
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Subject: Re: Mips Chips from NEC
To: riscy@pyramid.com
Date: Thu, 29 Jul 1993 11:35:45 -0500 (EDT)
In-Reply-To: <9307291451.AA00405@resi.waldorf-gmbh.de> from "Andreas Busse" at Jul 29, 93 04:51:03 pm
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 1332      
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

> So far, so good. But there's one important thing, probably a great
> disadvantage for some of the group: The VRX has ***NO*** FPU !!!

As far as I'm concerned we shouldn't worry about implementation as
much as the benchmarks.   Whatever has the best price performance
sounds good to me.  Although the fact that a 4000 can be upgraded
to a 4400 sounds like a good advantage to me.

The 4200 is around 55/30 specint/specfp
    4000 is around 62/63
    4400 is around 94/105
    601 is around 50/80

Of the chips we have prices for (4[04]00SC and 4200) the 4200 sounds the most
attractive, but (PC) prices may make the 4000 much more attractive.

I think another critical issue is whether we go with the ARC 100/150 
manufacturing kits.  If we do we can do 50 Mhz or higher, and have to use the 
mips chip.

Powerpc seems to be in the running as far as price/performance and
running gcc+gas is concerned.

> Any comments ? Please - if we really want to make hardware
> instead of vapor-ware, wakeup, guys. Here's something to discuss.

The cost difference between a 4000PC and the 4200 is the only thing I need
to decide.


-- 
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 Jul 29 11:44:44 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA03994; Thu, 29 Jul 93 11:44:44 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10874; Thu, 29 Jul 93 08:43:46 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA16348; Thu, 29 Jul 93 08:42:49 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA16339; Thu, 29 Jul 93 08:42:47 -0700
Received: from boulder.Colorado.EDU 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA10826; Thu, 29 Jul 93 08:43:31 -0700
Received: from agua.Colorado.EDU by boulder.Colorado.EDU with SMTP id AA16191
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Thu, 29 Jul 1993 09:41:23 -0600
Received: by agua.colorado.edu (cu.generic.890828)
Date: Thu, 29 Jul 93 09:38:58 MDT
From: Alan Krantz <atk@agua.Colorado.EDU>
Message-Id: <9307291538.AA27123@agua.colorado.edu>
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

  >>From owner-riscy@pyramid.com Thu Jul 29 09:31:32 1993
  >>From: Amos Shapira <amoss@cs.huji.ac.il>
  >>Date: Thu, 29 Jul 1993 18:32:14 +0300

  >>In message <9307291451.AA00405@resi.waldorf-gmbh.de> Andreas Busse writes:

  >> [ lots of info,  bottom line is that the R4200 doesn't have a FPU ]

  >>|Any comments ? Please - if we really want to make hardware
  >>|instead of vapor-ware, wakeup, guys. Here's something to discuss.

  >>Ok,  so shall we take a vote or what?

  >>I am not a hardware type but the R4200 seems to be enough for me it and it
  >>eats 486/66 raw even as it is (here is the table again:

  >>Vendor          Clock   Chip    Price   SPECint92       SPECfp92
  >>unspecified     100     R4000SC 640     61.7            63.4
  >>unspecified     150     R4400SC 1120    88(est)         97(est)
  >>NEC             80      R4200   70      55              30
  >>                                                        ^^
  >>And, for comparison
  >>Intel           66      486     542     32.2            16.1
  >>                                                        ^^^^

Price, performance, and ease of building a mother board how does some of the
other available risc processors compare such as 601 and the new sparc ?

atk

 
From owner-riscy@pyramid.com  Thu Jul 29 13:42:17 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA10285; Thu, 29 Jul 93 13:42:17 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA08648; Thu, 29 Jul 93 10:41:09 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28112; Thu, 29 Jul 93 10:40:02 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA28081; Thu, 29 Jul 93 10:39:59 -0700
Received: from ubiserver.ubitrex.mb.ca 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA08511; Thu, 29 Jul 93 10:40:36 -0700
Received: from ska.ubitrex.mb.ca ([192.75.16.23]) by ubitrex.mb.ca (4.1/SMI-4.1)
	id AA24770; Thu, 29 Jul 93 12:38:50 CDT
Date: Thu, 29 Jul 93 12:38:50 CDT
From: tim@ubitrex.mb.ca (Tim Braun)
Message-Id: <9307291738.AA24770@ubitrex.mb.ca>
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

My quick, off-the-cuff opinion:

1. A 3081/3051 design is still feasible, and may be a price/performance
   competitor - if the price is low enough to fit the lower performance.
   I note that IBM is now selling motherboards with 486slc/2's on them
   for about CDN$450.00.  That's a 486/33 class machine. (25/50 MHz clock
   doubled.)  16 KByte on-chip cache, no secondary cache.  We should
   hit 1.5-2 times that performance, with on-board scsi/serial/fdc/video
   for 1.5 times the price.  That's sensible to me.  The R3081/MOM3000/R3041
   combination is interesting.

2. An R4000SC design is too complex for my tastes, and secondary cache
   is too expensive (ca. $100 including support chips).  The performance
   information given is interesting as upper-bound numbers.

3. An R4000PC design may be feasible.  But we don't have cost information
   or performance information.  I really want to stay under $800 (no ram).

4. An R4200 design may be feasible, but no-one knows what it will cost.
   I'll start bugging the NEC distributor here.  We have a good relationship
   with him.

5. A PowerPC design is a dark horse.  Motorola will be aching to get design
   information out to design teams, but if it requires secondary caches
   and 60 MHz design, it will be tricky and expensive.

6. The SPARC chips look like they require significant design effort, and
   have historically been priced > $400.  Witness the Weitek "SPARC1 upgrade"
   that retails at $1500.  Certainly SMCC and other high-volume purchasers
   get better prices, but can we?

My 1992 NEC Databook on DSP's has a one-page table of the RISC products
which includes:

NEC part  Mkt name   Clock (MHz)  Pkg        pins
uPD30361  Vr3600     25, 33       PPGA       175
uPD30400  Vr4000PC   50, 66, 75   CPGA       179
uPD30401  Vr4000SC   50, 66, 75   CPGA, LGA  447
uPD30402  Vr4000MC   50, 66, 75   CPGA, LGA  447

I do have advance information on the Orion from IDT.  This is R4000PC
pin compatible (PGA) or 208-pin MQuad.  Hmmm. 5-stage pipeline like the
Vr4200.  And what is the R4400PC?  Note that some of these devices
(Orion) are 3.3V.
________________________________________________________________
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 Jul 29 15:20:49 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA15767; Thu, 29 Jul 93 15:20:49 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA25603; Thu, 29 Jul 93 12:19:45 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13813; Thu, 29 Jul 93 12:17:44 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13801; Thu, 29 Jul 93 12:17:39 -0700
Received: from ubiserver.ubitrex.mb.ca 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA25086; Thu, 29 Jul 93 12:18:21 -0700
Received: from ska.ubitrex.mb.ca ([192.75.16.23]) by ubitrex.mb.ca (4.1/SMI-4.1)
	id AA25084; Thu, 29 Jul 93 14:16:35 CDT
Date: Thu, 29 Jul 93 14:16:35 CDT
From: tim@ubitrex.mb.ca (Tim Braun)
Message-Id: <9307291916.AA25084@ubitrex.mb.ca>
To: riscy@pyramid.com
Subject: Re: Mips Chips from NEC
Sender: owner-riscy@pyramid.com
Reply-To: riscy@pyramid.com

Just perusing the latest (July 22nd) EDN -

Fujitsu MB86932 SPARClite - US$79 - 20 MHz,  US$120 - 40 MHz, 5K qty.
  DRAM controller, 2 KByte D-cache, 8 KByte I-cache, "1 MFlop" (no
  FPU), 2 DMA channels, timer on-chip, address decoder on-chip, 
  interrupt controller.

Cyrix Cx486s2/50 - US$249, 1k qty (just for $ comparison).

DEC 21064 - Alpha - US$1096, 1k qty.  I don't think we're talking
  Alpha this year.

Intel Pentium - "less than $1000".

AT&T Hobbit - $100 CPU & chipset (5 chips). 10k qty.  No gcc.

TI uSPARC - US$179, 10k qty.  Offchip primary cache required.

I don't think we should use any of these, for various price and support
reasons.  But the price/performance issue is part of what's driving this
list, so perspective is helpful.  But that SPARClite (targetted at
embedded/X/laser printers) is interesting.   Hmmm, a $500 SS-2
motherboard?  I guess that's the Classic's price.

________________________________________________________________
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
 
