[comp.sys.mac.hardware] Radius Rocket & Sys 7 question

george@hsvaic.boeing.com (George Williams) (06/26/91)

I have a Mac-II (vintage), and am considering adding a Radius Rocket
68040 board.  Questions:

Will I be able to run 32-bit virtual memory under system 7?
Will I need/be able to use the Connectix MODE32 solution?
Is the Rocket compatible with MCL 2.0 (Mac Common Lisp)?

Any help/suggestions/pointers would be appreciated.

George Williams
Boeing Computer Services   Internet: george@hsvaic.boeing.com
POBox 240002, M/S JY-58    UUCP: ...!uw-beaver!bcsaic!hsvaic!george
Huntsville AL 35824-6402   Phone: 205+464-4968 FAX: 205+464-4930 BTN: 464-4968

sjhg9320@uxa.cso.uiuc.edu (/dev/null ) (06/28/91)

If you can get ahold of the MacWeek Issues offered this week and last week,
they have a nice discussion of '040 and Radius Rocket compatibility issues.


--
================================================================================
|            Return unused portion to creator if not fully satisfied.          ||                  ----------------------------------------                    |================================================================================

johnL@radius.com (John Lee) (06/28/91)

george@hsvaic.boeing.com (George Williams)
Subject: Radius Rocket & Sys 7 question

> I have a Mac-II (vintage), and am considering adding a Radius Rocket
> 68040 board.  Questions:

> Will I be able to run 32-bit virtual memory under system 7?
> Will I need/be able to use the Connectix MODE32 solution?
> Is the Rocket compatible with MCL 2.0 (Mac Common Lisp)?


George,

I have some good news and some bad news.

First, the Rocket is not compatible with Rev A Mac ][ ROMs.  This is due to a 
well-known bug in the Apple ROM which was corrected in January 1980 on the Rev 
B ROM set.  The bug doesn't allow access to addresses above 1MB per NuBus slot.

Next, Macs of the ][cx variety and older inherently are not able to go into 32-
bit addressing mode and are stuck with 24-bit addressing.  Rocket does not 
change this fact as it uses the code in the motherboard's ROMs.

Next, while it is theoretically possible for MODE32 to work with Rocket, there 
is no announcement of such development being done.  However, I wouldn't be 
surprised if this actually were to come about in the not-too-distant future.

Finally, Radius has no report of Rocket and MCL bugs.  I guess I'll leave this 
one open to the community at large to correct me if such nasties actually occur.

As an aside, AbSoft told me yesterday that their FORTRAN package works 
wonderfully on the Rocket!

John Lee
Radius Developer Services

"Most opinions are mine, none are my employer's,
 and the facts belong to their respective owners!"

Internet:    johnL@radius.com
AppleLink:  RADIUS.QA  (ATTN: RDS)

paul@taniwha.UUCP (Paul Campbell) (06/29/91)

In article <1481@radius.com> johnL@radius.com (John Lee) writes:
>
>First, the Rocket is not compatible with Rev A Mac ][ ROMs.  This is due to a 
>well-known bug in the Apple ROM which was corrected in January 1980 on the Rev 
>B ROM set.  The bug doesn't allow access to addresses above 1MB per NuBus slot.

No - it's only a bug in the slot manager the II is quite capable of accessing
more than 1Mb/slot

>Next, Macs of the ][cx variety and older inherently are not able to go into 32-
>bit addressing mode and are stuck with 24-bit addressing.  Rocket does not 
>change this fact as it uses the code in the motherboard's ROMs.

of course they can - your company sells large screen boards that sit in
NuBus slots and have 4Mb of framestore - they do work on Mac IIs don't
they? A/UX runs in 32-bit mode on my II ......

The problem is not that the hardware inherently can't work in 32-bit mode
but that the memory manager in those systems can't - hence the need for
MODE32 or a ROM upgrade from Apple


	Paul

-- 
Paul Campbell    UUCP: ..!mtxinu!taniwha!paul     AppleLink: CAMPBELL.P

Tom Metzger's White Ayrian Resistance has been enjoined to stop selling Nazi
Bart Simpson t-shirts - Tom of course got it wrong, Bart is yellow, not white.