NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/sys/apple2/26036

Text File | 1992-12-27 | 5.9 KB | 154 lines

Newsgroups: comp.sys.apple2
Path: sparky!uunet!usc!rpi!sarah!cook!psinntp!psinntp!blkbox!jdb8042
From: jdb8042@blkbox (John D. Baker)
Subject: 65816 in //e, //c, //c+ works.
Organization: The Black Box, PO Box 591822 Houston, TX 77259-1822
Date: Sun, 27 Dec 1992 04:45:56 GMT
Message-ID: <1992Dec27.044556.20987@blkbox>
Summary: Here's the proof.
Keywords: 65816, 65c02, 65802, //e, //c, //c+
Lines: 142

I posted here the other day about receiving two reports of people using
65816 CPUs in their //e and //c (and //c+) and having it work just fine
as a 65802.

Well, here they are.

--------------- begin 65816 in //e, //c, //c+ messages -----------------


*** Message #1:  ***

From acme.fred.org!greggb Thu Sep 17 17:46:32 1992
Return-Path: <[e-mail]>
Received: by tamuts.tamu.edu (/\../\ Smail3.1.14.4 #14.6)
    id <[e-mail]>; Thu, 17 Sep 92 17:46 MDT
Received: by acme.fred.org (5.65/1.35)
    id AA01132; Thu, 17 Sep 92 15:39:44 -0700
From: [e-mail] (Gregg D. Brekke)
Message-Id: <[e-mail]>
Subject: Re: 65816 in //c
To: [e-mail] (John Donald Baker)
Date: Thu, 17 Sep 92 15:39:43 PDT
In-Reply-To: <[e-mail]>; from "John Donald Baker" at
 Sep 18, 92 05:28:00 pm
X-Mailer: ELM [version 2.4dev PL17]
Status: RO

John D. Baker writes:
> I have been told that you managed to install a 65816 processor in an
>   Apple //c
> and have it function as a 65802.
>
> I'd like to learn more about this, if you please.  Just a matter of popping
>    one
> in or was something more involved?
>
> Thanks for your help.
>

Yeah that is right, I just popped it in.  I took a 65816 out of my ROM3
GS when I replaced the 65816 with a TransWarpGS.  The 65816 worked fine
in an OLD (1984) //c and now works fine in my //c+.  Some have said it
is impossible, but I have had no problems with ANY software.  Certain
software that only runs on a 65816 or 65802 (such as Merlin 16) works
great.

If you have problems, tell me what they are like.  I may be able to help.

Good luck,
Gregg

---------------------------------------------------------------------------
 Gregg D. Brekke       Director of Network Operations, FrEdMail Foundation
 [e-mail]            FrEdMail - The K-12 link to the InterNet



*** Message #2:  ***


From socs.uts.EDU.AU!posty Fri Sep 18 00:05:04 1992
Return-Path: <[e-mail]>
Received: by tamuts.tamu.edu (/\../\ Smail3.1.14.4 #14.6)
    id <[e-mail]>; Fri, 18 Sep 92 00:04 MDT
Received: from syzygy.socs.uts.EDU.AU by ultima.socs.uts.EDU.AU
   (5.65+/SMI-3.3)
    id AA21445; Fri, 18 Sep 92 15:04:14 +1000
Received: by syzygy.socs.uts.EDU.AU (4.1/SMI-4.1)
    id AA15569; Fri, 18 Sep 92 15:03:57 EST
From: [e-mail]
Message-Id: <[e-mail]>
Subject: Re: ZIP CHip
To: [e-mail] (John Donald Baker)
Date: Fri, 18 Sep 92 15:03:55 EST
In-Reply-To: <[e-mail]>; from "John Donald Baker" at
 Sep 17, 92 5:46 pm
X-Mailer: ELM [version 2.3 PL11]
Status: RO

> In article <posty.716694778@syzygy> you write:
> >It's amazing how much nicer Merlin16 is to use on a //e over Merlin8.
> >What other applications do you run which are 16bit dependent?
>
> Actually, I don't run any 16-bit applications.  It's a matter of the block
> memory move instructions so that screen I/O is blazingly fast.

Well, whatever works best :).

> Thanks for the feedback.  tell me more about using an '816 in a //e,
>   please.
> I'm fascinated by this possibility.

Well, I knew about the 802 being pin compatible and being able to use Merlin
16 on the //e with it. One day was desperate to get Merlin 16 running on the
//e while testing an application on my //gs (they are linked thru a common
hard drive, so they can easily share files). I remebered the 65816 lying in
my TWGS box, so i pulled out the "Programming the 6502 including the 65802,
65816" reference (Eyes & Lichty) to check the pins on the '816 for
compatibility. The pins that had changed function were not connected on the
//e (thanks to the //e Tech Ref), so I became daring and plugged it straight
in.

Merlin 16 uses the native mode of the processor to improve performance while
still remaining a P8 application restricted to 128k of memory.

Even though the processor speed is faster (2.8 compared to 1 Mhz) there will
be no increase in system performance because everything is timed at 1Mhz.

As the Zip Chip cannot have the processor removed, the only way to 'attempt'
(because as far as I know, nobody has done it) to speed up a //e with a
65802 or 65816 is to replace the 6502 on the Transwarp //e board.

As for general use of the '802 or '816 in a //e, you just have to use an
assembler which understands the extended opcodes... Debugging is not much
fun because naturally the //e dissassembler in monitor knows nothing about
extended opcodes.

One problem that I've come across so far is that I can't boot from my CMS
hard drive using the CMS SCSI card. ProDOS has no problems in using the
drive once it is loaded, however. My gut feeling is that the boot code in
the card's ROM is timing dependent and uses codes that differ in cycle counts
between the 65c02 and the '816 in emulation mode. I'm probably wrong though.
It just doesn't work for me.

regards,
Andrew

------------------ end 65816 in //e, //c, //c+ messages --------------------


Now, are there any people out there who have installed accelerators in there
IIgs machines who would be willing to part with their 65816's (40-pin DIP)
that they removed from the motherboard.  How much would you ask for yours?


John D. Baker  ->An Apple CardZ180 Z-System nut//
Internet:  [e-mail]            [e-mail]
UUCP:      nuchat!blkbox!jdb8042         ...!taronga!jdbaker
BBSs:  JOHN BAKER on Z-Node #45 [[phone]],
The Vector Board [[phone]], PIC of the Mid-Town [[phone]]
Karnage: "I am certainly very happy for you, but--WHAT ARE YOU TALKING ABOUT?!"