This group...

106 views
Skip to first unread message

Mux

unread,
Oct 27, 2015, 4:11:50 PM10/27/15
to
is still alive and kicking and I think that's awesome! Anyone care to share any other groups you're part of that's 68k related?

-Mux

Zach Metzinger

unread,
Feb 24, 2016, 2:42:07 AM2/24/16
to
On 10/27/15 15:11, Mux wrote:
> is still alive and kicking and I think that's awesome! Anyone care to share any other groups you're part of that's 68k related?
>
> -Mux
>

Yep, pretty cool that m68k is alive and well.

I've come into a supply of 68000 and 68008 chips, and plan a couple of
projects around them. Just subscribed to this group tonight :-)

--- Zach

Tom Evans

unread,
Feb 24, 2016, 7:15:53 PM2/24/16
to
On Wednesday, February 24, 2016 at 6:42:07 PM UTC+11, Zach Metzinger wrote:
> Yep, pretty cool that m68k is alive and well.

As long as you don't mind a post/response time in this group that would have been considered slow back in the 17th century for mail between Australia and the UK...

I'm also reminded of the "I'm not dead" line in:

https://en.wikiquote.org/wiki/Monty_Python_and_the_Holy_Grail

If you have questions about your 68k chips, you might be better off here:

https://community.freescale.com/community/coldfire/content

ColdFire's "not quite dead yet" either...

Tom

Jeff Jonas

unread,
Feb 25, 2016, 1:03:09 AM2/25/16
to
>I've come into a supply of 68000 and 68008 chips, and plan a couple of
>projects around them. Just subscribed to this group tonight :-)

Very awesome!
Do you have some firmware and cross compiler for it?
I used a 68008 board not so long ago.
My web page includes the monitor source:
http://ferretronix.com/tech/sbc/index.html#68k

This may help:
Application Note AN897 "MC68008 minimum configuration"
http://ferretronix.com/tech/68k/an897.fsl.d.rev0.pdf

-- jeff jonas

Zach Metzinger

unread,
Feb 25, 2016, 1:51:06 AM2/25/16
to
Oh, I've learned the value of patience over the years. Nothing teaches
you that like cross-compiling on a Sparc 1. ;-)

While I appreciate the information on "forums", I'd rather spend my time
on Usenet with you fine folks.

--- Zach

Zach Metzinger

unread,
Feb 25, 2016, 1:55:01 AM2/25/16
to
I'm a *NIX geek and use GCC to cross-compile to my targets. By day, I'm
a ARM Cortex-M coder for a place that builds microcontrollers. By night,
I hack on ancient hardware! ^o^

Thanks for the pointers to that application note! Here's a great project
someone did with the 68008:

http://wandel.ca/homepage/mc68008/

He even lists that same app note. Do you, by chance, know where I might
find the AR226 article he mentions?

--- Zach

Zach Metzinger

unread,
Feb 25, 2016, 2:01:24 AM2/25/16
to
On 02/25/16 00:03, Jeff Jonas wrote:
> I used a 68008 board not so long ago.
> My web page includes the monitor source:
> http://ferretronix.com/tech/sbc/index.html#68k

I completely missed your link. Very cool site with great information!

My web presence isn't nearly that fancy:
https://www.praxibetel.org/index.html

When my wire-wrap order comes in, I plan to start wiring a
proof-of-concept up. Right now, I'm working on the 22V10 equations for
address decoding and DTACK_N.

I have an idea for something really oddball, but useful. We'll see. :-)

--- Zach

Bruce Mardle

unread,
Feb 25, 2016, 2:17:02 AM2/25/16
to
Hi, Zach. Welcome to the group!
Since you're a fan of *ix, you might like http://www.bigmessowires.com/68-katy/
Linux on a breadboard!

PS: My 68010 homebrew currently has ~DTACK permanently grounded. Memories are faster than they were when the CPU came out!

Marcopolo

unread,
Feb 25, 2016, 3:23:57 AM2/25/16
to
Le 25/02/2016 07:53, Zach Metzinger a écrit :
> He even lists that same app note. Do you, by chance, know where I might
> find the AR226 article he mentions?
>
> --- Zach
>
Hello,

Maybe AR226 is a reprint of this article:
Designing with the 68008 Microprocessor (WirelessWorld 1984)?

Available here: www.retronik.fr/motorola

Marc

Zach Metzinger

unread,
Feb 25, 2016, 9:34:12 AM2/25/16
to
I believe that this is the exact article.

Merci!

--- Zach

Ola

unread,
Feb 25, 2016, 4:15:39 PM2/25/16
to
Hi,
I recently got a 68008 working (i.e. loading scode) with 3 ttl chips, rom,
ram (512+128K) and 68681. But I have to admit - I used a pcb (10*10cm).
Amazing how many errors you can do in a project - even after you are sure it
MUST be correct.
But very rewarding to succeed!

Ola


"Zach Metzinger" <ple...@ask.me> wrote in message
news:najmmr$rt5$2...@dont-email.me...

Zach Metzinger

unread,
Feb 25, 2016, 10:03:19 PM2/25/16
to
On 02/25/16 15:15, Ola wrote:
> Hi,
> I recently got a 68008 working (i.e. loading scode) with 3 ttl chips, rom,
> ram (512+128K) and 68681. But I have to admit - I used a pcb (10*10cm).
> Amazing how many errors you can do in a project - even after you are sure it
> MUST be correct.
> But very rewarding to succeed!

Did you keep the ROM at 0x00000000 and use jumps in the interrupt vector
table, or did you do the cycle-count-and-swap-ROM-for-RAM trick?

It's pretty easy to do the latter in my address decoder PAL, and I
understand that most existing code expects that behavior.

--- Zach

Zach Metzinger

unread,
Feb 25, 2016, 10:28:43 PM2/25/16
to
Thanks Bruce!

I have a rail of HM628128BLP-7 70ns 128k x 8 SRAMs, so you're right
there on ~DTACK.

I'm debating on putting a slow EPROM, a DS1220Y-200 with write disabled,
or a more modern-ish DIP package flash I have floating around (I think
Intel). The last one is probably fast enough to read with no wait states.

Real EPROMs with the quartz window are so cool!

Yup! I've seen BMOW, and Magic-1. They're on the Homebuilt CPUs WebRing.
(http://members.iinet.net.au/~daveb/simplex/ringhome.html)

--- Zach

Marcopolo

unread,
Feb 26, 2016, 3:33:10 AM2/26/16
to
Le 26/02/2016 04:02, Zach Metzinger a écrit :
>
> Did you keep the ROM at 0x00000000 and use jumps in the interrupt vector
> table, or did you do the cycle-count-and-swap-ROM-for-RAM trick?
>
> It's pretty easy to do the latter in my address decoder PAL, and I
> understand that most existing code expects that behavior.
>
> --- Zach
>

A other solution for the ROM/RAM swap?
http://lell.se/hacks/p/adr-decode68000.png

--
Marc

Bruce Mardle

unread,
Feb 26, 2016, 7:49:15 AM2/26/16
to
On Thursday, 25 February 2016 21:15:39 UTC, Ola wrote:
> Hi,
> I recently got a 68008 working (i.e. loading scode) with 3 ttl chips, rom,
> ram (512+128K) and 68681. But I have to admit - I used a pcb (10*10cm).
> Amazing how many errors you can do in a project - even after you are sure it
> MUST be correct.
> But very rewarding to succeed!

Indeed!
I've been designing a board with a XR82C684 UART for, erm, longer than I care to think about! It claims to be 68000-compatible, though it still expects ME to turn its ~INT signal into ~IPL2/1/0 and generate an interrupt acknowledge [scowl]. Just brushing up on my vintage programmable logic skills!
I see from a Google search that Freescale (née Motorola Semiconductor) issues an End Of Life notice on the 68681 in 2009... but the NXP version seems readily available.

Bruce Mardle

unread,
Feb 26, 2016, 7:52:20 AM2/26/16
to
Or just use a 68010 or later :D
Too bad there wasn't an 8-bit one of those. (Actually, I only bought my 68010 cos it was a) on offer and b) the biggest DIP I knew of :) )
Actually actually, all the RAM on my 68010 at the moment is battery-backed RAM, so the vector table can stay at its default address 0.

Ola Hollum

unread,
Feb 29, 2016, 4:13:03 PM2/29/16
to
As the system was simple (3 TTL chips :-) I used a jump table. Never taxed the 640K ram anyway.
Ola

Tom Evans

unread,
Mar 3, 2016, 12:50:55 AM3/3/16
to
On Friday, February 26, 2016 at 11:49:15 PM UTC+11, Bruce Mardle wrote:
> On Thursday, 25 February 2016 21:15:39 UTC, Ola wrote:
> > Hi,
> > I recently got a 68008 working (i.e. loading scode) with 3 ttl chips, rom,
> > ram (512+128K) and 68681. But I have to admit - I used a pcb (10*10cm).
> > Amazing how many errors you can do in a project - even after you are sure it
> > MUST be correct.
> > But very rewarding to succeed!
>
> Indeed!
> I've been designing a board with a XR82C684 UART for, erm, longer than I care to think about! It claims to be 68000-compatible, though it still expects ME to turn its ~INT signal into ~IPL2/1/0 and generate an interrupt acknowledge [scowl].

They never "just plugged together". Something had to handle the IPL pins. Although I remember you could simply decode the space pins to get an IACK and connect that directly to VPA to get one of 8 Autovectors. I've also seem cheap hardware (looking at YOU, Apple) that had only three interrupting devices and hardwired the requests to IPL0, 1 and 2. So the only valid interrupts were 1, 2 and 4, but you could get the other vectors if more than one device interrupted at once. You could even get an IPL7 if you were unlucky. Better to use a priority encoder chip.

> I see from a Google search that Freescale (née Motorola Semiconductor) issues an End Of Life notice on the 68681 in 2009... but the NXP version seems readily available.

Distributors may have stock but that part is certainly EOL'd:

http://www.nxp.com/products/interface-and-connectivity/wired-connectivity/uarts/two-channel/dual-asynchronous-receiver-transmitter-duart:SCC68681

The above doesn't show the status, but if you then type "scc68681" into the Search Field you'll get a page listing four of them, all "End of Life".

Tom

Simon Ferber

unread,
Mar 24, 2016, 11:02:45 AM3/24/16
to
Hello,

Am 25.02.2016 um 22:15 schrieb Ola:

> Amazing how many errors you can do in a project - even after you are sure it
> MUST be correct.
> But very rewarding to succeed!

yes, pretty amazing. Some of the first lessons I have learned when
working on my Kiwi prototypes were:

- Avoid ground-bounce by increasing Vcc track width
- Double check inputs you don't need -> Give them a defined level
(either Vcc or GND)
- Take a break if you are stuck at a problem

Best
Simon

--
Kiwi - ein Selbstbau Computer

Projekt Webseite: http://www.ist-schlau.de
Projekt Forum: http://www.ist-schlau.de/phpBB3
Reply all
Reply to author
Forward
0 new messages