Does this project spark your interest?

Become a member to follow this project and don't miss any updates

MC68000 Backplane Computer

9 92 71
Enjoy this project?
Share on twitter   Share on Facebook

This project was created on 12/20/2013 and last updated 4 months ago.

The design and construction of a hombrew computer based on the Motorola 68000 CPU

​This project is a homebrew computer based on the Motorola 68000 CPU. Design includes four megabytes of RAM, 128kB of ROM, A Yamaha V9938 VDP (the video chip in the MSX2), two serial ports at 9600 baud, and eventually networking and a hard disk. Read more »

See all details

  • 2 × 28C256 Content/Electronic Components/Semiconductors and Integrated Circuits/Memory ICs/PROMs, OTP PROMs
  • 8 × AS6C4008 Content/Electronic Components/Semiconductors and Integrated Circuits/Memory ICs/Static RAM (SRAM)
  • 1 × MC68000 Content/Electronic Components/Semiconductors and Integrated Circuits/Microprocessors, Microcontrollers, DSPs/Microprocessors (MPUs)
  • 2 × 6850 ACIA Content/Electronic Components/Passive Components/Inductors, Chokes, Coils and Magnetics/Current, Voltage and Power Transformers
  • 1 × Yamaha V9938 VDP

Project logs

    a month ago • 0 comments

    This is something I've written about in a front page Hackaday post, but I think it's time to go over a little more of the theory of what I'm doing here. First, a video: Read more »

  • Theory of RAM and ROM modules

    3 months ago • 4 comments

    Although it might make sense to start this project by building a CPU module first, I decided it would make more sense to start with the memory for this system. This serves two purposes: as an explanation of how the 68000's memory-mapped I/O works, and to have a relatively simple circuit built before embarking on the more complex that include the CPU module. Read more »

  • Theory of a CPU module

    3 months ago • 0 comments

    Compared to the 8080, the Z80, the 6809, 6502, and all the other 8-bit microprocessors used in boxxen of yore, the CPU I’m using for this project - the Motorola 68000 is both extremely powerful and extraordinarily complex. The power comes from a huge address space and some neat features like a divide instruction. The complexity comes from it’s asynchronous nature. Read more »

View all 6 project logs

kurq wrote 4 days ago null point

It would be interesting if you get a cpu/bus performance like the fastest 68000 computer ever: The Sage.
Just Google it, you will find it was 68000 computer designed for efficient usage of its speed.

Are you sure? [yes] / [no]

tomcircuit wrote a month ago null point

The 6850 is so braindead... ugh... I'm happy to donate one or two 68681 and/or 68901 chips to your project. I've got at least a dozen of each scurried away... I'm sure I've got one or two 68230 as well.

As much as these 68xxx chips are convenient to use with the async bus of the 68K, I've come to the conclusion that the only real reason that I, personally, would ever use most of these devices is for software compatibility with some older system. The state logic required to interface a synchronous bus peripheral device (i.e. without DTACK) to the async 68K bus is straightforward and easy enough to implement in a GAL or CPLD.

I think the exception (pun intended) I would make to this statement is that the 68901 in particular makes a fine vectored interrupt controller. I have no desire to re-invent THAT wheel...

Are you sure? [yes] / [no]

Benchoff wrote a month ago null point

You're right that the 6850 is dumb. I'd really rather not bother with DTACK and 6800 stuff. I'm kinda in a bind with this project, though: I want to make it as simple as possible, but also give people a chance to replicate it. So far, all the chips can be ordered off jameco (with the exception of the 68000 and V9938/V9958). The other 68XXX chips... they're hard to find.

If you want to donate something, email me at (my last name) Give me your address and I'll send out a T-shirt and some stickers for your trouble. I'll also promise to use the '681... after I've brought the system up with the 6850. Beauty of the backplane, I guess.

Are you sure? [yes] / [no]

lennart.lindell wrote 2 months ago null point

Here is a diagram.

The idea is that the ROM and not the RAM is mapped at addr 0 at reset. When the ROM is addressed at its runtime address the RAM is mapped instead at addr 0.
The ROM is in this picture mapped at $500000-$5FFFFF, it is better to move that to a high address.

Are you sure? [yes] / [no]

lennart.lindell wrote 2 months ago null point

Try to find a 68681 DUART instead of the 6850. And a 68901 that includes UART, timer and an 8-bit port. You need a timer don't you? ;)
Move peripheral and ROM to as high address as possible to get continous RAM from address 0 and up.
And I have some comments how to simplify the ROM/RAM address decoding:
Use a simple flip-flop instead of the counter at reset and see comment in text about RAM.
I had better draw a figure than explain it in text.

Are you sure? [yes] / [no]

Benchoff wrote 2 months ago 1 point

I was originally planning on using the Motorola 68k educational/dev computer from 1979 as the ROM monitor, thus necessitating the use of the 6850. Guess I'll just roll my own with the 68681.

You're completely right about the RAM /BOOT decoding. I'll get that in my notebook, and eventually on a project log here.

You're always free to put a diagram on imgur and post a link in a comment.

Are you sure? [yes] / [no]

wrm wrote a month ago null point

That... is clever. I also used the counter-based system, it comes straight from Motorola AN-897, but I admire this one.

There's Pete Stark's HUMBUG and also TUTOR/TUTORNEW out there in source format, they work with the Quelo assembler (HUMBUG needs a bit of a massage). They both supports the 68681.

With hindsight I would also have mapped my ROM right at 0x00FFxxxx to keep RAM contiguous. It's easier to fill up a 16 M memory space these days than it was back in '87...

Are you sure? [yes] / [no]

Give feedback