6502.org

Has anyone ever created any 6502 based computers that use parts that are typically used on Z80 based systems?

For example, the SN76489 (audio), AY-3-8910 (audio) and the TMS9918 (video) were usually found in Z80 based systems like the ColecoVision, MSX, etc.

But there is nothing special about them that would prevent being used in a 6502 based computer.

Thanks

I haven't used the chips you mention, but I agree they should be 100% alright for use with 65xx. You just have to give them the RD\ and WR\ control signals they understand, and deal with any timing issues.

However, there is one general case where you will lose some capability, and that is stuff like Z-80-family peripheral chips that have vectored-interrupt capability and whatnot. Such features are Z-80-specific. There's little to no chance you'll be able to use them on a 65xx system, but of course you still get to use the rest of the stuff in the IC.

-- Jeff

I was told many years ago that Am9511 was easily interfaced to Z80 but a dog wrt. 6502. It was possible and I knew someone who did it, using rather a lot of interfacing chips but it was considered painful and expensive.

Well I've interfaced both audio chips to an Arduino and a shift-register. It was surprisingly easy once I figured out the timing.

So I would imagine just having the 6502 perform the same timing.

Also, I can't think of any commercial systems that would have mixed those. All of the commercial systems I can think of that used 6502's almost always used customized components for audio/video.

The SN/AY/TMS were off-the-shelf.

Here is an Apple //c peripheral with an AY-3-8910 (In fact it may have had two if I recall correctly):

http://apple2history.org/wp-content/upl ... ricket.jpg

There was a production complete computer that had a 6502 uP and AY-3-8910 as well, but which one it was escapes me.

I see the BBC Micro's default disk controller, the 8271, has a \RD and a \WR pin.
http://mdfs.net/Info/Comp/BBC/Circuits/BBC/bbc.gif (Big image - disk controller is top left corner.)
(This was a poor choice of chip and a lot of people used a 1770 on a daughterboard. In which case the daughterboard, in the 8271 socket, has to make a RnW signal. )

It was the Oric-1 and Oric-Atmos. They used an AY-3-8912 (a 28-pin version of the AY-3-8910 with one of the I/O ports not brought out to pins) , although it was accessed through the 6522 VIA rather than being directly on the 6502 bus.

There were two peripheral boards for the Apple // which were publicly available:
The Sprite board from Synetix and the Arcade board
Both boards include a TMS9918 graphics chip and at least a AY38910 chip (depending of configuration) for sound fatter than result from native speakers. Some of them also include a speech chip.
Further details can be found here:
http://www.atarimagazines.com/creative/ ... phics_.php

Ah, thanks.

I know what you're talking about for the Apple. It was the Mockingboard. And there were others. Like the Phasor. But they were after-market cards that weren't very popular. In fact, I built a Mockingboard on a breadboard for my Apple IIe.

I do remember the TMS card for the Apple now that you mentioned it. That was an odd beast because almost no one had it. And, it wasn't a replacement for the Apple's video. It could be used for a second screen. So that would have been a lot of money back then. Even though the TMS9918 has a video "pass through", I don't know if it would be compatible with Apple's video. Would have been cool if it could, though. Which would have given the Apple 2 playfields along with 32 sprites. But alas, it wasn't meant to be.

The Acorn Atom runs on a 6502 but uses an 8255 to decode the keyboard. Does that count?

===Jac

Check out the N8VEM group. We have the 6502 processor interfacing to several of the Z80 based cards including the TMS9918, several other video cards and several disk controller chips.

Dan Werner

I always thought it was fair to say that with enough glue logic nearly any peripheral IC can be made to work with any CPU? The GI bus is pretty weird, yet the AY was widely used. It's probably easiest to talk about combinations of parts which cant be made to work together? Of course I mean within the sphere of 8 bit devices.

Ohio Scientific used an 6850 for serial i/O and a 6820 + 6850 for a FD interface. Not too unusual to use 68XX chips with 6502s I suppose, but the disk interface was a bit unique.

The Xetec Lt. Kernal used a Motorola 6821 PIA to drive the SCSI bus. Minor bit of trivia: the 6821 has two 8 bit I/O ports like the 6522, giving 16 bits to drive the bus. However, the "narrow" SCSI bus needs 18 lines. So the parity and reset lines were not implemented.

I came across an image which makes it look easy: an expansion card for an AppleII which apparently could have been even simpler:

The product is from 1983, comes with a Forth and proclaims 100x faster floating point. There's more here including followup on Garth's spotting of a similar offering announced in 1980: